Abstract
Accessing the martian deep subsurface is a long-standing scientific priority for astrobiology, climate reconstruction, and planetary evolution, yet robotic drilling missions have historically been limited by wellbore instability, loss of working-fluid circulation, and the risk of irrecoverable tool entrapment. This work presents and evaluates a wireline, downhole-actuated pneumatic drilling architecture designed to directly mitigate these mission-ending risks through active wellbore pressure support and continuous cuttings removal within a single, sealed CO2 circulation system. The proposed system combines a rotary-percussive bottomhole assembly with a deployable sealing membrane and a closed CO2 pneumatic circuit that provides both mechanical support to the borehole wall and transport of generated cuttings to the surface. Reduced-order flow physics models are developed to capture compressible gas transport, particle entrainment, porous leak-off, junction losses, incompressible liquid tether flow, and phase-change thermodynamics. These models are assembled into section-wise drilling and cleanout cycles and integrated into a mission-level simulator that enforces realistic sol-level constraints on time, energy, battery usage, and working-fluid mass. Mission simulations demonstrate that cleanout operations dominate both energy and CO2 mass budgets, establishing wellbore pressure support as a first-order design variable rather than a secondary constraint. Modest relaxation of the maintained back-pressure from an overburden-matched level to a derated fraction substantially reduces cleanout energy demand and idle leak-off penalties while preserving effective particle transport. Under an InSight/Mars Life Explorer-class mission envelope, the architecture exceeds a 30 m baseline depth target well within the nominal operational window, with favorable scaling toward ∼100 m depths through increased mission duration and resource allocation. By explicitly coupling drilling, cuttings removal, and wellbore stability within a single operational framework, this architecture targets the primary failure modes identified in deep martian subsurface access. The results indicate, at the concept and reduced-order sizing level, that pressure-supported pneumatic drilling may provide a scalable pathway for deep drilling on Mars and other low-pressure planetary bodies, while identifying the subsystem validation needed before flight-system viability can be assessed.
Introduction
Science objectives
The primary scientific motivation for drilling on Mars is to access subsurface environments that offer substantially improved conditions for preserving evidence of life on early or extant Mars relative to the planet’s surface. While orbital observations and rover missions have demonstrated that early Mars supported widespread liquid water and habitable surface conditions, the modern martian surface is characterized by intense ionizing radiation, large thermal gradients, highly oxidizing chemistry, limited availability of stable liquid water, and continuous impact gardening. Acting together, these processes are expected to efficiently degrade or destroy mineralogical, molecular, and morphological biosignatures at or near the surface over geologic timescales (Dartnell et al., 2007; Pavlov et al., 2012; Hassler et al., 2014; Gardner et al., 2017). As a result, the lack of unambiguous biosignatures in surface materials may reflect post-depositional destruction rather than a definitive absence of life on Mars (Summons et al., 2011).
In contrast, the martian subsurface provides natural shielding from radiation and environmental variability, while maintaining comparatively stable thermal and geochemical conditions. On Earth, the deep subsurface hosts an extensive and diverse microbial biosphere sustained by water-rock interactions, redox gradients, and low, though persistent, energy fluxes (Edwards et al., 2012). These observations strongly support the hypothesis that subsurface environments on Mars represent the most plausible settings for the persistence of extant microbial life or the long-term preservation of biosignatures from extinct communities (Michalski et al., 2013; Westall et al., 2015; Cockell et al., 2016). Consequently, access to subsurface materials is a prerequisite for advancing Mars exploration from assessments of habitability toward robust biological validation. This priority is reflected in a recent Keck Institute for Space Studies (KISS) workshop on biosignature detection, which identified subsurface sampling as essential for advancing Mars life-detection experiments beyond surface habitability assessments and toward definitive biological validation (Perl et al., 2024).
The depth required to achieve these objectives is governed by the attenuation of surface-driven degradation processes and the likelihood of intersecting preserved or potentially habitable environments. Radiation modeling and laboratory studies indicate that biologically relevant organic molecules and mineral biosignatures may persist at depths of order one meter beneath the surface, where radiation fluxes are already substantially reduced (Dartnell et al., 2007; Pavlov et al., 2012). However, drilling to greater depths significantly expands the accessible scientific return. At depths of several meters to tens of meters, the subsurface may intersect buried ice, permafrost, salt-stabilized brines, or groundwater-altered and lacustrine strata that record extended periods of aqueous activity and offer improved preservation potential (Michalski et al., 2013; Vago et al., 2017).
Deeper subsurface access further enables direct stratigraphic interrogation that allows reconstruction of environmental evolution rather than reliance on reworked or incomplete surface exposures. From both biological and geological perspectives, access to these depth ranges is, therefore, essential. At the same time, depth targets must remain consistent with the stringent power, mass, reliability, and cost constraints of Mars missions. The science objective is, thus, not depth for its own sake but access to subsurface environments that maximize biological and geological return while remaining achievable within a low-cost, power- and mass-limited mission architecture (Zacny et al., 2013; Badescu et al., 2019b).
Planetary drilling and subsurface access
Planetary subsurface access on Mars has historically been limited to surface abrasion, shallow drilling, and near-surface sampling. Early rover missions, including the Mars Exploration Rovers (MER), employed rotary abrasion tools designed to remove weathered rinds and expose fresh rock surfaces for in situ analysis, providing essential mineralogical and geochemical context but no true subsurface access (Paulsen et al., 2006; Cannon et al., 2007; Bar-Cohen and Zacny, 2021; Wang et al., 2024). Subsequent missions incrementally expanded interaction with the shallow subsurface. The Phoenix lander excavated unconsolidated regolith to depths of several tens of centimeters, while the Curiosity rover introduced a rotary-percussive drill capable of acquiring powdered samples from competent rock at centimeter-scale depths. These systems marked important advances in surface science but remained fundamentally constrained to shallow penetration and surface-based sample handling (Cannon et al., 2007).
The most advanced flown drilling architecture to date is the coring system aboard the Perseverance rover, which enables the extraction and caching of intact rock cores for eventual return to Earth. While this represents a major advance in sample integrity and stratigraphic context, penetration depths remain limited to centimeters, and access to environments that offer substantial radiation shielding and long-term biosignature preservation is not achieved (Vago et al., 2017). Efforts to extend subsurface access beyond rover-mounted drilling, such as the HP3 mole deployed on the InSight mission, have highlighted the sensitivity of meter-scale penetration to regolith mechanical properties, frictional coupling, and deployment conditions. HP3 was designed to emplace thermal sensors to a target depth of 3–5 m, but the mole ultimately reached a tip depth of only about 0.37 m on Mars, which underscores the challenges of achieving reliable depth access without active borehole control or cuttings management (Wippermann et al., 2020; Spohn et al., 2022). Collectively, heritage systems demonstrate steady progress in drilling capability, while revealing a persistent gap between shallow subsurface interaction and the depths required to address key astrobiological and geologic objectives.
Beyond flown heritage systems, a diverse set of planetary drilling and subsurface access concepts has been developed and tested over the past two decades that target depths from meters to tens of meters under Mars-relevant environmental constraints. Early deep-drilling architectures largely adapted terrestrial rotary drilling approaches to robotic platforms, with surface-mounted actuation and mechanical transmission of drilling forces through a drill string. Representative examples include the DeeDri drill, developed as a rotary system for Mars applications, and the ExoMars drill program, which advanced a rotary coring architecture for competent rock and cemented materials. As summarized in Table 1, these systems were primarily top-driven and relied on auger-based cuttings removal with discrete extraction cycles, achieving limited demonstrated depths despite deeper design targets (Magnani et al., 2004; Van Winnendael et al., 2005; Vago et al., 2017). An extensive review of these systems is provided by Bar-Cohen and Zacny (2021), as well as by Wang et al. ( 2024).
Representative Planetary Drilling and Subsurface Access Systems and Demonstrations, Comparing Actuation Method, Deployment Architecture, Design Versus Demonstrated Depth Capability, and Cuttings Removal and Extraction Approaches Reported in the Literature
LISTER, Lunar Instrumentation for Subsurface Thermal Exploration with Rapidity.
In parallel, alternative actuation and deployment strategies were explored to reduce surface mass and decouple downhole drilling mechanics from lander or rover structures. Ultrasonic and percussive concepts, such as the Ultrasonic/Sonic Gopher and the Auto-Gopher series, employed downhole actuation with wireline or tethered deployment to penetrate ice and rock substrates. These systems demonstrated increased depth capability relative to top-driven approaches but relied on either in-hole compaction or bailing-based cuttings extraction via wireline retrieval, which limits continuous drilling and introduces operational complexity as depth increases (Badescu et al., 2006; Bar-Cohen et al., 2017; Badescu et al., 2019a). Related field work with the Planetary Deep Drill and the Wireline Analysis Tool for the Subsurface Observation of Northern ice sheets (WATSON) provides an important deep-access benchmark: in Greenland, the system extended an existing borehole from 84.5 to 110.5 m and acquired in situ borehole-wall spectroscopy over depths that exceeded 100 m (Paulsen et al., 2016; Malaska et al., 2020). This result demonstrates substantial terrestrial depth capability for a wireline drill-instrument architecture in ice, while remaining distinct from autonomous drilling in martian regolith and rock under lander-scale mass, power, and working-fluid constraints.
Additional staged and modular concepts, which include the Autonomous Tethered Corer (ATC), the Modular Planetary Drill System (MPDS), and the RedWater system, further expanded the design space by targeting deeper access through tethered coring, modular surface-driven drilling, or coil-tubing deployment. As reflected in Table 1, these approaches span a range of cutting-handling strategies, including bailing, auger-based extraction, and pneumatic transport in ice-rich environments. Despite achieving increased design depth targets, demonstrated depths remain substantially lower across most systems, which highlights a persistent gap between intended subsurface access and experimentally validated performance under Mars-relevant conditions (Shenhar et al., 2005,2006; Guerrero, 2008; Palmowski et al., 2022). These comparisons should be interpreted cautiously, however, because full-depth Mars-relevant testbeds are difficult to build, and field analog sites rarely provide both the desired depth interval and the specific rock, ice, or regolith properties assumed for a flight scenario.
