Exploring, sampling, and interpreting lunar volatiles in polar cold traps

Significance

Volatile deposits associated with permanently shadowed regions, and other surface volatile deposits of the Moon are important sampling targets during the return of humans to the lunar surface. They are of high scientific value and potentially shift strategies for sustained human activity on the Moon. However, sampling these extremely cold deposits, returning the samples to Earth, and interpreting their volatile record are challenging issues for the Artemis Program.

Abstract

Numerous missions to the Moon have identified and documented volatile deposits associated with permanently shadowed regions. A series of science goals for the Artemis Program is to explore these volatile deposits and return samples to Earth. Volatiles in these reservoirs may consist of a variety of species whose stable isotope characteristics could elucidate both their sources and the processes instrumental in their formation. For example, the δD of potential contributors to the deposits can be used to identify a uniquely light solar wind component. Because of the exceptionally low temperatures of these volatile deposits, examining and interpreting their stable isotope systems to fulfill Artemis science goals through sampling, preserving, curating, and analyzing these samples are far more difficult than for other sample return missions. Collecting and preserving the samples at cryogenic temperatures dramatically increases science yield but is technologically demanding and poses increased risk during transport.

Permanently shadowed regions (PSRs), transiently shadowed regions (TSRs), and ice stability regions (ISRs) are potential lunar environments for the sequestration of both exogenous and endogenous volatiles. In the south pole region, temperatures range from 20 K to 120 K for PSRs (e.g., 1–3). These temperatures are low enough that minimal sublimation occurs, even over billions of years (4). PSRs are primary targets for lunar missions this decade (e.g., VIPER, Artemis III) and their potential as resources are game changers for sustainable human surface activities on the Moon and beyond (5). They are of high science value as recorders of degassing of the lunar interior, sources, and processes of volatile additions to airless planetary bodies and may answer vital terrestrial and Solar System questions (1, 5). An important science objective for Artemis III (6) and follow-on Artemis missions is to understand the characteristics and origin of lunar polar volatiles. Although in situ measurements will fulfill some science goals, returned samples will provide great insights into the compositional state and distribution, sources, and depositional histories of volatile deposits. Many outstanding questions remain: At what conditions should these volatile-rich samples be collected and maintained? How do the evolving capabilities of Artemis missions impact our examination of volatile deposits? What is the trade-space for sample containment-temperature control-science realization? Here, we examine sources for lunar surface volatiles (endogenous and exogenous) and their stable isotopic characteristics. We also consider the trapping of volatiles at the lunar surface in PSRs, TSRs, and ISRs and consider the stability of volatiles in cold traps. We place the preservation of this volatile record within the context of evolving strategies for collecting volatiles during Artemis Program surface activities Finally, we consider the effects of Artemis collecting strategies on preserving the volatile isotopic record, and volatile record and human sustainability on the Moon and beyond.

Results

Nature of Lunar Surface Deposits.

Trapping volatiles at the lunar surface.

The lunar poles are exceptional environments where topography and solar angle combine to create ultracold locales favorable to the sequestration and long-term retention of volatiles within the near surface, as well as the formation of metastable surface deposits (7, 8). For the upcoming Artemis and south polar surface missions, key science objectives include investigations of volatile locations, compositions, ages, sources, trapping mechanisms, loss mechanisms, cycles of deposition and loss, distribution, and volumes, as well as any interactions and chemical processes occurring within the regolith and near-surface locations containing volatiles (e.g., 5, 6). Such investigations will likely be accomplished through in situ surface measurements as well as eventual sample return.

PSRs.

PSRs are areas where there is no direct solar illumination and thus surface radiative temperatures are very low, in some places less than 40 K (3, 7). PSR locations are highly dependent on the topography and are particularly found within craters (e.g., Fig. 1) at both lunar poles (8–11), but can also be found wherever shadows are persistent (12). The low temperatures within many polar PSRs make it possible for a variety of volatile species, including water-ice and CO₂ ice, to be harbored or sequestered over relatively long-time scales, depending on the age and stability of the PSR over time, as well as secondary illumination and double shadowing (13, 14). Notably, nonpolar PSRs are too warm to be persistent cold traps for volatiles (e.g., 15, 16).

Fig. 1.

PSRs at the lunar south pole as determined from Lunar Orbiter Laser Altimeter (LOLA) topography (17) shown as red polygons. PSRs overlain on a map of average surface illumination (10) and digital elevation model (17). Whiter areas indicate more annual illumination, and darker areas indicate low amounts of illumination. Everywhere but the pure white and red pixels are TSRs. The green dot at center marks the geographic south pole on the rim of Shackleton crater. The rendered perspective looks toward Mons Mouton between Malapert and Nobile craters. Crater diameters provide scale for this image: Shackleton crater = 21 km, Faustini crater = 39 km, and Sverdrup = 35 km.

