Foundations for Lunar Nuclear Reactors, and the Engineering Subgrades in Extreme Conditions

The deployment of compact nuclear reactors on the Moon is no longer speculative. Concepts from Lockheed Martin, Westinghouse, and Rolls-Royce presented in 2022 are progressing toward mission-ready systems under initiatives such as NASA’s Fission Surface Power (FSP) project. These reactors, typically designed to produce 10 to 40 kW of continuous electrical output, are essential for enabling long-term habitation, ISRU, and nighttime operations on the lunar surface.

What remains almost entirely unaddressed is how these reactors will interface with the ground.

Design teams have focused on reactor shielding, redundancy, and thermal rejection but have yet to confront the engineering realities of placing a high-value, long-duration fission system on granular, compressible, and potentially unstable terrain. No reactor system on Earth would be fielded without rigorous subgrade design, but current lunar reactor concepts remain silent on foundation performance, bearing behavior, settlement tolerances, and thermal–mechanical interaction with the regolith.

This omission is not minor. Lunar regolith lacks cohesion, exhibits shallow stress diffusion, and behaves unpredictably under vibration, compaction, and thermal cycling. Without a validated method for stabilizing the reactor base, whether slab-mounted, buried, or vault-installed, the integrity of the power system is exposed to unacceptable risk.

The objective of this article is to directly address this engineering blind spot. It outlines feasible subsurface foundation strategies for compact reactors, tailored to lunar conditions, with reference to terrestrial analogs, mission design parameters, and observed mechanical responses of regolith under load. This is not a discussion of reactor design, it is a demand for structural realism in the systems that support them.

Design Constraints and Load Cases

Foundations for portable nuclear reactors on the Moon are governed by load characteristics that differ substantially from terrestrial analogs, not due to gravity alone, but because of regolith behavior, vacuum exposure, lack of moisture-dependent mechanisms, and thermal extremes (Figure 1).

Figure 1: Cross-sectional conceptual illustration of a portable microreactor embedded in lunar regolith. The reactor is shown buried at an approximate depth of 3 meters, spanning regolith layers L1 to L3. This preliminary depth assumes average regolith stratigraphy but must be confirmed through in-situ subsurface investigations to validate bearing capacity, thermal insulation, and radiation shielding performance.

The structural load paths must account for:

1. Dead Load and Reactor Mass

   Reactor designs such as those proposed by Lockheed Martin and Westinghouse estimate system masses in the range of 3 to 6 metric tons, including radiators, shielding, and support structures. Even under reduced lunar gravity, the regolith must support a concentrated footprint with limited stress diffusion. On Earth, this weight would be distributed through granular or concrete foundations with known bearing parameters. On the Moon, load transfer depends on interparticle friction, angular interlock, and shallow contact redistribution, all of which remain poorly understood under reactor-scale pressures.

   
   

2. Dynamic Loads and Vibration

   Internal reactor components generate operational vibrations, while external events such as micro-impacts or equipment traffic contribute additional dynamic loading. Regolith's low confining stress and lack of damping properties may lead to grain rearrangement, settlement, or tilt over time. These effects cannot be mitigated using terrestrial isolation pads or elastomeric bearings; novel mechanical decoupling strategies must be developed.

   
   

3. Thermal Gradients and Expansion

   Reactor operations induce steep thermal gradients into the surrounding ground, both through direct conduction and radiative coupling. Unlike Earth soils, regolith lacks moisture buffering and thermal conductivity varies sharply with compaction. Differential heating can cause localized expansion, grain displacement, and void formation beneath or around the base, especially if mounted on slabs or shallow footings.

   
   

4. Shielding Requirements and Burial Depth

   To meet human-rated shielding limits, some concepts propose partial burial or berming of the reactor module using regolith. This introduces lateral earth pressures, thermal insulation layers, and potential differential settlement. Load-bearing assumptions for buried vaults or bermed installations must account for heterogeneous regolith stratigraphy and the risk of non-uniform compaction or collapse.