Prior work has also examined mechanically deployed borehole-support elements for lightweight planetary drills. For example, the Ultralight Mobile Drilling System used C-shaped tubular booms to guide the drilling subsystem, transfer load, and secure shallow boreholes in loose formation while reducing the mass and stowed volume relative to conventional drill-string and casing approaches (Seweryn et al., 2015). Such deployable mechanical supports provide important precedent for borehole stabilization under planetary mass constraints, but they address a different architectural regime than the pressure-supported approach considered here: shallow mechanically supported coring rather than tens-of-meters-class drilling with pneumatic cuttings transport and a sealed pressure-support conduit.
Despite sustained development efforts, deep drilling on Mars remains fundamentally constrained by a combination of power, mass, reliability, and mission cost limitations. Mechanical specific energy (MSE) or energy needed to remove a unit volume of material provides a useful framework for assessing drilling efficiency and highlights the challenge of achieving meaningful penetration under the tight power budgets of Mars robotic platforms. Available mechanical power is typically limited to tens of watts, which implies extended drilling durations even for modest penetration depths in competent rock and cemented materials. As a result, many planetary drilling systems exhibit a substantial gap between design depth targets and experimentally demonstrated performance, which reflects the difficulty of reducing MSE sufficiently within Mars-relevant power envelopes (Teale, 1965; Badescu et al., 2019a).
Mass constraints further restrict feasible drilling architectures. Terrestrial drilling systems rely on heavy surface infrastructure that includes top drives, drill strings, casing, and circulating fluids to transmit force and torque, stabilize the borehole, and mitigate failure modes during drilling. On Mars, these approaches are largely infeasible due to launch mass limits and deployment complexity. Consequently, planetary drilling systems are typically forced to operate with lightweight, top-driven, or downhole-actuated architectures that provide limited reaction force, reduced borehole control, and constrained cuttings transport capability. As summarized in Table 1, these constraints limit achievable depth before loss of drilling efficiency, mechanical instability, or operational risk dominates system behavior (Shenhar et al., 2005; Vago et al., 2017).
Reliability represents an equally critical challenge, as recovery from downhole failures is severely limited in robotic planetary missions. In terrestrial drilling, stuck pipe is a well-characterized failure mode associated with borehole instability, cuttings accumulation, and differential sticking, and is commonly mitigated through overpull, circulation, or casing strategies (Isambourg et al., 1999). Under planetary mission constraints, however, such mitigation approaches are unavailable or severely limited by the available in situ hardware and mass constraints. Field and laboratory demonstrations of planetary drilling systems have shown that tool immobilization, borehole collapse, and loss of cuttings conveyance can rapidly terminate drilling operations, particularly as depth increases and continuous cuttings removal is not maintained (Shenhar et al., 2006; Wippermann et al., 2020). These risks are further amplified in low-cost missions, where redundancy, contingency operations, and extended surface infrastructure are necessarily limited. Collectively, these factors have constrained demonstrated subsurface access to depths well below those required to fully address key astrobiological and geologic objectives.
The challenges outlined above motivate the exploration of alternative drilling architectures that better align with the power, mass, reliability, and cost constraints of Mars missions. The present study, however, focuses on a wireline, downhole-actuated variant of the Deep Access Subsurface Extraction & Retrieval (DASER) system. In this architecture, the primary drilling forces, energy conversion, and cuttings transport mechanisms are located at the bottomhole assembly (BHA), rather than being transmitted through a surface-driven drill string. This approach significantly reduces limitations associated with top-driven systems, in which reaction force requirements, drill-string buckling, and total system mass scale unfavorably with depth (Howe et al., 2022; Tosi et al., 2024). Although the wireline mass itself also scales with depth, it does so more favorably than a torsion- and compression-loaded connection to the surface (e.g., drill pipe).
At the same time, a wireline architecture provides limited mechanical capability to recover from stuck pipe, stuck BHA, or failed cuttings-retrieval cycles. To address this limitation, the DASER concept explored here employs pneumatic cuttings extraction through a deployable conduit, which enables a single-trip BHA to drill to total well depth without the need for repeated bailing operations. The same conduit can also provide pressure support to the wellbore wall, mitigating collapse in weak or unconsolidated formations. This approach is enabled by the use of the martian atmosphere as the working fluid, together with an efficient compression system and a concept of operations (ConOps) designed to minimize power usage. The integrated DASER architecture should, therefore, be interpreted as a low technology readiness level (TRL) concept, approximately TRL 2 at the system level, rather than as a validated flight design. Though we do not attempt to describe efforts to optimize fully all aspects of this architecture, we provide an evaluation with regard to whether the mass and power requirements for a wireline pneumatic drilling system compatible with low-cost Mars mission constraints can close under representative operating assumptions.
The remainder of this article is organized as follows. The Mission Architecture section presents the overall mission architecture and science-driven requirements and introduces the DASER drilling system, including subsystem-level design choices, actuation methods, pneumatic cuttings transport, wellbore pressure-support concepts, and interfaces with the surface platform. The Primary Hardware and Concept of Operations section describes the primary flight hardware and ConOps, including the surface station, atmospheric CO2 capture and compression subsystem, downhole drilling assembly, deployable sealing membrane, power delivery architecture, and nominal sol-level operational sequencing. The Modeling Architecture section introduces the reduced-order modeling architecture that links surface resources, downhole flow, drilling and cleanout cycles, and mission-level resource accounting. The Modeling CO2 Capture, Compression, and Liquefaction and Modeling Drilling, Cuttings Removal, and Wellbore Support sections develop the atmospheric CO2 acquisition and drilling/cleanout models used to establish mass and energy requirements under Mars-relevant operating conditions. The Full Mission Simulator section integrates these models into a mission-level simulator that enforces realistic sol-level constraints on time, energy, battery usage, and working-fluid mass, and presents results that demonstrate achievable drilling depth and resource scaling for InSight/Mars Life Explorer-class missions. Finally, the Conclusion section summarizes the implications of this architecture for future Mars subsurface exploration missions and outlines key areas for further system-level development, testing, and validation.
Mission Architecture
Subsurface access mission requirements
The proposed mission architecture is intentionally constrained by a stationary Mars lander consistent with New Frontiers-class mission resources. The primary system-level baseline is the Mars Life Explorer (MLE) mission concept, a Planetary Science and Astrobiology Decadal Survey mission concept study. The relevant New Frontiers-class cost frame is the Phase A-D PI-managed mission cost cap of $850 M FY2015, excluding launch vehicle and contributions, represented in the MLE study as $1.1B FY25 (National Aeronautics and Space Administration, 2016,2021). MLE provides a recent, integrated reference for lander architecture, power system design, aeroshell constraints, and surface operations under Mars-relevant environmental conditions (National Aeronautics and Space Administration, 2021), as well as a drilling platform. InSight and Phoenix are used as flown heritage missions to bound operational realism and environmental sensitivity, rather than as direct drivers of new mission requirements.
Operational heritage from InSight further constrains feasible deployment and sequencing of subsurface hardware. Analyses of its robotic arm activities, regolith interaction, and long-duration surface operations demonstrate that subsurface interaction on Mars is highly sensitive to force limits, sequencing, and power margin, particularly when operating over extended mission durations (Zacny et al., 2009; Golombek et al., 2023a). These constraints motivate a conservative approach in which subsurface access hardware is deployed incrementally from the lander deck and operates within the force and power envelopes demonstrated by InSight-class systems.
The MLE concept employs a solar-powered lander architecture with rechargeable lithium-ion batteries to buffer diurnal power availability and support nighttime operations. The baseline MLE mass budget allocates ∼21 kg to onboard batteries, together with a solar array sized to provide an average daily energy availability of order 3300 Wh per sol under nominal conditions (National Aeronautics and Space Administration, 2021). This power architecture is adopted directly herein and defines the upper bound on sustained electrical power available for subsurface access activities, including drilling, cuttings transport, sample handling, and analysis.
Mission timing and latitude introduce additional constraints on power and subsurface access. InSight, which landed at ∼4.5
Subsurface interaction heritage from InSight’s HP3 experiment further informs the architectural requirements. Although HP3 did not achieve its intended penetration depth, it clearly demonstrated that meter-scale subsurface access is highly sensitive to regolith mechanical properties, frictional coupling, and deployment conditions (Wippermann et al., 2020). These observations motivate an architecture that minimizes repeated insertion and retrieval cycles and avoids reliance on unsupported borehole stability.
Science-driven subsurface requirements are derived from the references discussed in the Science objectives subsection , which support access to depths that exceed 30 m as a robust threshold for advancing biological and geochemical investigations beyond surface-altered materials. Extensibility toward depths approaching 100 m is treated as a reachable stretch goal. The KISS workshop also emphasizes the importance of depth-resolved sampling, contamination control between sampling intervals, and preservation of sample provenance, particularly when intact cores are not returned (Perl et al., 2024).
Consistent with these constraints, the architecture considered here assumes a single-trip drilling system that returns cuttings to the surface rather than intact cores. This choice represents an architectural trade rather than a science preference and necessitates explicit consideration of cuttings handling, depth tagging, and contamination control. In particular, sterility and sample alteration risks are treated as time-dependent, motivating requirements on the duration of bottomhole-to-surface transport and subsequent surface processing. These considerations are reflected in the mission-level requirements summarized in Table 2 and are discussed further in the context of system architecture and operations in subsequent sections.
Proposed Mission and Science Requirements for an InSight/MLE-Class Subsurface Access Lander, with Phoenix and the KISS Workshop Used as Bounding References Where Applicable
KISS, Keck Institute for Space Studies; MLE, Mars Life Explorer.