The volatiles that are thought to be potentially present within PSRs are related to their sources (e.g., solar wind, endogenous water and CO₂, carbonaceous chondrite water and organics, comet ice and organics) and the surface temperature within the PSR (e.g., 3). The surface temperatures within PSRs vary with the Moon’s orbit, with some periods (i.e., when the subsolar latitude is above the equator) having more area with low surface temperatures (7). However, the range of surface temperatures and the duration of any periods warmer than the sublimation temperature, are what most influence the stability (or residence time) and migration or loss processes occurring within the PSR (e.g., 4, 12). Furthermore, the local temperatures within a PSR are highly dependent on the local topography and illumination, and high-precision temporal thermal modeling is challenging. In addition, some PSRs receive secondary or tertiary reflected illumination or radiation from nearby terrain, such as crater walls (18–20). Finally, the past extent of polar PSRs — dependent on the Moon’s orbit — may have been nearly nonexistent prior to ~3.4 Ga (21).

TSRs.

Any temporarily, seasonally, or TSR receives partial solar illumination (e.g., 11). Similar to PSRs, TSRs can occur outside of the polar regions, such as on the night side or adjacent to boulders. However, at the poles, the continuously low Sun elevation (from the perspective of someone on the lunar surface) results in many shadowed areas. They are highly influenced by the shape of the nearby terrain, and the positions of these shadows vary over time. Only a few areas near the poles receive illumination for more than 50% of the time (either a yearly average or averaged over multiple years) (Fig. 1), with most areas near the poles receiving much less (8, 10, 17, 22).

ISRs.

A volatile stability region is a region where temperatures are low enough for a particular volatile phase to persist for long durations (e.g., 23). For the lunar case, an ISR is an area where temperatures are ~100 K or less, sufficiently cold to preserve water-ice on the million-year timescale (e.g., 3, 24). An ISR could occur at the surface of the Moon or at depth within the regolith (e.g., 3, 23, 25–31). Thus, while an ISR can theoretically harbor water-ice, further measurements and observations are needed to determine whether water ice is actually present. Presently, the LCROSS impact is our most direct measurement of the volatile contents of a PSR (1), but it is a single point, and PSRs and ISRs are potentially highly variable in their volatile contents. Because of the highly variable conditions at spatial and temporal scales near the poles, different ISRs might have dissimilar expressions of water-ice, if any is present at all (e.g., 2, 23, 25, 32). A variety of processes could lead to an ISR having lost some or all its volatiles. Many PSRs, both large and small, are thought to host ISRs (Fig. 2). TSRs, on the other hand, might host surface volatiles on shorter timescales or might have volatile stability zones at depth (e.g., 3, 4, 26, 30). Other volatiles with lower sublimation temperatures than H₂O (e.g., CO₂ ice) will have corresponding stability zones based on the surface temperature over time, and because there is less surface and near surface area with lower temperatures, the extents of those stability zones are also less (4) (Fig. 3). Carbon dioxide (CO₂) and ammonia, for example, require temperatures as low as 62 and 66 K, respectively, for thermal stability (26), limiting their possible preservation to only the coldest parts of PSRs, such as the floors of Amundsen, Haworth, de Gerlache, and Faustini (26, 30). Notably, these areas are not adjacent to or within the more easily accessible Artemis III candidate landing regions and would thus require long or challenging traverses to reach.

Fig. 2.

Polar stereographic map of the south polar region. (A) Map of mean summer surface temperatures at the lunar south pole determined from Diviner observations (7). (B) Stability regions at the south pole calculated from time-integrated maximum surface temperatures (data from ref. 26). Colors correspond to different chemical species stable at different surface temperatures: sulfur (202 K), water (107 K), ammonia (66 K), carbon dioxide (62 K), and pink outlines indicate boundaries of large (>1 km²) PSRs (17). Basemap is the Lunar Reconnaissance Orbiter Camera (LROC) Wide Angle Camera (WAC) South Pole Summer Mosaic (33). (C) Map of ISRs, over WAC South Pole Summer Mosaic, based on mean time-integrated temperatures as well as sublimation rates for water-ice (data from ref. 4). (D) Map of water ice stability with depth (m) in the subsurface (data from ref. 4). The spatial relationship between Fig. 1 and this figure is illustrated with crater names indexed in 2D.

Fig. 3.

Stability of volatile species in PSRs under cold trap conditions on the lunar surface (e.g., 4, 24), melting, and boiling points of volatile species at 1 atmosphere, stability of ices in vacuum-sealed conditions for 11 d during an Aretemis mission, and collection-storage-curation conditions currently available based on previous ANGSA studies and proven ISS capabilities. Sublimation point for the various species represents the temperature needed to retain volatiles at the lunar surface for billions of years. Melting and boiling point temperatures and reactions represent conditions required to maintain the species at 1 atmosphere. Stability of ices and conditions of collection, storage, and curation reflect strategies discussed in the text. Approximate transition from amorphous to crystalline water ice at approximately 130 K is dependent upon ice depositional processes. Volatiles within amorphous ice water locked into its structure at 80 K (34).