   
   

5. Installation Constraints

   All lunar foundation solutions must be compatible with autonomous emplacement, remote verification, and minimal equipment logistics. There is no tolerance for on-site adjustment, re-leveling, or post-placement grouting. The design must succeed the first time. These constraints define a new class of load cases: high-value static systems operating under unknown ground conditions, with no margin for structural instability. Subsurface engineering for such systems cannot be retrofitted; it must be embedded from the start of reactor design.

   
   

Subsurface Interaction and Regolith Response

Microreactors are not conceptual devices, they are critical enablers of sustained lunar operations. Yet, current mission planning neglects a foundational premise: power infrastructure must be grounded, and the ground itself is a variable not yet understood. No terrestrial reactor would be commissioned without rigorous subgrade design. The same principle must apply to the Moon.

Portable nuclear reactors, such as those proposed by Lockheed Martin and Westinghouse, exert concentrated static loads in the range of 5,000 to 10,000 kg, depending on shielding configuration and heat rejection systems. Unlike broad-based structures, these reactors have small footprints, often cylindrical or hexagonal, resulting in elevated bearing stress at the interface. The required stability margin must also accommodate lateral thrust during thermal expansion, anchorage of heat sinks, and seismic response to moonquakes or meteoroid impacts.

As established previously, stress propagation into regolith under lunar gravity is minimal. There is no effective overburden to confine load paths, no interstitial fluid to redistribute stress, and no pore pressure to dissipate. Load transfer mechanisms are driven by interparticle contact, angular friction, and compaction resistance within a granular field of undefined density and anisotropy.

Standard elastic or plastic foundation theories break down. Terzaghi’s bearing capacity equations, for instance, assume a confining medium with shear failure envelopes. On the Moon, stress quickly collapses into shallow, particle-scale rearrangements. This implies that subgrade stiffness, not bearing capacity, becomes the primary performance criterion. The reactor must not tilt, sink, or lose alignment over time. Without these considerations, operational safety is compromised.

Subsurface heat transfer is nontrivial. Unlike radiators exposed to vacuum, buried reactors create thermal plumes that disrupt adjacent regolith. The low thermal conductivity of regolith (on the order of 10⁻³ to 10⁻² W/m·K) leads to localized heating, potential expansion, and weakening of the support matrix. Without isolation layers or engineered backfill, the reactor itself can induce its own destabilization.

Foundation systems must incorporate thermal break layers, possibly sintered panels or ISRU-derived insulating mats. These must be integrated with anchorage elements, such as micro-piles, mechanical skirts, or interlocked trench bases. Conceptual renderings of “drop-in” reactors placed on raw regolith are insufficient. Construction-grade performance requires stratified engagement with the ground.

The decision to bury reactors is rooted in shielding and thermal regulation. However, the depth to which burial occurs, typically 2.5 to 3.5 meters, intersects with multiple regolith zones. According to the L1–L5 excavation zoning framework:

• L1 is the upper disturbed surface (fine, volatile-rich, electrostatically active)

• L2 includes compacted surface regolith with higher angularity and voids

• L3 comprises transitional megaregolith fines, fractured aggregates, and ancient ejecta

Anchorage into these layers demands a zoning-aware approach. Excavation should target the L3 horizon where particle size increases and compaction potential rises. But without subsurface validation, through GPR, Absolute Vector Gravimeter, penetrometer, or core extraction, depth planning remains speculative.

Engineering Scenarios for Reactor Foundations

The conceptual placement of nuclear microreactors on the Moon must transition from renderings to constructible, validated foundation systems. At present, no standardized framework exists for anchoring such reactors in the lunar subsurface. The following scenarios explore practical, mission-adaptable alternatives grounded in terrestrial geotechnical logic, adjusted for lunar constraints.

Shallow Trench with Compacted Base

Method: Excavation of a 3–4-meter trench, reaching mid-L3 layer, followed by mechanical compaction of the trench base using ISRU-grade tampers or robotic compactors. The reactor is placed directly on the compacted surface, optionally with an intermediate isolation layer (e.g., sintered basalt slab or fabricated ceramic mat).