Proposed drill system architecture
To construct a wellbore that exceeds 30 m in the martian subsurface, the system architecture must overcome the mass constraints of robotic missions by minimizing the depth-to-system-mass scale coefficient (Badescu et al., 2019c, 2019b). Conventional load- and torsion-bearing drill strings typically require adding heavy segments to reach greater depths, making them untenable for depths much greater than 10 m. A wireline or coiled-tubing architecture, in which the BHA is suspended on a lightweight tether or tube, is more favorable for this purpose and has been adopted by multiple deep-access concepts, including Auto-Gopher and RedWater (Badescu et al., 2013, 2019b; Mellerowicz et al., 2022; Palmowski et al., 2022).
The key mechanical penalty of this architecture is limited weight-on-bit (WOB), because the drill cannot react thrust forces through a massive lander or rigid drill pipe (Bar-Cohen and Zacny, 2021). This constraint conflicts with the expected high-strength basalt and ice-cemented regolith targets, which require high static WOB for efficient conventional rotary drilling (Han et al., 2005; Paulsen et al., 2011; Golombek et al., 2023a). Under these conditions, percussive drilling is preferred because high-frequency impulsive blows fracture rock with substantially lower static axial load than rotary crushing (Teale, 1965; Bruno, 2005; Zacny et al., 2008a; Paulsen et al., 2011; Zacny et al., 2013).
The second architectural constraint is avoiding repeated BHA retrieval. Conventional wireline tripping for cuttings disposal exposes the borehole to collapse, debris accumulation above the tool, and permanent downhole entrapment (Isambourg et al., 1999; Zacny et al., 2008a; Badescu et al., 2019b). Recovering a stuck BHA would require a load-bearing tether and surface actuation capable of delivering high hook-loads, adding mass and power requirements that are incompatible with constrained robotic missions (Bar-Cohen and Zacny, 2021). The architecture, therefore, adopts a single-trip BHA capable of reaching total depth without repeated retraction cycles (Palmowski et al., 2022).
Single-trip drilling requires continuous cuttings transport. Mechanical augers are prone to choking and jamming at depth because friction and cuttings expansion can drive power consumption beyond spacecraft limits (Cannon et al., 2007; Guerrero, 2008; Zacny et al., 2008a). In the proposed pneumatic circulation system, compressed gas is flushed through the drill-bit to entrain cuttings and return them to the surface, which reduces torque demand and limits cuttings accumulation (Zacny and Cooper, 2005; Zacny et al., 2008a; Mellerowicz et al., 2022). The same circulation path also enables internal wellbore pressure support that helps stabilize unconsolidated formations during drilling and cleanout.
The establishment of a pneumatic circulation path to the drill-bit creates a secondary capability critical for the initial phases of penetration: the ability to jet through unconsolidated surface media. By leveraging the momentum exchange between high-velocity gas and soil particles, the system functions similarly to “jet-lift dredging” or pneumatic excavation tested for lunar regolith, where gas injection into the top centimeters of soil creates high-pressure pockets that lift granular material without mechanical augering (Zacny et al., 2008a, 2008b). This pneumatic-excavation approach now has flight heritage: the Lunar Instrumentation for Subsurface Thermal Exploration with Rapidity on Blue Ghost Mission 1 pneumatically excavated lunar regolith with pressurized N2 to a depth of 1 m, while making thermal measurements at multiple depths (Nagihara et al., 2025). This result supports the practical use of gas-jet excavation in loose planetary regolith, but it remains a shallow lunar analog rather than a demonstration of deep martian pressure-supported drilling. When coupled with a robust deployment mechanism capable of reacting vertical loads and positioning the injector head (Tosi et al., 2024), this pneumatic jetting approach enables robust penetration of the estimated 3 to 10 meters of unconsolidated regolith overlying the target formations (Golombek et al., 2023a, 2023b), effectively casing the hole with the tool body as it advances.
However, transitioning into the deep subsurface introduces challenges associated with the martian megaregolith, which is characterized as highly fractured, porous, and permeable (Beaty et al., 2000; Cannon et al., 2007). Hydrological modeling of the martian crust suggests that near-surface permeability may be as high as
A “barefoot” section, the unlined interval between the drill-bit and the membrane deployment sub, remains a necessary feature of this architecture to allow bit advancement. Pressure management in this zone must be handled dynamically, balancing the need for cuttings clearing against gas loss into the formation. Design parameters prioritize minimizing the length of this rock-exposed face; limiting this interval reduces the surface area available for gas diffusion into the regolith and minimizes the volume of cuttings that must be cleared before they enter the safety of the membrane-lined transport tube. Because of the presumed upper bound on permeability in the formation, it is also necessary to tune the percussive drilling parameters to reduce leakage. The Open-hole section flow subsection discusses how the system can effectively be tuned to change formation permeability to an acceptable level so that internal wellbore pressure can be maintained over the mission lifetime.
Finally, a pneumatic gas system can provide a means for dust mitigation on solar panels, building on compressed-gas dust-removal flight heritage from the Mars 2020 Perseverance SuperCam Gas Dust Removal Tool (Maurice et al., 2021) and complementary evidence from intentional solar-array cleaning experiments on InSight (Lorenz and Reiss, 2015; Golombek et al., 2023a). This capability may effectively extend the operational window to near the 160 sol maximum.
Primary Hardware and Concept of Operations
The lander-side mission-critical hardware consists of two primary subsystems: the surface station and the BHA, together with the hardware that remains in the wellbore. This section briefly discusses key segments within each and how they operate together in the concept of operations (ConOps).
The hardware elements described here are at different maturity levels. The atmospheric compression subsystem is anchored in Mars-analog reciprocating-compressor experiments and associated scaling models (Veismann and Tosi, 2025). The BHA geometry, pneumatic actuation sequence, and valve/hammer behavior are informed by prior DASER and compressed-CO2 drill prototype studies (Howe et al., 2022; Tosi et al., 2024,2026). By contrast, the integrated pressure-supported cleanout architecture, deployable sealing membrane, sample-handling interface, and sol-level ConOps are concept-level designs extrapolated from component studies, heritage lander systems, and first-order sizing rather than from a fully integrated flight-like testbed.
Surface station hardware
The drilling system is deployed by using a lander-mounted robotic arm that provides positioning, surface contact, and limited reaction force during drilling operations, as shown in Fig. 1a. In this architecture, the drill assembly is mechanically supported and guided by the arm during initial placement and alignment with the surface, after which active drilling forces are generated internally within the BHA rather than transmitted through the arm structure. This configuration uses the arm for gross positioning and stability while avoiding the need for the arm to sustain continuous high axial or torsional loads during deep drilling, a strategy successfully employed by the Phoenix and InSight missions for shallow regolith interaction (Zacny et al., 2008a; Golombek et al., 2023b). By decoupling drilling actuation from the arm, the system avoids the mass and structural penalties associated with top-driven architectures and reduces sensitivity to arm stiffness, backlash, and load limits (Zacny et al., 2013; Tosi et al., 2024). The arm-mounted deployment mechanism also enables controlled engagement with uneven terrain and allows the drill to be repositioned or redeployed at multiple surface locations within the arm’s reachable workspace, increasing the probability of successful borehole initiation (Golombek et al., 2023a; Tosi et al., 2024).
Drill deployment system from the lander platform. The deployable surface casing in panel
System-level schematic of the three-stage atmospheric CO2 compression, liquefaction, storage, and delivery subsystem for the pneumatic subsurface drill.
As outlined in the Mission Architecture section, initial penetration through unconsolidated regolith is achieved by using pneumatic jetting rather than mechanical augering. In hardware terms, compressed gas is directed through the drill head to fluidize near-surface granular material using the “jet-lift” effect, clearing soil without relying on high WOB (Zacny et al., 2008a). As the drill advances through this interval, a deployable surface casing mechanism integrated with the BHA, such as a bi-STEM tube, extends over the unconsolidated section (Fig. 1b). This deployable liner provides geometric confinement and pressure support to the surrounding formation, preventing borehole collapse in the absence of hydrostatic drilling fluid pressure and isolating the flow path for efficient pneumatic cuttings transport (Cannon et al., 2007; Hossain and Islam, 2018; Tosi et al., 2024; Wang et al., 2024). At the borehole mouth, the deployed casing also forms the top seal of the pressure-supported conduit, tying the membrane-lined return path to the surface flow-control hardware. Once the deployed assembly reaches consolidated rock, the BHA progresses downward and the membrane deployment system is the only outer borehole-wall-interacting section that continues into the solid formation with the BHA.
To prevent solar-power output from degrading over time, a critical addition to the system is a pneumatic solar-panel dust-removal system. One concept, consistent with mass, power, and complexity minimization, uses a single hard tube with equidistant perforations along its length that spans from the solar-panel midpoint to its outer rim. At the midpoint, the tube is connected to a flexible pneumatic line and to the existing solar-panel deployment actuator. The flexible pneumatic line follows the harnessing for that same actuator to the lander body and provides enough slack at the center-point joint for 360
ISRU compressor
The pneumatic subsurface drill uses liquefied CO2, locally sourced from the martian atmosphere by using a dedicated compression and liquefaction system for mechanical actuation, which eliminates the need for imported consumable working fluids while minimizing contamination risks associated with terrestrial propellants. Figure 2 summarizes the corresponding three-stage compression, liquefaction, storage, and delivery architecture.
Atmospheric compression represents a dominant contribution to the system energy budget, necessitating careful selection of the fundamental compression method and system design among candidate approaches for martian application, including cryogenic, sorption-based and mechanical compression. Because cryogenic and sorption-based compression approaches are constrained by phase-change energy penalties, batch operation, strong coupling to the diurnal thermal cycle, and limited demonstrated normalized yield, mechanical compression is adopted for its continuous-flow architecture, scalability in mass throughput, and higher potential overall efficiency, despite increased mechanical complexity. At martian ambient pressures, frictional work drives efficiency losses because of the large required volumetric flow rates. In related work, we experimentally demonstrated that by strategically minimizing mechanical friction at the expense of seal tightness at low pressures, mechanical compression achieved ∼35% single-stage efficiency under Mars-analog conditions using a prototype reciprocating system. Modeled performance informed by these results indicated efficiencies exceeding 70% through optimization, which is comparable to terrestrial systems (Veismann and Tosi, 2025).