Ages of volatiles.

The formation history and ages of volatile deposits are relevant to many science objectives addressing the sources of water in the Solar System. Several studies suggest that polar ISRs, and the persistence of any ice within them, will have varied over time, with some areas possibly preserving larger deposits within the subsurface (23, 31, 35). For ancient ice to persist, it needs to be protected by burial under impact ejecta or regolith. Otherwise, significant loss and destruction mechanisms including increasing temperature and illumination as well as exposure or destruction through impact cratering, space weathering, and ongoing bombardment will occur. However, recent analysis of the Moon’s orbital evolution suggests that periods of higher obliquity before about 3 Ga would correspond to a time when there were virtually no ISRs or PSRs, and any ice inside a crater must be younger than the host crater and younger than ~3 Ga (21, 36). The composition and modeling of the volatiles within Cabeus crater as revealed by LCROSS, for example, suggest ages less than 1 Ga (21, 31).

Interactions between volatiles and regolith.

While the presence of surface water-ice has been proposed for parts of the polar regions and PSRs (19, 28, 36, 37) and a global monolayer of metastable water (specifically OH) has been detected (38–41), water-ice is thought to be much more stable and longer-lasting if it is buried by at least a thin regolith covering to protect and insulate it (18, 29, 42). Most models of ice stability at depth (Fig. 2D) are based on the inferred surface temperature and sublimation rate (e.g., (4)). Presumed areas of stability are generally greater in extent if mean surface temperatures, rather than maximum surface temperatures, are considered (4, 26, 28).

Temperature variations or differentials between the surface and subsurface can drive a variety of processes including sequestration and cold trapping (e.g., 4, 25), diffusion and chemical reactions (43, 44), or chemical processes such as proton loss and oxidation (45–47). Surface volatiles are also subject to modification by radiation, solar wind influx, and changes in illumination and temperature over time (e.g., 48–51). Meanwhile, ongoing bombardment, and perhaps thermal variations, also work to cause physical and chemical changes to near surface materials (e.g., 52). Infrared spectra suggest that widespread H₂O could be stored in glass or voids between grains, possibly as part of a regolith gardening process (39). Widespread hematite spectral detections observed in the near-infrared suggest a relatively common chemical process to produce oxidation reactions in polar soils, although it could be a reaction using oxygen sourced from the Earth’s magnetotail (46).

Sources for Lunar Volatiles and Stable Isotopic Characteristics.

These aforementioned environments on the Moon potentially record volatile contributions to the lunar surface. The isotopic composition of water and other volatiles provides a fingerprint for the sources of the volatiles and the timing of their contribution. In order to discuss water in terms of lunar sources and abundance, it is necessary to define water as it applies to the Moon. Molecular water (in the form of ice) is likely to occur in the PSRs of the poles (e.g., 1, 3). It is sourced from comets, hydrous chondrites (53, 54), or solar wind and perhaps as adsorbed H₂O on minerals surfaces (45, 55). Other “so-called” water exists as structural or trace hydroxyl (OH⁻) in minerals such as apatite, feldspar, or glass (56–58) or as H⁺ or OH⁻ as the result of solar wind bombardment of the lunar regolith (59–62). In magmas, water is a mixture of molecular H₂O and hydroxyl with the proportion being a function of total water content (63). In lunar basalts, a significant amount of water is hydroxyl (63, 64). Direct degassing of water vapor during volcanism is unlikely as the oxygen fugacity of lunar basalts is so low that H₂ is the stable vapor phase rather than H₂O (65). H₂ gas evolved during volcanic degassing would likely be lost to space. The relative contributions of these different potential sources of water to the PSRs of the Moon could be better understood from the chemistry and isotopic composition of returned Artemis samples.

Analysis of Apollo samples led to the conclusion that the Moon is “bone-dry” (66) containing less than 10 ppm lunar interior water (67), although the lunar regolith was found to contain trace amounts of H, generally believed to be the result of solar wind implantation. The discovery of high H-contents in volcanic glass beads (58) raised the possibility that endogenous lunar water contents were far higher than previously assumed. High hydrogen contents have since been found in apatite (68), melt inclusions (56), and nominally anhydrous minerals (57).

The hydrogen isotope composition of lunar materials spans an enormous range, from nearly pure protium (δD < −950 ‰, 69) to over 1,000‰ for some lunar apatite (Fig. 4). The very low values are attributed to solar wind bombardment. At the other extreme, high δD values in apatite grains have been interpreted as addition from comets or chondrites (70, 71), but are more likely the result of degassing of H₂ gas prior to crystallization (65). Apatite is an extremely late crystallizing phase, such that significant vapor loss and isotope fractionation should occur prior to its formation (72). Assuming that the lowest δD values of basalts unaffected by solar wind represent the least degassed samples, Robinson et al. (73) proposed three apatite groups: a very low δD population in quartz monzodiorites averaging −630‰; a second group between −350 and −50 ‰; and a high δD group between +150 to +850‰ found in some mare basalts. Petrologic constraints argue against a solar wind contribution to the lightest samples in the first group. Instead, they may preserve the signature of primordial nebular ingassing (74). The intermediate group is plausibly the overall lunar mantle value, while the highest values are most likely affected by preferential loss of light hydrogen during cooling and crystallization.