Benefits:

• Minimal material import

• Leverages in-situ regolith

• Rapid deployment using excavation-class rovers

Risks:

• Variable subgrade stiffness

• Unquantified settlement behavior

• Thermal feedback into surrounding regolith

  
  

Recommended Use: Short-term or mobile reactor missions where relocation or redundancy mitigates foundation degradation.

Modular Vault with Engineered Fill

Method: Construct a prefabricated vault or shell using imported or ISRU-fabricated materials, placed within a prepared excavation. Void space around and beneath the structure is backfilled with engineered particulate (crushed sintered regolith or graded simulant), layered and compacted to specification. Anchoring fins or skirts are embedded within the vault walls.

Benefits:

• Controlled boundary conditions

• Reduced differential movement

• Potential to integrate thermal and radiation shielding within the vault

Risks:

• Requires material import or high ISRU capacity

• Complexity in assembly and sealing

• Increased excavation volume

Recommended Use: Permanent installations at hubs, power plants, or ISRU processing zones.

Mechanical Skirt with Micro-Pile Anchorage

Method: Utilize mechanical skirts extending radially from the reactor base, penetrating the regolith up to 1 meter. These skirts transfer load through bearing and shear resistance. Vertical micro-piles (titanium alloy or compacted regolith columns) provide deeper anchorage and thermal isolation.

Benefits:

• Lateral stability without full trench

• Minimizes excavation

• Good resistance to thermal distortion and vibration

Risks:

• Requires precise deployment mechanism

• Load transfer highly sensitive to local regolith composition

• Limited in very fine or cohesionless material

Recommended use: Modular systems deployed in multiple locations with varying terrain conditions.

Sintered Pad

Method: Sinter a 3–5-meter diameter pad using concentrated solar or microwave energy, creating a semi-rigid regolith slab. Excavate a shallow dome below the slab and position the reactor into this form-fitting cavity, shielded by overhead regolith berms or thermal blankets.

Benefits:

• Reduced thermal penetration

• Passive structural confinement

• Demonstrates early-stage ISRU fabrication utility

Risks:

• High energy demand

• Sintered strength dependent on mineral composition

• Long preparation time

Recommended Use: Demonstrator missions or strategic proof-of-concept deployments with ISRU focus.

Note: Each of these foundation strategies must be validated through physical testing in vacuum chambers, with regolith simulants under variable thermal and gravitational analogs. At present, no existing test program, public or private, has demonstrated full-scale performance for reactor foundations in lunar conditions. This is an unacceptable blind spot in mission design. Lunar surface systems will not fail due to reactor design, but rather due to foundation instability, differential movement, or heat-induced degradation at the subgrade. These risks can be controlled, but only if structural engagement with the ground is prioritized, not after the fact, but from the first phase of system design.

Potential Risk Classes and Operational Consequences

Designing foundations for microreactors on the Moon without a rigorous understanding of regolith behavior exposes mission-critical systems to unacceptable risks. These risks are not abstract, they stem from poorly defined load paths, unpredictable thermal responses, and inadequate subgrade confinement. To support risk-informed planning, this section proposes a provisional classification of foundation risks based on site conditions, design assumptions, and operational requirements.

Class I - Controlled Excavation with Engineered Fill

Site Condition: High-resolution terrain mapping and in-situ sampling confirm a stable regolith profile with predictable layering and compaction response.

Foundation Type: Vault with engineered fill or reinforced trench with compaction protocols.

Operational Risk: Low. Load transfer is well-characterized. Settlement and thermal distortion are minimized. Suitable for long-duration deployment of critical energy infrastructure.

Class II - Partial Knowledge with Unverified Interfaces

Site Condition: Remote sensing and analog data provide partial confidence in stratigraphy, but no in-situ probing or load testing is available.

Foundation Type: Shallow trench with mechanical skirt, or sintered pad with anchoring enhancements.