Building on these considerations, the mission concept adopts a multistage mechanical compression system to achieve the pressure levels required for downstream CO2 liquefaction on Mars (∼0.6–2 MPa, depending on ambient temperature), while remaining within acceptable power and system mass limits. The architecture uses previously validated low-friction reciprocating compressor designs in a three-stage configuration, selected as a compromise between per-stage compression ratio, thermodynamic efficiency, and mechanical complexity. Although the resulting stage pressure ratio (
To meet displacement and packaging constraints, each stage employs a multi-cylinder configuration (4/2/1 cylinders for stages 1–3, respectively). This approach improves packaging efficiency and torque uniformity relative to a single large cylinder, at the expense of increased frictional losses. The present configuration provides a closed design point with respect to mass, power, and yield that supports mission feasibility (see the Modeling CO2 Capture, Compression, and Liquefaction section), while additional system-level trade-offs remain to be explored in future work.
Intercooling is implemented between stages to manage thermal loads and improve efficiency, while inlet filtration protects moving components from particulate contamination. The high-pressure CO2 is routed through an aftercooler and subsequently condensed and stored in two 6 L pressurized tanks, providing multi-sol storage capacity and decoupled operation of the compression and drilling subsystems. The tanks are externally mounted and uninsulated, with auxiliary radiators to enhance heat rejection and facilitate liquefaction. Liquefaction requires rejection of ∼48 W during operation, which is achievable based on radiative heat transfer analysis with a radiator area of only ∼0.1 m2 in combination with the effective tank surface area, indicating high liquefaction efficiency. Each storage tank has an estimated mass of ∼1.45 kg, consistent with Mars flight heritage.
Strategic nighttime operation enables recovery of rejected frictional heat to offset thermal management requirements while exploiting favorable thermodynamic conditions for compression, including higher atmospheric density and lower temperatures. The compressor is intentionally sized to operate at low power over extended nighttime periods, which allows compression work to partially substitute conventional survival heating.
To deliver the high-pressure liquid to the drill, a compact liquid pump provides the required pressure increase and transports the fluid through a tether. Minimal additional hardware is required, such as valves and bleed lines to support auxiliary pneumatic functions, including periodic solar panel cleaning.
Sample handling
Samples from the current platform are expected to come in the form of cuttings, likely between 1 and 500
At the shallow-sampling end of Mars flight heritage, the Phoenix robotic arm excavated trenches, acquired soil and icy-soil samples using its scoop, scraper, and rasp, and delivered samples to the Thermal and Evolved-Gas Analyzer (TEGA) and Microscopy, Electrochemistry, and Conductivity Analyzer (MECA) for in situ analysis (Arvidson et al., 2009; Bonitz et al., 2009). That experience provides a useful reference for lander-side sample positioning and delivery verification, although the DASER architecture differs because material arrives as pneumatically transported cuttings from depth rather than as mechanically scooped near-surface soil.
Although not considered further in the current article beyond this section, the expected baseline sample-handling hardware stack is similar to prior ocean-world lander sample-handling concepts, in which pneumatic action transfers processed sample material into a set of instrument-fed cups for analysis (Backes et al., 2020; Mercer et al., 2022). The cups themselves sit in a lander-positioned carousel or handling mechanism that can be scaled to meet the required processing timeframe (Malespin et al., 2020; Hand et al., 2022). For the Europa Lander concept, the Collaborative Acceptance and Distribution for Measuring Europan Samples (CADMES) provides a representative architecture in which a filling station, facing the sampling system, is fed with a set of cups from a swing arm or carousel on the instrument side (Malespin et al., 2020; Mercer et al., 2022). In another pneumatic-powered system that provides a relevant baseline for the connection between a sampling/exhaust stream and the lander-side handling hardware, the dual-rasp sampling system for an Enceladus lander concept describes integrated sample-acquisition and delivery elements that can inform interface design trades (Backes et al., 2020).
A primary challenge in sample handling is accommodating the volume of cuttings generated during drilling: a wellbore of ∼0.04 m diameter drilled to 30 m in basalt can produce ∼100 kg of material (∼0.5–1 kg per sol). A concept for managing this is discussed in the KISS report, which describes a sample-management architecture that features a “Triage” or “Waiting Room” chamber (Perl et al., 2024). This intermediate stage receives cuttings from the drill and holds them in a controlled environment to prevent alteration, such as the formation of eutectic brines or the loss of volatiles. Within this triage suite, samples undergo preliminary, nondestructive “surveillance” using sensors for volatile monitoring (e.g., tunable laser spectrometry), mineralogical characterization, and imaging. This process allows the science team to assess the value of each depth interval and selectively advance high-priority samples to resource-intensive life-detection instruments, such as radiotracer incubations or chemical extraction, while archiving or discarding others to conserve mission resources.
Bottomhole assembly
The BHA, which constructs the wellbore and remains in the subsurface, contains two primary segments: the drilling assembly that carries out the percussive action and delivers gas to the drill-bit for cuttings removal, and the stowage-drive assembly that carries the downhole driving unit and the deployable membrane for wellbore sealing. The wireline tether connects the surface station to the BHA, and the sealing membrane is connected to a surface flow/pressure-control valve that feeds the sample-handling system as cuttings are recovered to the surface.
The downhole assembly is connected to the surface station via a hybrid tether that integrates fluid delivery, electrical power transmission, and data communication within a compact, mechanically robust package. Pneumatic and liquid CO2 transport is provided by a pressure-rated polytetrafluoroethylene (PTFE) line selected for chemical compatibility with CO2, a smooth internal surface, and repeated pressurization and depressurization cycles. Electrical power and data are carried by a lightweight, high-strength hybrid cable based on W. L. Gore & Associates (GORE) tethered cable technology, which combines silver-plated copper conductors and single-mode optical fiber within a low-friction fluoropolymer jacket. This construction provides high-voltage power delivery (up to 600 V), low electrical resistance, and high-bandwidth data transmission, while maintaining excellent abrasion resistance, crush resistance, and bend durability under repeated reeling and deployment. The composite tether is further enclosed within a protective outer sheath that provides mechanical isolation between the fluid line and electrical elements, limits wear against the borehole wall, and ensures stable mass and handling characteristics during both drilling and cleanout operations.
Drilling assembly
The drilling assembly portion of the BHA, shown in Fig. 3, is a compact rotary-percussive system designed to operate under limited WOB while enabling continuous cuttings removal during deep subsurface drilling. Percussive energy is delivered by a pneumatically driven hammer that impacts the drill-bit following the controlled release of pressurized gas through a magnetically latched flapper valve. Slow, continuous rotation of the bit (e.g., ∼1 rpm) is provided by an electric motor and serves solely to re-index the percussion buttons between impacts; rotation is not intended to supply primary cutting torque.
Cross-section of the drill assembly.
The drilling assembly operates in two primary regimes, drilling and cuttings removal, which are selected by adjusting the opening pressure of the flapper valve that separates the vent and hammer chambers. During drilling, liquid CO2 is delivered from the surface through the tether and enters the phase-change chamber, where embedded heaters convert the working fluid to gas. The gas then expands through a pressure-control valve that sets the phase-change expansion conditions and flows through the motor assembly into the vent chamber upstream of the flapper valve. As gas accumulates in the vent chamber, pressure rises until the flapper opening threshold is reached, at which point the valve opens and rapidly transfers mass into the hammer chamber. The resulting momentum flux accelerates the hammer toward the drill-bit, delivering an impulsive impact that fractures the formation. Following impact, the flapper valve recloses due to recoil, establishing the vent chamber closing pressure for the subsequent cycle, while a return spring repositions the hammer at its initial location. Gas downstream of the hammer is simultaneously exhausted through the bit, which provides localized clearing of cuttings at the cutting face to ensure effective rock-button contact. The dynamics of the hammer-valve interaction and associated pressure evolution are treated in detail by Tosi et al. (2026).
In the cuttings-removal regime, the flapper valve opening pressure is reduced to a minimum-pressure state well below the phase-change expansion pressure set by the upstream control valve. Under these conditions, liquid CO2 is delivered at higher mass flow rate, undergoes phase change to gas, and flows continuously through the drilling assembly and out of the drill-bit without pressure accumulation in the vent chamber. The resulting high-velocity gas stream entrains cuttings and transports them uphole through the sealed conduit to the surface-handling system. The flow dynamics and performance of this regime are discussed further in the Cuttings removal cycle subsection.
Stowage-drive assembly
The stowage-drive assembly, shown in Fig. 4, forms the structural and functional interface between the drilling assembly and the upper tethered system. This section provides reaction torque for rotary indexing of the percussive drill-bit, supplies controlled axial force during drilling (primarily to ensure drill-bit-formation contact rather than WOB), and houses the mechanisms responsible for deployment of the wellbore sealing membrane. All drilling reaction forces are closed internally within the BHA rather than transmitted to the surface platform, which enables operation under the limited WOB available in a wireline architecture. The assembly further contains internal power, control, and sensing electronics, as well as a continuous internal conduit that routes entrained cuttings from the drilling assembly into the sealed uphole transport path.
Cross-cut of the stowage-drive section.
A central feature of the stowage-drive assembly is the membrane stowage and deployment system, which occupies an annular volume that surrounds the internal cuttings passthrough, as illustrated in Fig. 4. In the representative configuration considered here, the sealing membrane is implemented as a thin polymer laminate with a total thickness of ∼25
The membrane is stored in a compact, axially distributed cartridge within the stowage-drive section and is deployed continuously as the BHA advances downward, forming a sealed conduit between the drilling assembly and the surface, with the roll-out deployment wheel also serving as the mechanical seal between the two wellbore segments. For a wellbore inner diameter of 4.13 cm and a BHA outer diameter of ∼3.8 cm, deployment over a 45 m interval corresponds to a membrane material volume of ∼0.14 L (∼0.30 L for 100 m). When packaged within an annular stowage volume with a radial thickness of ∼0.5 cm, this membrane volume can be accommodated within roughly 0.5–0.9 m of axial length for a 45 m well (∼1.0–1.6 m for 100 m), assuming realistic packing efficiencies of 35–55% appropriate for Z-fold or accordion-style storage. These bounds account for finite fold radii, interply voids, and handling tolerances rather than idealized void-free packing. As the membrane is deployed, gas and entrained cuttings exiting the drilling assembly are directed into the lined conduit through inlet ports and flapper valves, which isolate the circulating working fluid from the surrounding formation and enable continuous pneumatic cuttings transport to the surface.