Fig. 4.

Hydrogen isotope composition and water content of lunar apatite (53) with additional H₂ and H₂O from lunar regolith from ref. 69. The lowest δD values are related to solar wind implantation. The highest values are explained by degassing and loss of H₂ to space (65). The yellow band defines the likely bulk endogenous lunar range. Devolatilization will raise the δD of a given sample. Chondrite compilation from ref. 75; comet from ref. 76.

Evidence for proton implantation via solar wind in the lunar surface comes from both spectroscopic measurements (e.g., 41) and direct measurements of the lunar regolith (59, 60). Proton implantation can result in hydroxyl incorporation into silicates or glasses (e.g., 77) or as isolated water molecules (78). Low D/H ratios suggest that hydroxyl in lunar soil grains is sourced by solar wind rather than chondritic addition (79). He et al. (77) measured the H concentration and isotope profile in impact glass beads from the Chang’e-5 mission and found high H concentrations and δD values as low as −990‰ in the outer 10 s of microns of individual beads. The global water-equivalent contribution from the glass beads is estimated at 4 to 78 mg/g water, in general agreement with spectroscopic data (80), equal to as much as 2.7 × 10¹⁴ kg H₂O to the lunar surface (less than 1 billionth of Earth’s ocean water).

Exogenous sources (other than solar wind) of water include asteroids and comets. Modeling suggests that a cometary impact will lead to a transient H₂O atmosphere that will persist long enough for a small fraction of cometary water to be trapped in lunar cold traps (81, 82). Both cometary and chondritic (asteroidal) additions should have distinct hydrogen isotope ratios. The δD values of volatile-rich carbonaceous chondrites generally overlap with Earth values (except for rare CI, CR, and TL chondrites which have δD values between 200 and 700‰) (53). Comets are more variable. The initial spectroscopic data for D/H in Oort cloud comets gave extremely high values, negating the possibility of it being a major source for Earth’s water. Subsequent measurements of Jupiter-family comets overlapped with bulk Earth values (83). Recently, Lis et al. (76) demonstrated a negative correlation between the D/H ratio and the ratio of the active surface area to the total nucleus surface of a given comet. They suggested that all comets may have D/H ratios similar to Earth. This result has two important implications. First, it eliminates the unique isotopic signature of comets and, second, removes the constraints on the contribution of cometary water to Earth (and by analogy, the Moon).

The water residing in lunar cold traps may have distinctive hydrogen isotopic compositions that could be used to identify their source. Endogenous water likely has a δD ranging from ~−500 to as high as −50‰ (Fig. 4) if we exclude analyses of material that has been modified by extreme degassing. Solar wind sources are uniquely low, with δD values from bulk mature regolith and lunar glass bead rinds that are less than −900 ‰. Chondrite sources (exclusive of the rare CI, CR, and TL chondrites) are similar to Earth values of ~−60 ± 100‰. Cometary sources are more ambiguous and the δD values may overlap other sources or be extremely high. How can the isotope values be reconciled with measurements that will be made on Artemis return samples? Extremely low δD values approaching −900‰ should be definitive of a solar wind source. δD values greater than 0‰ are most likely a signature of chondritic or cometary material, although there are other possible explanations for high values. One is that lunar apatite, which has an anomalously high δD compared to the primitive Moon, may be preferentially volatilized during impact and provide a source for deuterium-rich water, although long-term meteorite bombardment may lead to an overall loss of endogenous water (53). A second is that migration to the lunar cold traps should lead to partial water loss (84, 85), which would result in preferential retention of isotopically heavy water. Schorghofer (86) suggested that the lower thermal speed of HDO molecules relative to H₂O would result in the retention of isotopically light water, although the effect is very small (86). Intermediate δD values may be a signature of endogenous water, although it should be noted that at the low f(O₂) of lunar samples, H₂ gas released during degassing is unlikely to be retained on the lunar surface. Finally, intermediate δD values (~−500 to −100‰) may be a mixture of a light solar source and a heavier exogenous source. If frozen samples are eventually returned to Earth, different layers may preserve different compositions, thus recording secular variations in the delivery of water to the permanently shadowed cold traps.

The Δ′¹⁷O (=δ′¹⁷O-λδ′¹⁸O) values of extraterrestrial materials are distinct. Compared to the Earth and Moon (Δ′¹⁷O = 0‰, λ = 0.528), chondrites straddle Earth values, ranging from ~ −5‰ to +3‰ (75). Oxygen isotope values of comets are extreme with the δ¹⁸O, δ¹⁷O, and Δ′¹⁷O values of the coma of comet 67P/Churyumov-Gerasimenko estimated to be ${96}_{-79}^{+91}$%, ${209}_{-90}^{+90}$%, and $139\pm 37$% (87). The Δ′¹⁷O is enormous compared to all previously measured materials and therefore even a 2-3% contribution of this cometary water would raise the overall lunar ice Δ′¹⁷O value by ~5‰, giving it a uniquely heavy signature.