Operational Risk: Moderate. Bearing behavior is sensitive to subsurface anomalies (voids, vesicular zones, or fractured regolith). Reactor systems may experience tilt, uneven settlement, or anchor slippage under thermal cycling or secondary loading (e.g., regolith deposition, lander activity nearby).

Class III - No Prior Characterization

Site Condition: No ground truth. Deployment on untested terrain or emergency/reactive positioning.

Foundation Type: Direct surface placement with minimal anchorage. Load supported by contact-only interface.

Operational Risk: High. Stress localization, differential thermal expansion, and regolith reorganization under load may lead to structural instability, system decoupling, or partial subsidence.

Escalating Consequences

Failure in lunar foundations does not follow a linear risk profile. Minor deviations in bearing or tilt can compromise:

• Heat exchanger performance, due to uneven regolith contact or poor conduction paths.

• Power distribution, when cable strain or anchor movement disrupts transmission.

• Reactor integrity, through structural fatigue or shock amplification under regolith deformation.

No fault tree, no digital twin, and no control algorithm can mitigate a foundation failure once deployed. Risk reduction must happen in the physical domain, at the regolith interface, not in simulations after launch.

Toward Codified Lunar Foundation Design

There is currently no accepted standard, code, or methodology for designing and verifying foundations on the Moon. While structural engineers continue to advance concepts for power reactors, antennae, and surface systems, the interface between those structures and lunar ground remains undefined. This is no longer a conceptual gap—it is a structural liability.

On Earth, foundation design is guided by codified frameworks such as Eurocode 7, AASHTO, and ASTM standards, each rooted in decades of empirical field behavior, subsurface testing, and material predictability. Lunar regolith violates nearly every underlying assumption of these standards:

• No effective stress behavior

• No Cohesion from moisture

• No deep confinement

• No known bearing capacity forces

Attempts to adapt these models to lunar conditions result in untraceable safety factors and unverifiable assumptions. Without ground-truth calibration, “design” becomes extrapolation.

A future lunar foundation standard must abandon terrestrial inheritance and instead develop from first principles and site-specific behavior. It must include:

• Stratigraphy-based classification (e.g., L1–L5 zoning with physical parameters)

• Stress distribution models calibrated for low-confinement granular media

• Bearing capacity expressions derived from load testing in simulants and in-situ probes

• Deformation envelopes for angular, interlocked, and fragmented media

• Anchorage guidelines accounting for shallow pull-out resistance and sliding thresholds

• Thermal-structural interfaces based on cycling-induced microstructural fatigue

Moreover, codification cannot occur without data. Prototype deployments must be fully instrumented, capturing real-time responses to:

• Static loads

• Lateral stress redistribution

• Thermal expansion and contraction

• Regolith motion, vibration, and settlement

Nuclear reactors, unlike modular payloads or robotic rovers, demand long-duration stability, shielding, and heat dissipation. Their foundations are not temporary supports, but structural lifelines. Codifying foundation design for reactors is the logical starting point for all future lunar infrastructure.

This codification effort must begin now, or the Moon’s next energy systems will be deployed on uncertain ground, literally and structurally.

Open Questions and Research Priorities

The deployment of portable nuclear reactors on the Moon presents a unique set of subsurface engineering challenges that remain unresolved. While terrestrial analogs provide limited guidance, the extreme environmental divergence on the Moon introduces fundamental unknowns that cannot be bypassed by simulation or assumption. A rigorous research agenda is required to close these gaps.

Unresolved Questions in Regolith-Structure Interaction

• What are the true in-situ mechanical properties of compacted L3–L5 regolith under thermal cycling and operational loads?

  No mission to date has retrieved representative data at the necessary depth or loading regime to validate long-term subgrade stability.

• How does regolith rearrange under localized point loads exceeding operational thermal gradients (reactor footprints)?

  While theoretical stress bulbs are informative, no empirical evidence confirms stress propagation under these hybrid loading conditions.