Electrical power generation and storage
The available energy budget depends strongly on the landing site through both power generation and power demand, with solar-energy generation governed by local solar flux and seasonal variation, and survival-heating and compression power driven by the thermodynamic state of the ambient atmosphere, including temperature and density. Figure 5 shows representative diurnal variations in atmospheric temperature, pressure, density, and incident solar flux for candidate equatorial and polar landing sites based on InSight and Phoenix mission data, respectively. Electrical power is generated by a photovoltaic system and is either used directly by active subsystems, including science operations and drilling, or stored in onboard batteries to support nighttime survival heating and atmospheric processing under more favorable thermodynamic conditions. The lander adopts a solar array architecture with direct heritage from the Phoenix and InSight landers, which consists of two circular deployable arrays with a diameter of ∼2.2 m each and provides a total active area of ∼5.1 m2. The arrays are deployed symmetrically on opposing sides of the lander after landing and have a mass of ∼6.2 kg each (Romero-Guzmán et al., 2023), which results in a total solar array mass of 12.4 kg.
Representative diurnal climatology across seasons (solar longitudes,
Concept of operations (ConOps) illustrating a representative daily operations timeline for a deep-access mission concept. Exact timing and duration vary with local solar time, power and thermal constraints, mission phase and operational objectives, as well as orbital geometry and Deep Space Network (DSN) availability.
Assuming a conservative multi-junction photovoltaic efficiency ≥22% under martian surface conditions (Stella et al., 2006), the system is expected to generate a peak electrical power of ∼600–700 W near local noon for an equatorial landing site at
System-Level Compression Performance and Energy Budget Under Representative Martian Conditions for Selected Landing Sites and Seasons
Results assume nighttime compression and allocation of a dedicated fraction of the battery capacity.
The system employs two flight-proven lithium-ion batteries with a mass of ∼10.5 kg each and a combined capacity of 80 Ah, which corresponds to ∼2415 Wh based on a state-of-the-art specific energy of 115 Wh kg
As an additional point of reference, the MLE mission concept baseline flight system likewise adopts an InSight/Phoenix-heritage direct-energy-transfer (DET) architecture (28 V primary with auxiliary rails) with centralized switching and battery/array control implemented within a power distribution and drive unit (PDDU). The corresponding MLE Master Equipment List similarly allocates 21.0 kg (CBE) to the two Li-ion batteries and 16.1 kg (CBE) to the PDDU (National Aeronautics and Space Administration, 2021), which reflects a heritage low-voltage distribution approach intended for year-class surface operations rather than sustained multi-kilowatt pulsed loads.
High-power delivery for cuttings removal
Power delivery to downhole heaters for liquid-to-gas phase change is the primary power driver for both drilling and cleanout operations. To enable brief high-power delivery (
Concept of operations
Figure 6 illustrates the representative daily operations timeline used for this ConOps. The ConOps follows a notional diurnal operational schedule aligned with the martian day-night cycle, leveraging natural variations in temperature and atmospheric density to improve system performance. The operational cycle begins with the capture, compression, liquefaction, and storage of atmospheric CO2 for later use as the working fluid that drives the subsurface drill.
Atmospheric CO2 capture and compression are performed primarily during the nighttime period after fully charging the onboard batteries during the day, when surface temperatures are lowest and atmospheric density is highest. These conditions reduce the required volumetric flow rate for a given mass throughput and increase compression efficiency, while also improving the effectiveness of interstage cooling. Heat generated by mechanical and frictional losses during compression could be recovered and used to offset spacecraft thermal control demands, reducing or partially substituting conventional survival heating, which typically constitutes a significant fraction of the nighttime energy budget discussed in the Budgets subsection. Thus, nighttime operation of the compression system offers a strategic advantage for using the constrained energy budget effectively. Future mission trades may explore shifting the compression process into daylight hours to directly use solar power generation and reduce battery mass requirements. During noncompressing nighttime periods, conventional survival heating is employed to maintain the spacecraft’s internal temperature limits. The high-pressure compressor output is delivered to externally mounted tanks to condense and store the fluid for later use. After sunrise, passive solar heating increases the temperature and pressure of the stored CO2 within the sealed tanks, raising the available working pressure without additional electrical power input. In parallel, batteries are recharged throughout the day using the lander’s photovoltaic system. As dust accumulation on solar arrays has been shown to significantly reduce power generation over time for previous Mars missions, periodic pneumatic cleaning could enable a sustained high-energy budget.
As batteries recharge during the morning hours, primary sample handling and science operations are conducted. Periodic ultra-high frequency (UHF) communications sessions using orbiter relay are scheduled for uplink and downlink of science and engineering data. These events are brief (typically 5–15 min per pass) and can be executed alongside other activities provided power demand remains within operational limits.
Drilling operations are performed in the afternoon during peak solar-power availability, when tank temperatures and pressures are elevated by solar heating and ambient atmospheric pressures are lowest. Fully decoupling atmospheric compression from drilling operations provides operational flexibility in mass flow rate, pressure, and duty cycle for more effective drilling action and cuttings removal, while also leveraging favorable atmospheric conditions.
Modeling Architecture
The hardware elements and concept of operations described above define a coupled resource, flow, drilling, and scheduling problem. The modeling framework is, therefore, organized to connect surface CO2 acquisition and power availability to downhole thermofluid behavior, drilling and cleanout-cycle performance, and mission-level resource accounting.
Figure 7 summarizes the information flow used throughout the reduced-order analysis. Environmental and power states determine CO2 capture, compression, liquefaction, and stored working-fluid inventory. These surface resources set the boundary conditions for the downhole thermofluid models, including liquid delivery through the tether, liquid-to-gas phase change, electrical transmission losses, gas transport, particle entrainment, wellbore pressure support, and formation leak-off. The resulting cycle-level quantities are passed to the mission simulator, which accumulates depth progress, energy use, CO2 consumption, active time, idle time, and battery usage over successive sols.
Reduced-order drilling and cleanout model guide showing the information flow from surface resources and CO2 supply through the downhole thermofluid and transport models to cycle-level and mission-level outputs. The shaded region identifies the coupled surface-to-downhole model whose state variables and outputs are passed to the mission simulator.
The remainder of the modeling analysis follows this architecture from surface resources to mission outcomes. First, the atmospheric CO2 capture, compression, and liquefaction model defines the available working-fluid and energy budget. Next, the drilling, cleanout, and pressure-support models define the downhole state and cycle-level resource demands. Finally, the mission simulator integrates these outputs over sol-level operational constraints to evaluate depth progress and system feasibility.
The framework is intended as a reduced-order sizing and coupling analysis. Component-level inputs are tied to prior compressor and DASER prototype studies where available, while integrated pressure-supported cleanout, membrane sealing, permeability control, and mission-level operations remain concept-level assumptions that require future validation.
Modeling CO2 Capture, Compression, and Liquefaction
The compressor energy and mass budget is evaluated by using a reduced-order simulator (Compressor Simulator appendix) that couples environmental conditions with thermodynamic compression relations, geometric design parameters, and experimentally informed compressor-scaling laws.
The energy budget available for atmospheric compression is determined for the selected landing site by accounting for local and seasonal solar irradiance, solar panel performance, and reserve margins associated with battery storage and additional operational power draws.
Model inputs in this section combine three levels of support. Compressor leakage, friction, and stage-mass scaling are tied to Mars-analog low-pressure compressor experiments (Veismann and Tosi, 2025). Drill-cycle timing, BHA pressure levels, and internal pneumatic restrictions are based on prototype-informed DASER drill studies (Howe et al., 2022; Tosi et al., 2024,2026). Integrated quantities that have not yet been validated in a flight-like system, including long-duration pressure-supported cleanout, membrane sealing and deployment, permeability-control effectiveness, and high-voltage pulsed-power delivery, are treated as concept-level assumptions and are varied through reduced-order sizing rather than calibrated against integrated hardware data.
Compressor performance is modeled using a multistage polytropic framework (Rapp and Hinterman, 2023; Rapp, 2018) that conservatively captures nonideal thermal behavior while remaining suitable for system-level analysis. Geometric and operational parameters determine effective displacement and volumetric performance, while leakage and friction losses are incorporated through scaling relations derived from prior low-pressure compressor experiments using geometric similarity. Stage mass is similarly scaled relative to the same reference design, with appropriate margins applied. The total system mass is computed as the sum of the individual stage masses and auxiliary components, including the motor, drive shaft, storage tanks, valves, and associated hardware.
The full set of governing relations and the evaluation algorithm are provided in the Compressor Simulator appendix. While the model fidelity can be expanded in future work to explore alternative architectures and system-level trade-offs, the present formulation is sufficient to capture first-order performance trends. Model results indicate that, for a fixed energy input, nighttime compression yields ∼30% higher CO2 production than daytime operation, even without accounting for reduced survival heating during this period. Although this advantage is not strictly required during the early mission phase when energy margins are high, progressive tightening of the energy budget over mission lifetime motivates nighttime operation as the baseline mode. Accordingly, atmospheric compression is assumed to occur exclusively during nighttime hours using a fully charged battery while maintaining minimum state-of-charge constraints, with optional daytime CO2 production and extended drilling permitted early in the mission during periods of energy surplus.
System-level results for the baseline design, including total power draw, operating duration, processed mass, and overall efficiency, are summarized in Table 3. For the selected landing sites, early-mission energy availability is relatively consistent at ∼4 kWh per sol, providing sufficient margin to support atmospheric compression, with degradation expected to be less severe than in comparable missions due to periodic solar panel cleaning. The energy available for compression is defined as a fixed fraction of the battery capacity after accounting for operational margins and ancillary loads.