Several distinct isotopic reservoirs for S have been identified. S associated with mare basalts has a remarkable degree of homogeneity (δ³⁴S = 0.58 ± 0.05‰, Δ³³S = 0.008 ± 0.006‰, and Δ³⁶S = 0.2 ± 0.2‰) (88), although there are more recent work to suggest some deviations from this in lunar pyroclastic deposits (89, 90) associated with degassing or mantle sources. There are heavy sulfur enrichments in mature regolith and light enrichments in some highland crustal rocks (91). These subtle variations in sources could be preserved in PSRs in volatile sulfur phases such as H₂S.

It has been proposed that organic molecules in cold traps might be a sensitive indicator of a chondritic or cometary source (24). The abundance of organic molecules in comets is generally less than 1% relative to water (X/H₂O in %), with the exception of methanol which may be up to several % (92). The LCROSS sampling mission detected other volatile species (e.g., H₂S, NH₃, C₂H₄, CH₃OH) in abundance relative to water that is higher than in chondrites or comets (1). Colaprete et al. (1) suggest that these molecules could form in situ on mineral grain surfaces. If this is the case, then the proportions of organic molecules to water may not indicate a unique source. Carbon isotope measurements of return samples may help resolve this issue. The δ¹³C value of organic matter in chondrites is −20 ± 10‰ (75), while spectra from 67P/Churyumov-Gerasimenko indicate far heavier values of +60‰ (93). The nitrogen isotope composition δ¹⁵N value of comets is extremely heavy at 950 ± 500‰ (94). Such high values in lunar samples would strongly indicate a significant cometary source.

Strategies for Collecting Volatiles During Artemis Program Surface Activities.

Collecting, preserving, curating, and allocating cold volatile-rich samples will evolve during the growth of the Artemis program. It is likely that Artemis III will have volatile sampling capabilities similar to the Apollo Program. During the Apollo program, special samples were isolated from the crew cabin and terrestrial atmosphere with a variety of sample containers [e.g., Gas Analysis Sample Container (GASC), Special Environmental Sample Container (SESC), Core Sample Vacuum Container (CSVC)]. All of these containers were sealed on the surface of the Moon using an indium seal (90% Indium, 10% Silver) pressed into a stainless-steel knife edge of the sample tube (95). These seals worked to varying degrees of success (87). For Apollo 16 and Apollo 17, single (69,001) and double drive tubes (73,001 to 73,002) were hammered into the lunar regolith and the drive tubes were placed in the CSVC and sealed on the lunar surface. Only one CSVC was used for each of the Apollo landing sites. Whereas for Apollo 16, the single drive tube was placed in a CSVC, for Apollo 17 only the lower part of the double drive tube (73,001) was placed in the CSVC. Below 20 cm, temperatures of the regolith collected by the double drive tube at Station 3 during Apollo 17 were estimated to be approximately 250 ± 2 K (96). These core sample containers were kept at ambient crew cabin temperature and upon their return to Earth each was placed within their own outer vacuum containers and stored in a nitrogen glove box at room temperature. Only 50 y later was the Apollo 17 CSVC opened during the NASA Apollo Next Generation Sample Analysis (ANGSA) initiative and internal gases sampled. The gases preserved were a mixture of N₂ lab gas and gases of lunar origin (95, 97, 98).

Although it is likely that similar containers and temperature conditions will be used during Artemis III, samples will be distinctly different. For the Artemis III mission, samples placed in sealed containers will contain regolith with a higher expected percentage of volatiles. This has profound implications for preservation of regolith stratigraphy, volatile–regolith interactions, and conceivably, crew safety.

Samples collected from PSRs could potentially be stored in the crew cabin at low temperatures. Temperatures of 150 K are sufficient to preserve 100 µm diameter ice crystals for 10⁶ s (11 d) (99), which would preserve ice during transport from the Moon back to the permanent storage facility. Coated ice grains would be able to survive for these time periods at higher temperatures (100). The International Space Station (ISS) has cold stowage hardware (e.g., Merlin, Polar, MELFI, Glacier, Rapid Freeze) with temperature capability down to 178 K for launch and return and 113 K for experiments on the ISS (101–103). Capabilities in missions for preserving cold samples are demonstrated by these ISS-ready freezers although mass (730 kg) and volume (exterior dimensions 50.1 cm × 55.4 cm × 45.7 cm with a volume 0.127 m³) are current limiting factors for early Artemis missions. However, Gateway (first space station orbiting the Moon that will provide access to the lunar surface) (e.g., 104) or permanent surface structures will be able to accommodate these larger freezer capacities for extended surface activities.