• What are the thresholds for differential settlement and lateral displacement over time under continuous heat generation?

  The combination of low gravity, particle angularity, and insulation effects creates behaviors not captured in any current analytical model.

Research Priorities for Mission-Critical Deployment

1. High-Fidelity In-Situ Characterization

   Instrumented coring, penetrometers, and shear probes capable of penetrating at least 3–5 meters are required for each reactor site. These must operate under lunar vacuum, dust, and thermal conditions to capture actual stress–strain profiles.

2. Thermomechanical Subgrade Testing

   Controlled experiments using lunar simulants subjected to reactor-grade heating cycles are necessary to simulate subsurface thermal fatigue, layer separation, and void development.

3. Scaled Load Testing with Simulated Mass

   Physical models of reactor bases should be deployed in large vacuum chambers or reduced-gravity simulators to measure tilt, displacement, and contact degradation over time.

4. Digital Ground Behavior Modeling Linked to Real Data

   Finite element models and discrete element simulations must be constrained with physical data from in-situ tests. Model calibration without empirical ground truth will perpetuate uncertainty.

5. Stress Decay Mapping under Combined Loads

   The interaction between mechanical loads and induced thermal plumes needs to be captured in detail to define the actual depth of influence and lateral spread. This is essential for long-term vault stability and proximity planning.

Until these gaps are addressed, no foundation design for surface-based nuclear reactors on the Moon can be considered validated. Agencies and contractors must treat subsurface engineering as a primary, not auxiliary, requirement. The current absence of defined criteria is not due to unsolved physics, but due to under-prioritization of ground mechanics in mission architectures.

Bibliography

1. Dyke, J., Lum, S., Polanskey, C., et al. (2024). Establishing Standards for Lunar In-Situ Resource Utilization Structural Materials. NASA Technical Report, ISRU-Material Working Group.

2. Terzaghi, K. (1943). Theoretical Soil Mechanics. John Wiley & Sons.

3. Terzaghi, K., & Peck, R.B. (1967). Soil Mechanics in Engineering Practice, 2nd ed. John Wiley & Sons.

4. Bieniawski, Z.T. (1989). Engineering Rock Mass Classifications: A Complete Manual for Engineers and Geologists in Mining, Civil, and Petroleum Engineering. Wiley.

5. Hoek, E., & Brown, E.T. (1980). Underground Excavations in Rock. Institution of Mining and Metallurgy.

6. Peck, R.B. (1969). Advantages and Limitations of the Observational Method in Applied Soil Mechanics. Geotechnique, 19(2), 171–187.

7. Siekmeier, J.A., et al. (1993). Behavior of Non-Standard Subgrade Soils for Pavement Structures. Federal Highway Administration.

8. Schrader, C. (1973). Lunar Soil Mechanics Report. NASA Technical Note NASA-TN-D-7051.

9. Wang, J., Gao, Y., Zhang, N., et al. (2024). Thermal–Mechanical Behavior of Regolith-Based Protective Structures for Lunar Surface Habitats. Acta Astronautica, 212, 333–347.

10. McLean, C., et al. (2021). Arched Structures for Lunar Application: Design, Optimization, and Material Considerations. Proceedings of the IAC.

11. Mottaghi, M., & Benaroya, H. (2015). Structural Design of Pressurized Lunar Habitats: Load Cases and Failure Criteria. Journal of Aerospace Engineering, 28(5), 04014105.

12. Rolls-Royce. (2022). UK Space Nuclear Power Program: Micro-Reactor Concept for Lunar Deployment. Rolls-Royce Holdings Technical Brief.

13. Lockheed Martin. (2022). Compact Nuclear Power Systems for Lunar and Mars Bases. Company Technical Dossier (internal release).

14. Westinghouse Electric Company. (2022). eVinci™ Microreactor Program Overview. Company White Paper.

15. Canup, R.M., & Esposito, L.W. (1996). Accretion of the Moon from an Impact-Generated Disk. Icarus, 119(2), 427–446.