The compressor operates at an average power of ∼130–150 W, which is consistent with system power limits and sized to favor extended runtime at low power, partially substituting survival heating, rather than maximizing peak efficiency. Under this operating regime, a total CO2 production of ∼6 kg per sol is achievable, depending on the landing site and season. During early mission phases, additional daytime compression may be enabled during periods of energy surplus, while later in the mission, production can be reduced as available energy becomes more constrained. Estimated frictional losses (friction & motor inefficiencies) are comparable in magnitude to nominal survival heating requirements, which indicates that rejected compression heat may partially offset thermal management demands when appropriately utilized, which further increases the liquid CO2 yield per sol. The total mass of the baseline design is estimated at ∼26 kg, with the majority of the mass concentrated in the first and largest compression stage. A breakdown of the mass is provided in Table 4.
Mass Breakdown of the Atmospheric Compression and Storage System
Modeling Drilling, Cuttings Removal, and Wellbore Support
This section develops a reduced-order modeling framework for the coupled processes of percussive drilling, pneumatic cuttings transport, and wellbore pressure support. The model architecture is organized around the physical flow path of the BHA, shown schematically in Fig. 8. The diagram defines both the geometric segmentation of the system and the sequence of thermodynamic and fluid-dynamic transformations experienced by the working fluid as it circulates between the surface station and the bottomhole.
Bottomhole assembly flow path diagram.
Because no integrated, flight-like DASER cleanout test has yet been performed, the model should be interpreted as a system-sizing framework rather than a validated predictive simulator. Component-level prototype work informs the internal BHA pneumatic cycle and pressure ranges, while annular cuttings transport, pressure-support leak-off, membrane sealing behavior, and permeability reduction are represented with conservative reduced-order closures that require future integrated testing.
As illustrated in Fig. 8, the flow path may be partitioned into three primary regions: (1) the internal BHA flow path, which includes the phase-change chamber, vent chamber, flapper valve, hammer, and drill-bit exhaust; (2) the open-hole annular section, where gas exits the bit, entrains cuttings, and interacts directly with the surrounding formation; and (3) the sealed-hole annular section, which provides a pressure-supported conduit for upward transport to the surface. Liquid CO2 is delivered from the surface through the tether to the BHA, converted to gas at controlled pressure and temperature, and subsequently exhausted into the open-hole section. The returning gas-cutting mixture flows upward through the open-hole and sealed-hole regions to a surface outlet, where back-pressure regulation provides wellbore support. Abrupt expansions, internal restrictions, distributed wall friction, particle entrainment, and porous leak-off act at different points along this path and motivate the reduced-order models introduced below.
The reduced-order drilling and cleanout model relies on a shared closure set, detailed in the Flow physics appendix subsection. That appendix material defines the carrier-flow state, particle-transport response, pressure-support target, formation-loss estimate, restriction losses, liquid-delivery pressure drop, and phase-change energy terms required by the section-wise models below. The main text, therefore, focuses on how these relations are coupled into section-level and cycle-level models, while the governing relations, parameter definitions, and algorithms are collected in the Reduced-Order Drilling and Cleanout Model Relations appendix.
Section-wise models
The reduced-order physics relations collected in the Flow physics appendix subsection are assembled into section-wise models by mapping each physical region of the pneumatic circuit to (i) a constant-area compressible flow segment (Fanno formulation), (ii) particle entrainment integration over the resulting axial flow field, and (iii) localized junction closures (abrupt expansion and orifice relations) at interfaces between segments. In the open-hole section, formation leak-off is added as a porous-reservoir sink term based on a conservative radial Darcy-flow approximation. In all cases, the Fanno-flow solution fields are treated as internal state variables that define the carrier-gas profiles required by the particle acceleration model, rather than as externally reported outputs. The following subsections summarize model construction and enumerate the reported outputs used by the drilling-cycle and cuttings-removal-cycle algorithms.
Open-hole section flow
The open-hole section is modeled as an annular constant-area compressible segment with an upstream boundary condition inferred from the sealed-hole return path across an abrupt expansion junction. The open-hole annulus geometry is defined by
Geometric and Boundary-Condition Relations Used to Define the Sealed-Hole Annulus and Compute the Outlet Mach Number for Constant-Area, Adiabatic, and Compressible Flow
Junction Relations for Coupling Flow Segments: (i) Borda–Carnot Sudden Expansion (Constant-Density Loss) and (ii) Compressible Orifice Flow (Choked and Unchoked)
Particle entrainment is evaluated by integrating the one-dimensional particle equations of motion (Table A8) over the open-hole axial flow field using the local interpolated carrier-gas properties and velocity. This yields the particle exit time and exit state, which are used for assessing transport feasibility and determining the required flow-time margin when the global value is not externally prescribed.
Particle Entrainment and Trajectory Relations Used to Compute Particle Position, Velocity, and Exit Time in a Prescribed 1D Gas Flow Field
Leak-off into the formation is modeled as a porous-reservoir sink using the radial Darcy-flow coefficient with an isothermal compressible-gas density correction (Table A9). The open-hole inlet pressure is applied as a conservative driving pressure level for leak-off over the full open-hole length, yielding a total leak-off mass flow rate.
Porous/Reservoir Leak-Off Relations Used to Estimate Compressible Gas Loss from an Open-Hole Section into a Permeable Formation
Sealed-hole section flow
The sealed-hole section is modeled as an annular constant-area compressible segment surrounding the tether. The annulus geometry is defined by
Particle entrainment is evaluated by integrating the particle equations of motion (Table A8) over the sealed-hole axial flow field. When the global flow time
BHA section flow
The BHA flow model provides the coupling between the wellbore annulus state and the internal drill pneumatic circuit and supplies the thermodynamic power requirements for phase change and transmission losses. The BHA is treated as a sequence of localized restrictions represented using compressible orifice relations (Table A10). First, the drill-bit exhaust is modeled as an orifice that maps the open-hole inlet state to an upstream pressure level, yielding the exhaust-chamber-side pressure
Thermodynamics within the phase-change chamber are modeled by computing the heater power required to (i) vaporize the incoming liquid CO2 at the controlled chamber pressure and (ii) condition the vapor to the target stagnation temperature (Table A12). The corresponding energy per cycle is computed by using
Incompressible Liquid-CO2 Tether Flow Relations Used to Compute Pressure Loss from
Phase-Change Chamber Thermodynamics and Electrical Transmission-Loss Relations
Thermophysical properties are evaluated using CoolProp.
Drilling/Cleanout cycle with wellbore pressure control
A complete drilling operation is composed of a repeated sequence of percussive drilling followed by pneumatic cuttings removal, with both processes coupled through wellbore pressure control. In this framework, drilling proceeds through a multi-impact cycle that advances the bit by a prescribed depth increment, after which a cleanout (blow-out) cycle is executed to evacuate accumulated cuttings from the wellbore. The defining system parameter that governs this sequence is the cleanout depth interval, which sets the ratio between the number of percussive impacts and the frequency of cuttings removal events. This ratio directly influences (i) the effective near-wellbore permeability through particulate generation and compaction, as described in the Permeability control appendix subsection, and (ii) the overall energy consumption of the system, since the cleanout cycles dominate the pneumatic mass flow and phase-change heater power requirements.
The cleanout depth interval must, therefore, be traded against drilling efficiency. Increasing the interval reduces the frequency of energy-intensive cleanout events, but allows cuttings to accumulate within the annulus, progressively attenuating impact energy transfer from the bit to the formation and reducing penetration efficiency. Conversely, more frequent cleanout improves impact effectiveness and permeability control at the cost of higher gas consumption and electrical energy expenditure. Throughout both phases, wellbore pressure support is maintained by regulating the back-pressure at the surface outlet of the sealed-hole section, which ensures that internal pressures remain above collapse thresholds while enabling controlled transition between drilling and cleanout modes. The following subsections formalize the multi-impact drilling cycle and the cuttings removal cycle within this coupled framework.
Multi-impact cycle
The multi-impact drilling cycle aggregates the per-strike performance of the pneumatic percussive mechanism into a depth-advanced drilling interval of size
Master Parameter List for Modeling Drilling, Cuttings Removal, and Wellbore Support
Quantities derived from reduced-order models are referenced by governing equations.
BHA, bottomhole assembly.
For a prescribed mass flow rate
The energetic cost of each strike is dominated by the phase-change heater power required to vaporize the incoming liquid CO2 and condition the vapor to the target stagnation temperature
Penetration per strike is obtained from the definition of mechanical specific energy, using the impact energy
Cycle-integrated quantities, which include heater energy, transmission-loss energy, actuation energy, and CO2 mass consumption, are computed by integrating the corresponding per-strike or per-time quantities over
Cuttings removal cycle
The cuttings removal (cleanout) cycle models a sustained pneumatic blow-out event in which CO2 gas is routed through the drill-bit to entrain accumulated cuttings and transport them to the surface through the sealed conduit. The cleanout model is implemented as a coupled pneumatic circuit that assembles the flow-physics building blocks defined in the Flow physics appendix subsection into three section-wise components: sealed-hole transport, open-hole transport with formation leak-off, and BHA internal pressure levels and thermodynamics. In this subsection, the governing relations are not re-derived; instead, the cleanout cycle is defined by its coupling logic and reported outputs.
For a specified cleanout mass flow rate
In the particle transport model, a representative cuttings diameter of
The particle-trajectory calculation is treated as one-way coupled, as detailed in the Particle entrainment flow appendix subsection. This assumption is used as a dilute-transport screening model: the particle density is retained in the trajectory dynamics, while feedback of the transported solids on the carrier-gas pressure field is neglected. Dense cuttings beds or slugging events would require a two-way or multiphase transport model and are outside the present sizing analysis.
Formation leak-off is computed by using the porous/reservoir approximation in the Flow physics appendix subsection. Consistent with the conservative treatment used in the code, the inlet open-hole pressure level is applied as the driving pressure over the full open-hole length to obtain
Leak-off is, therefore, treated as a mass-balance and control-authority constraint rather than as an automatic collapse condition. If the supplied flow can satisfy
Throughout the cleanout cycle, wellbore pressure support is maintained by regulating the surface back-pressure at the sealed-hole outlet, as described in the Wellbore pressure control appendix subsection. This control keeps internal pressures near the selected support target, which is high enough to support weak intervals while remaining below the derated overburden-based upper bound, and allows sufficiently high annular flow rates to mobilize and evacuate cuttings.