Cold curation of lunar samples has been limited. Apollo samples were not returned under temperature control, but a subset of samples from Apollo 17 were placed in a freezer within a month following their return (105). The samples remained in cold storage (250 ±2 K) for 50 y, although there are some records that indicate there were times at which the freezer did not maintain this temperature. During the ANGSA initiative, these samples were curated in a temporary cold curation facility at approximately 250 K and allocated samples were shipped to the ANGSA science team members at similar conditions. These were the first Apollo samples curated and allocated under cold conditions (92).

Currently, there are no permanent facilities for the curation and storage of Apollo samples in cold conditions, although such facilities are planned for the Johnson Space Center in the future. Such facilities do exist elsewhere. The Subzero Curation Facility for Astromaterials at the University of Alberta has an Ar gas glove box housed within a temperature-controlled environmental chamber capable of processing meteorite materials at temperatures between 243 to 263 K (106). Storage, curating, and study of terrestrial ice cores is routine, being done in walk-in freezers at many research facilities (e.g., NSF Ice Core Facility) (100). The NSF_ICF archive freezer is held at a temperature of 237 K and contains an examination room and a cold clean lab that are held at 248 K. A permanent Artemis cold facility would require capabilities at least at these temperature levels, and preferentially significantly lower, but within a vacuum or N₂ glovebox to prevent interactions between samples and the terrestrial atmosphere. What is the impact of collection, storage, and curation temperature capabilities on preserving the volatiles and stable isotope characteristics of the volatiles collected in PSR and other cold traps on the Moon?

Discussion

Approach to Accessing Volatiles.

The range of conditions and environments potentially hosting volatiles vary over even relatively small spatial and temporal scales. This implies that volatiles in materials near the lunar surface could have formed from, and perhaps preserved evidence of, a variety of processes, such as sublimation, chemical processing, sequestration, and trapping, gardening during meteorite impact, and recurring cycles of adsorption and loss. Some science objectives for upcoming missions focus on seeking a range of samples that reflect this expected natural variability. Others seek deposits that will best address a more focused objective such as measuring the oldest volatiles, or those with the lowest temperature stability or simply the chemistry and abundance of the bulk volatile species extracted at room temperature. Thus, the approach to measurements and sampling will need to be finely tuned to the defined objectives. Accessing distinct locations and types of environments (e.g., PSRs, TSRs, and ISRs) for sampling will result in encounters with different surface temperatures and illumination conditions, as well as different occurrences of volatile deposits (Fig. 5). Planning tools such as the Artemis Extravehicular Activities (EVA) Geographic Information System (AEGIS) can be used to precisely map out time-specific routes using input geospatial maps tailored to specific sampling objectives (107).

Fig. 5.

Simple traverses as straight-line profiles within two Artemis III candidate regions with plots showing maximum summer temperatures (7), depths to water-ice stability (4), and terrain slopes and elevations from 10-m LOLA maps (24) along the profile. On the elevation profile, boundaries of water-ice (H₂O) stability with depth from Schorghofer and Williams (4), volatile stabilities from Landis et al. (26), and PSRs (24) are plotted along the profile. Maps show ice depth stability using the color scheme of Fig. 2D. Both profiles begin on “dry” TSR terrain where water ice is not stable over the long term and move toward a nearby PSR. (A) Profile a-b on the rim of Shackleton crater. (B) Profile c–d on the rim of Faustini crater.

Therefore, initial collection of surface volatiles by humans may involve sampling a broad range of environments such as micro-PSRs, the edges of large PSRs, TSRs, and ISRs. In addition to reducing human risk in extremely hostile environments, such sampling may influence the design of capabilities required to preserve samples. Detailed investigation of the larger and more hostile PSRs may be accomplished remotely. Sampling of these volatile deposits in the foreseeable future may be accomplished using robotic technologies, although the extremely low temperatures of large PSRs exceed the capabilities of current lunar rovers such as Volatile Investigating Polar Exploration Rover (VIPER).

Consequence of Artemis Collecting Strategies on the Volatile Record.

Here, we evaluate the consequence of collecting and preservation strategies at four different temperature conditions: 1) ambient crew cabin and temperature and terrestrial curation; 2) conditions similar to ANGSA lunar sample storage and curation; 3) cold storage capabilities that have been developed and tested on the ISS; and 4) ultracold conditions associated with large PSR environments.

Return of Samples in Sealed Containers but Not Temperature Controlled.

Presumably, Artemis III mission to the south polar regions of the Moon will sample some portions of cold traps and place samples in sealed containers such as updated versions of SESC, GASC, or CSVC. Samples will be collected in the lunar vacuum and processed and curated at NASA in an N₂ glovebox under room temperature conditions.

There are profound implications for science and curation for the collection of cold trap samples under this scenario. Warming of the samples will lead to a loss of information concerning horizontal and lateral ice stratigraphy in the cores as well as the overall morphology of ice within the regolith. Melting, boiling, and sublimation reactions (Fig. 3) in the sealed container will allow for only average isotopic compositions of cold trap volatiles to be measured. One case in point is the isotopic composition of H as numerous H-bearing species are predicted to exist in these deposits (3) with distinct stable isotope compositions. Mixing of these components at crew cabin conditions will homogenize any small-scale variations in sources illustrated in Fig. 4. Small-scale volatile deposits previously identified on mineral and glass surfaces (e.g., 108, 109) will be partially to totally lost during the interactions between water vapor/other volatiles and the regolith, although it appears unlikely that whole mineral grains (e.g., agglutinates, volcanic or impact glasses) would exchange with water vapor over the timescale of a mission and subsequent curation. The mineral phases that may have formed by ice–regolith interactions caused by changing conditions over time (radiation, solar wind influx, and illumination and temperature) (39, 48–52) could be altered. Room-temperature collection will result in the loss of valuable information concerning the distribution, crystalline state, and composition of water ice (e.g., amorphous ices, normal hexagonal ice, cubic crystalline ice). The transition from amorphous to crystalline water ice occurs at approximately 135 K, although this is dependent upon ice depositional processes and conditions., Amorphous water ice does not fully “lock in its volatiles above 80 K (34). Further, this sampling and curation approach will lead to difficulty for interpreting human contamination of volatiles during past and current human activities on the lunar surface. Finally, it will be necessary to update sealed container curation approaches beyond capabilities of gas extraction techniques used by the ANGSA program (95).

Within the context of the earliest Artemis missions to the Moon, there are mitigation steps that can be taken to offer partial solutions to the issues above for samples collected and preserved at crew cabin and room temperature. Coldbag technologies (101–103) could be explored for preservation of cold lunar samples for extended periods of time. Coldbags used in the ISS have the capability to keep a payload frozen below −253 K for 130 h. Developing sealed containers with more insulating capabilities with limited contamination problems should be investigated. If passive preservation is not available, there are other steps that could be taken to preserve science. These include a) conducting in situ observation and measurements to gain ground truth insights into original regolith volatile characteristics; b) rather than taking a single core, distinct samples at the surface and subsurface could be taken and placed within smaller individual sealed containers (e.g., GASC, SESC). These steps may preserve stratigraphy of deposits; c) there are volatile-bearing phases that are not affected by thawing (e.g., hematite, lawrencite, akageneite, apatite, sulfides, impact-volcanic glasses, solar wind implanted H). However, some of these phases may be vulnerable to reaction/exchange with terrestrial atmospheres and should be stored in sealed containers within an inert environment. To reduce risk to the astronauts, it is critical to rigorously assess seals on containers to be used for PSR samples under a variety of conditions and provide an outer container.

Return of samples in sealed containers preserved at a temperature of 248 K.

Although samples have not been returned from the Moon or other planetary surfaces at subfreezing temperatures (at 1 atmosphere), lunar, other extraterrestrial, and terrestrial samples have been curated under these conditions (e.g., 95, 105, 106). In addition, ISS activities have demonstrated the use of cold storage and experiments on the ISS and return of cold samples from the ISS that are directly applicable to Artemis’s sampling of cold traps.

Preservation of volatiles would be limited at this temperature. Distribution of water ice within the PSR stratigraphy and within the regolith would be preserved if the ice is in the mm to cm size range and dust covered. The H and O isotopic composition of water ice can be determined and placed within a core stratigraphy. However, even in this optimum occurrence, the water ice could experience substantial sublimation during transport to Earth (Fig. 3). There will be no liquid water, so that alteration processes resulting from potential water–regolith interactions will be eliminated. Still, there will be science that will not be accomplished. For example, the low-temperature structural state of water ice (cubic ice) may not be preserved at temperatures above its presumed stability range. Distribution of other volatiles with melting or boiling points below 248 K will be lost (Fig. 3). Gas composition and isotopic composition will be a mixture of these components. Original organic compounds and clathrates (110) may not be stable (Fig. 3). The basic curation of these samples will be easier with water in a solid state (e.g., 95, 105, 106). The ANGSA initiative designed, assessed, and used a gas manifold and piercing tool to extract gas from a CSVC at room temperature (95) could easily be extended to these lower temperature conditions.

There are solutions for some of the problems identified for sample preservation at 248 K. During the initial stages of Artemis, sampling temperature environments that are approximately 248 K (edges of large PSRs, micro-PSRs, TSRs, or ISRs) will allow unmodified samples of these environments to be returned. In situ measurements will provide ground truth for samples collected in all volatile-rich environments. For all temperatures considered here (248 K, 188 K, 50 K), a strategy must be developed to preserve samples collected at cold trap conditions and then transported to higher temperature terrains during an EVA. Transporting the mass of a refrigerated unit is not practical during nonrover missions expected for Artemis III and IV. For 248 K samples, potentially coldbag technologies could be utilized during EVAs to keep sealed containers cold. For future missions with added transportation infrastructure, a freezer unit could be added to the lunar roving vehicle.