Thermodynamic power consumption during cleanout is dominated by the phase-change chamber heaters that are required to fully vaporize the incoming liquid CO2 and condition the vapor to the target stagnation temperature. Heater power and energy, as well as resistive transmission losses in the electrical tether, are computed by using the phase-change thermodynamic relations in the Flow physics appendix subsection. The delivered liquid pressure at the BHA is computed from the incompressible tether pressure-drop model. Finally, the total CO2 mass consumed during a cleanout event is computed from the sum of transported flow and leak-off (Eq. (107)) integrated over the cleanout duration (Eq. (108)).
Cleanout-Cycle Bookkeeping Relations Used to Compute Total CO2 Mass Consumption from the Coupled Pneumatic Circuit Outputs
Full Mission Simulator
The reduced-order closures in the Reduced-Order Drilling and Cleanout Model Relations appendix and the cycle models developed in the section-wise and cycle-level model subsections describe the instantaneous behavior of drilling, cuttings removal, and wellbore pressure support under prescribed operating conditions. However, the feasibility of deep subsurface access on Mars is ultimately determined at the mission level, where these processes must be executed repeatedly under strict constraints on available time, electrical energy, battery usage, and working-fluid mass. This section, therefore, extends the cycle-level models into a mission-level simulator that evaluates cumulative depth progress and resource consumption over successive sols.
Using the model architecture summarized in the Modeling Architecture section, the mission simulator explicitly couples drilling and cleanout cycles with sol-level scheduling, charging constraints, and idle-time behavior, while enforcing maintained wellbore pressure support throughout operations. This framework enables quantitative assessment of how architectural design choices, particularly the selected pressure-support level, propagate into mission outcomes such as achievable depth, pacing of operations, and dominant resource limits. By comparing representative pressure-support strategies within a fixed mission envelope, the simulator provides a system-level perspective on scalability, risk mitigation, and mission feasibility that cannot be captured by cycle-level analysis alone.
Intra-sol power
Figure 9 illustrates representative intra-sol power and energy flows for equatorial and polar landing sites under the proposed ConOps (see the Concept of operations subsection). During nighttime, the battery state of charge (SOC) decreases as atmospheric compression and residual survival heating dominate the power demand, reaching a minimum of ∼40% SOC for the equatorial case, while remaining above roughly 60% SOC for the polar spring scenario due to extended solar availability and reduced survival heating requirements. Following sunrise, available solar power is initially allocated exclusively to battery recharge before science and drilling activities commence, with generation exceeding daytime demand and restoring full SOC prior to peak daytime operations, indicating substantial energy margin during the early mission phase. Science operations and drilling are temporally separated to limit peak system power draw and preserve margin for nighttime compression. Daytime energy surpluses enable complementary surface activities, including drilling and opportunistic compression during early mission stages, while maintaining sufficient margin for nighttime operation.
Representative sol-level power and energy timeline for the mission concept for selected landing sites and seasons.
Mission simulator
To evaluate the feasibility of sustained deep drilling within an InSight/MLE-class mission envelope, we implement a mission-level simulator that couples the cycle models in the Multi-impact cycle and Cuttings removal cycle subsections with sol-level resource budgets and operational scheduling. The simulator propagates depth and resource consumption over a fixed mission window of
Mission-Simulator Inputs for Daily Drilling/Cleanout Scheduling with Resource Constraints and Wellbore Pressure Support (Used in Algorithm 4)
A key mission-level control lever is the maintained internal pressure support, imposed through surface back-pressure regulation as described in the Wellbore pressure control appendix subsection. In the mission simulator, this control is parameterized using the overburden scaling factor
Within each sol, the simulator iteratively executes a drilling increment followed by a cleanout event. Drilling increments are computed using the multi-impact cycle model described in the Multi-impact cycle subsection, which returns the drilling time and associated actuation and heater energy. Cleanout requirements are depth-dependent and supplied through interpolation functions obtained from the cleanout-cycle circuit model described in the Cuttings removal cycle subsection, returning the cleanout mass flow rate, duration, heater energy, transmission-loss energy, and leak-off mass flow rate at the current depth. Sol-level resource totals are accumulated as daily CO2 mass (Eq. (110)), daily energy (Eq. (111)), and daily active time (Eq. (112)), and compared against the mission budgets in Table A17 via the feasibility conditions in Eq. (115).
A distinguishing feature of the simulator is the explicit modeling of idle time required to supply cleanout heater energy under a finite power availability constraint. If the average charging power implied by a cleanout event exceeds the allowable power during drilling (Eq. (113)), additional idle time is introduced to deliver the required energy within the combined baseline and battery-assist power limits. During this idle period, the system continues to maintain the prescribed pressure-support level and, therefore, experiences additional CO2 leak-off and heater/transmission energy expenditures, evaluated using the idle/leak-off submodel at the current depth. Finally, any remaining time within the sol is treated as end-of-sol idle time (Eq. (116)), which may contribute additional leak-off and energy draw depending on depth and whether pressure support is maintained.
Algorithm 4 summarizes the mission simulator procedure, and Tables A16 and A17 define its inputs and governing accounting relations. Subsequent subsections use this simulator to quantify depth progress, resource consumption, and sensitivity to pressure-support level across the mission window.
Mission-Simulator Accounting Relations for Sol-Level Scheduling, Resource Constraints, and Pressure-Support Target
Mass flow rate selection
The mission simulator requires a depth-dependent cleanout mass flow rate,
At each depth, candidate cleanout mass flow rates are filtered by using two constraint classes. First, particle transport is constrained by a permissible velocity band,
Among the candidate mass flow rates that satisfy both the velocity and power constraints, the cleanout schedule is selected by minimizing the total cleanout energy,
Results
Figures 10 and 11 summarize cleanout-cycle behavior over depth and mass flow rate for two wellbore pressure-support strategies: a conservative overburden-matched case (
Model results over a range of mass flow rates and well depths at
Model results over a range of mass flow rates and well depths at
For
Reducing the maintained back-pressure to
Figure 12 illustrates the competing contributions of phase-change heating and tether transmission losses. Heater energy decreases with increasing mass flow rate over an initial range as cleanout duration shortens, while transmission losses increase monotonically due to their quadratic dependence on delivered electrical power. The superposition of these effects yields a clear minimum in total energy at an intermediate mass flow rate, with the minimum shifted to lower energy values for
Representative energy breakdown for heating, transmission, and their sum at a depth of 25 m.
Figures 13 and 14 show the cleanout mass flow schedules selected by the minimum-energy criterion subject to the velocity and power constraints defined in the Mass flow rate selection subsection. Over the depth range considered, the particle velocity constraint is the dominant limiter that shapes the selected mass flow rate, while power constraints remain inactive once transport feasibility is enforced. Consistent with earlier trends, the reduced back-pressure case achieves lower total energy consumption at all depths, at the expense of longer cleanout durations and increased particle velocities.
Selected cleanout mass flow rate and resulting particle velocity and cleanout duration vs. depth for
Selected phase-change heating, transmission loss, and total energy consumption vs. depth for
Recent RedWater pneumatic chip-clearing tests provide a useful order-of-magnitude comparison but not a direct validation case (Stolov et al., 2026). Reported RedWater tests cleared chips at 0.4–1.0 g s
Taken together, these results indicate that wellbore pressure support is a first-order mission design variable: modest relaxation of the maintained back-pressure substantially reduces cleanout energy and CO2 consumption while preserving effective particle transport. This trade directly expands the achievable drilling depth and operational margin under fixed sol-level mass and energy budgets, motivating the use of reduced pressure-support strategies where formation stability permits.
Full simulation results
Figure 15 shows mission progress in depth over the operational window for two wellbore pressure-support strategies,
Mission depth progress versus sol index for
The sol-level energy accounting that drives this divergence is summarized in Fig. 16. For both pressure-support cases, total daily energy expenditure approaches the imposed budget after an initial transient. The dominant contributor throughout the mission is cleanout energy, which reflects the phase-change heater and tether transmission losses required to sustain high-mass-flow blow-out events, as described in the Cuttings removal cycle subsection. Drilling energy is comparatively front-loaded, with larger contributions during early sols and diminishing contributions as depth increases, and a larger fraction of operational time is allocated to cleanout and charging. The modulation in total energy reflects repeated intraday drill-cleanout sequences and the associated idle time required to satisfy power delivery constraints, consistent with the mission simulator logic summarized in Algorithm 4.
Per-sol energy expenditure for drilling, cleanout, and idle periods over the mission window. Cleanout dominates the daily energy budget in both pressure-support cases.
Figure 17 shows the corresponding sol-level CO2 mass accounting. As with the energy budget, CO2 consumption is dominated by cleanout operations, while drilling mass consumption is largest during early sols and decreases as cleanout becomes the pacing mechanism for daily progress. Idle contributions represent additional leak-off and associated thermodynamic overhead incurred while maintaining pressure support during charging periods and during end-of-sol idle time, as modeled using the closures in the Flow physics appendix subsection and evaluated using Algorithm 5. The growth of these idle contributions with sol index reflects the increasing maintained pressure level and leak-off propensity with depth, consistent with the pressure-support prescription in the Wellbore pressure control appendix subsection. The reduced back-pressure case exhibits systematically lower idle penalties, contributing directly to the improved depth progress in Fig. 15.
Per-sol CO2 mass consumption for drilling, cleanout, and idle periods over the mission window. Cleanout dominates the daily mass budget, with idle leak-off increasing with depth due to maintained pressure support.