Return of samples in sealed containers preserved at a temperature of 188 K.

Sample collection, preservation, curation, and science at this temperature and lower are limited by current technology. Although lab measurements on water ice have been made at these conditions, and ISS activities have mission-tested freezer capabilities in space, the integration into lunar surface and curation activities has not been attempted on lunar samples. No testing on sealed sample containers exists at these temperatures. Further, Artemis spacesuits will not allow human activities at 188 K or lower.

Water would be well preserved at this temperature for the duration of transport from the Moon to curation. Other expected cold-trap phases would not be preserved. The sublimation rate of H₂O at 150 K (temperature to preserve 100 µm diameter ice grains for 10 d) is 0.8 µg cm^–2 h^–1. For CO₂, the equivalent sublimation rate occurs at 81 K (111). N₂O and Xe require temperatures of 74 and 53 K, respectively. SO₂, NH₃, and H₂S require temperatures of 91 K, 81 K, and 60 K, respectively, to preserve these phases as solids (30).

Return of samples in sealed containers preserved at a temperature of cold traps (20 to 90 K).

Sampling the coldest environments in PSR is the ultimate goal for evaluating the volatile reservoirs on the Moon. These environments retain lunar volatiles over billions of years. The return and analysis of samples at temperatures of the cold traps would potentially fill many of the science goals identified for the Artemis missions. Although sampling is possible with current technologies for the exploration of shadowed regions of micro-PSRs, perimeters of large PSRs, and TSR, the more hostile regions of large PSRs are technologically challenging (Fig. 5). These environments can be explored through human sampling on their perimeters, remote observations from the surface, and perhaps robotic return of samples. Further, advances could be made through the development of infrastructure on the lunar surface during advanced stages of lunar activities.

Astronaut safety issues.

In addition to the impact on science goals, there are implications for astronaut safety and surface activities associated with volatile-rich samples in containers sealed on the Moon. Assuming the composition of volatiles in Cabeus crater detected by the LCROSS mission (3), upon heating to crew cabin conditions or even 248 K in a simple refrigeration container, the PSR regolith sample could release at crew cabin conditions a gas phase consisting of H₂O, CO₂, C₂H₄, CH₄, CH₃OH, NH₃, H₂S, and SO₂ at pressures greater than initially conditions of collection (10^–15 bar) and, while unlikely, could theoretically exceed the cabin-crew pressure. The ultimate pressure is a function of the amount of non-H₂O volatile species (e.g., CO₂, CH₄) in the container. Most likely, the vapor pressures associated with sealed containers will not be high enough to result in container failure for the expected concentrations of volatiles such as CH₄. However, a larger risk is potential leakage of a toxic gas through container seals into the crew cabin. For example, several vacuum-sealed containers used during the Apollo Program did experience long-term leakage across the indium seals due to dust or other impediments. At lower temperatures of storage, there are probably fewer astronaut safety issues with samples, but depending on the storage temperature, some melting and boiling of components within the CSVC or other containers could occur. There are solutions to this potential risk issue, and containers to be used on lunar surface should undergo testing at a variety of conditions. The sealed containers used during EVAs should be placed within outer vacuum containers. Following the Apollo 16 and 17 missions, the sealed CSVC were placed within outer vacuum containers. Analysis of the gas from the outer vacuum container used for the Apollo 17 CSVC exhibited limited leaking during the 50 years it contained samples (87, 89, 90).

There are clearly risks to astronauts during surface activities involved in accessing, exploring, and sampling some of the more extreme conditions associated with large PSRs. This includes both temperature and the slope of the terrane (Figs. 2 and 5) that will exceed capabilities of spacesuits, sampling tools, sample storage, and slope traverse safety. Especially during the initial Artemis missions, sampling of the boundaries of large PSRs, small PSRs, TSRs, and ISRs present less risk as sample targets and better fit the technologies that will be available on the lunar surface.

Conclusions

Volatile-rich regolith associated with PSRs, TSRs, and ISRs provide important sampling targets for future human missions to the Moon (Artemis). The integration of observations from orbit, on the surface, and of samples returned to the lab will answer many fundamental science questions concerning the nature of volatile reservoirs on the Moon and the origin of those reservoirs. Stable isotopes of the various volatile components captured in these cold traps provide a fingerprint for their origin. However, there are numerous mass, technology, and safety issues with retrieving this information. On the one hand, collecting, preserving, curating, and analyzing samples at the extreme conditions of the large PSRs is a great technology challenge. On the other hand, returning the samples to ambient conditions is a challenge to fundamental science advocated for exploring these environments. There are potential fixes that enhance the information retrieved from these samples. Using technologies already documented in ISS activities may preserve significant science if the sample selection sites are strategically identified. Simply keeping samples only slightly below the freezing point of H₂O (at one atmosphere) would reduce the potential information (scientific value) from the returned Artemis samples.

Data, Materials, and Software Availability

There are no data underlying this work.