Taken together, these mission-level results reinforce the system-level coupling identified across the Reduced-Order Drilling and Cleanout Model Relations appendix and the section-wise, cycle-level, and mission-simulator model subsections: cleanout operations dominate both energy and mass budgets, and the maintained wellbore pressure-support level is a first-order control lever that influences cleanout energy demand, idle leak-off penalties, and thus achievable depth under fixed sol-level constraints. In the present simulations, reducing the pressure-support target from overburden-level (
The dominance of cleanout energy also identifies an alternate operations trade that is not included in the baseline simulations. The present budget assumes that each cleanout interval is a full surface-return event, with cuttings transported to lander-side handling hardware for depth-tagged analysis. If science and contamination-control requirements allowed BHA-based screening or sampling, full cleanouts could be reserved for selected depth intervals while intervening intervals use shorter local purge events only to maintain drill-bit contact, limit cuttings-bed buildup, and preserve pressure support. This mode could reduce CO2 use and phase-change energy in proportion to the reduction in full surface-return events, but it would move sensing, sample triage, and planetary-protection functions into the BHA and reduce direct surface-instrument access to every drilled increment. Partial cleanout and in situ BHA sampling are, therefore, high-leverage mission trades rather than part of the current sizing baseline.
Budgets
Table 5 lists a notional per-sol energy budget used to bound spacecraft power generation and storage requirements for the baseline mission concept across multiple mission phases. The early-mission budget assumes aggressive utilization of available power, enabling multiple science objectives to be executed within a single sol through sustained drilling, compression, and instrument operations. As available energy decreases with seasonal progression and environmental effects, operational activity levels are progressively reduced. In the advanced mission phase, drilling and compression activities are limited to ∼80% of early-mission duty cycles, while science operations are reduced to a single primary science objective per sol supplemented by lower-order complementary measurements to comply with the required budget of 3300 Wh/sol. In the late mission phase, high-power mechanical activities are further reduced to ∼40% of early-mission levels, with science operations correspondingly curtailed. Assuming an equatorial landing near northern spring (
Estimated Energy Budget per Sol
Table 6 provides an estimated mass budget for the science payload and key surface hardware required to support the mission’s baseline subsurface-access objectives, excluding the lander bus, landing hardware, mobility hardware, and entry, descent, and landing (EDL) elements. Because mission mass budgets use different accounting conventions, the table reports subsystem-level allocations rather than inherited spacecraft totals and lists the source or sizing basis for each row. The power subsystem architecture is based on InSight and Phoenix heritage and the MLE mission concept, including solar-array, battery, and PDDU elements, augmented by a dedicated high-voltage power module to support drilling operations (National Aeronautics and Space Administration, 2021; Golombek et al., 2023a). Compressor and drilling hardware masses were estimated from reduced-order sizing and prior prototype or compressor-scaling developments, while the drill deployment mechanism and sample-handling system are concept-level subsystem estimates. Reported values include local ancillary hardware within each row, such as plumbing, cabling, small tanks, and local structural supports, but not the lander bus or EDL system. The resulting 161 kg whole-row subtotal should, therefore, be compared against payload and surface-system allocations rather than against full landed mass. For context, MLE provides a dry-lander maximum-possible-value allocation of 565 kg, with a current-best-estimate dry-lander mass of ∼450 kg. The nonscience surface-system subtotal in Table 6 is 107 kg, of which 24 kg is drilling and sample-handling hardware, or 50 kg if the compressor is included as drilling-support hardware (National Aeronautics and Space Administration, 2021; Tosi et al., 2024).
Estimated Mass Budget for the Science Payload and Key Surface Systems (Power, Compressor, Drilling, and Sample Handling), Excluding the Lander Bus (Structure, Avionics, Telecom, Thermal), Mobility/Landing Hardware, and EDL Elements
The 8 kg BHA entry in Table 6 is not directly comparable to complete flight rover end-of-arm systems. For scale, Curiosity used a 30 kg turret at the end of its robotic arm, and Perseverance used a 45 kg turret; those turret masses include instruments, sample-handling interfaces, and drill hardware (NASA/Jet Propulsion Laboratory, 2020a). The comparable DASER drilling and sampling subtotal is 24 kg, or 50 kg if the compressor is included as drilling-support hardware, and the BHA mass remains a concept-level estimate pending detailed mechanical design and qualification margins.
Conclusion
This work evaluates a concept-level architectural approach to deep subsurface access on Mars, one that directly targets the dominant failure modes that have historically limited planetary drilling depth: wellbore instability, loss of working-fluid circulation, and irreversible tool immobilization. Rather than attempting to adapt terrestrial drilling paradigms under severe mass and power constraints, the proposed wireline, downhole-actuated pneumatic system reframes deep drilling as a coupled problem of pressure management, cuttings transport, and energy allocation.
A central contribution of this study is the explicit coupling of wellbore pressure support and cuttings removal within a single, closed pneumatic circuit. By maintaining a controlled internal pressure along the sealed wellbore, the architecture provides active mechanical support to the borehole wall in weak or unconsolidated formations, mitigating collapse without reliance on heavy drilling fluids, rigid casing strings, or repeated tripping operations. At the same time, continuous pneumatic cuttings transport eliminates the need for bailing or auger-based extraction, removing one of the primary precursors to stuck tools in both terrestrial and planetary drilling experience. The resulting single-trip BHA avoids the most mission-ending failure mode for robotic drilling systems: irrecoverable downhole entrapment.
These capabilities directly respond to priorities articulated in recent Planetary Science Decadal Surveys and KISS studies, which consistently identify deep subsurface access as a critical enabling technology for astrobiology, climate reconstruction, and planetary evolution science, while emphasizing risk reduction, operational robustness, and scalability under flight constraints. In particular, the Decadal emphasis on accessing well-preserved biosignatures below the radiation-altered near surface, and KISS findings highlighting drilling reliability and fault tolerance as gating challenges for Mars missions, are addressed here through architectural risk mitigation rather than incremental component hardening.
Through reduced-order modeling and mission-level simulation, this work indicates that these capabilities could be compatible with the mass, power, and operational envelopes of an InSight/Mars Life Explorer-class lander. Cleanout operations are shown to dominate both energy and CO2 mass budgets, establishing wellbore pressure support as a first-order mission design variable rather than a secondary constraint. Modest relaxation of the maintained back-pressure level, from overburden-matched support to a derated fraction, yields substantial reductions in cleanout energy and idle leak-off penalties while preserving effective particle transport. This trade directly translates into increased achievable depth within a fixed sol-level budget, as demonstrated by mission simulations exceeding the 30 m baseline target well within the nominal operational window.
Importantly, the architecture exhibits favorable scaling behavior that aligns with Decadal guidance on progressive mission complexity. Extension from the baseline 30 m depth to ∼100 m does not require fundamental changes to the drilling mechanism, but rather increased mission duration, stored working fluid, and energy throughput, parameters that scale linearly or sublinearly with depth when wellbore walls remain self-supporting. In this regime, the deployable sealing membrane and pressure-supported conduit continue to isolate the working-fluid circulation path from the surrounding formation, allowing cuttings recovery and pressure control to remain tightly coupled and operational risk to remain bounded.
Changes in mission cost cap would, therefore, be expected to modify margin and science scope more directly than the basic drilling architecture. A smaller mission class could retain the core value of the approach as a subsurface-access demonstration, but would likely trade down target depth, stored CO2 and power margin, sample-return frequency, payload breadth, and subsystem redundancy. A larger mission class would most naturally buy down operational risk and increase science return through larger energy-storage and CO2 reserves, redundant flow-control or compression elements, wider thermal and operations margins, and additional downhole or surface instruments. The 30 m baseline should, therefore, be interpreted as a New Frontiers-class reference case: lower caps would emphasize technology demonstration and selected high-value depth intervals, whereas higher caps would support greater depth, stronger fault tolerance, and broader payload accommodation.
At kilometer scales, additional architectural evolution would be required, particularly to decouple wellbore stability from cuttings recovery over long intervals. Nevertheless, the core principles analyzed here (downhole actuation, continuous pneumatic transport, pressure-supported conduits, and single-trip operation) remain directly applicable. Future architectures may segment pressure support, introduce staged sealing, or hybridize pneumatic transport with other conveyance methods, but the governing insight persists: robust deep subsurface access on Mars requires controlling the borehole environment, not merely penetrating it.
The present analysis is, therefore, best viewed as a system architecture and sizing study rather than a subsystem-level validation. Key next steps include breadboard tests of rotary-percussive penetration rate and MSE in basaltic simulants, Mars-relevant measurements of chip transport as a function of gas pressure and flow rate, sealing-membrane deployment and pressure-support tests in weak and competent media, integrated drill-cleanout demonstrations that close the coupled gas-use, leak-off, and cuttings-removal budgets under representative CO2 working-fluid conditions, and operations trades for partial cleanout or BHA-based in situ sample screening.
Beyond Mars, the concepts developed here are broadly relevant to subsurface access across planetary bodies with low gravity, low ambient pressure, and limited operational margins, including the Moon, icy satellites, and small bodies. By treating drilling, cuttings removal, and wellbore stability as a unified system rather than independent subsystems, this work supports a technology-development pathway aligned with Decadal and KISS objectives for sustained, risk-tolerant planetary subsurface exploration: transforming depth from a mission-ending risk into a manageable and scalable design variable.
Footnotes
Acknowledgments
The authors gratefully acknowledge the KISS at the California Institute of Technology for supporting the writing process, with particular thanks to Janet Seid and Antonio Soriano.
Authors' Contributions
Luis Phillipe Tosi led the conceptualization, methodology, model development, formal analysis, visualization, and original manuscript drafting. Marcel Veismann contributed to the atmospheric-compression modeling, subsystem sizing, and manuscript review and editing. Scott M. Perl contributed to science requirements, mission context, and manuscript review and editing. Kristopher Sherrill contributed to the DASER architecture, drilling-system design context, and manuscript review and editing. Scott Howe contributed to drill-system concept development, hardware architecture, and manuscript review and editing. Marcello Gori contributed to power-system and mission-integration analysis, supervision, and manuscript review and editing. All authors reviewed and approved the final manuscript.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
The research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004).
Abbreviations Used
Appendix
Associate Editor: Michael A. Meyer