The Mars 2020 rover mission is currently collecting and caching samples from the Martian surface for future retrieval and return to Earth. The return of those samples would present our first opportunity to analyze Mars samples with detailed knowledge of their geologic context. In addition, these samples could contain definitive information about the existence of past or present life on Mars, and the types of pre-biotic materials that may impact our understanding of life on Mars.
The process of examining these returned samples presents unique challenges for the curation and scientific investigation of these samples. In order to achieve all Mars Sample Return (MSR) and astromaterials sample return science goals, reliable analyses will depend on overcoming challenging organic signal/noise-related issues. This is especially important since sparse organic compounds that have been identified from the regolith of Mars are detected in the parts per billion range (Eigenbrode et al., 2018; Freissinet et al., 2016). For the Mars 2020 cached samples, X-ray computed tomography (XCT) has been identified as the initial procedure to characterize the interior contents because it can account for a variety of materials that may, or may not, contain evidence of past or present martian life without disassembling the sample (Rummel et al., 2002; Harrington et al., 2018; Tait et al, 2023).
X-ray computed tomography is considered a non-destructive technique that enables examination of whole samples without disassembly. Developed in the 1970’s primarily as a diagnostic tool for medicine, it uses very low energy X-rays that won’t alter organic compounds (e.g. human flesh). Yet it found application in the material sciences, with the earliest examination of a planetary material being the Allende meteorite in 1982 (Arnold et al., 1982). In the last two decades, X-ray computed tomography (XCT) of planetary materials has seen a significant increase in application to planetary science research (Flynn et al., 2000; Friedrich et al., 2016; Glavin et al., 2024; Hezel et al., 2013b; Zeigler et al., 2013; Kminek et al., 2014; Beaulieu et al., 2017; iMARS-II Report, 2018; Schoonen et al., 2018; Tait et al., 2022), with more institutional and laboratory facilities and instrumentation now dedicated to support an increasing demand (Hanna and Ketcham, 2017).
X-ray computed tomography for astromaterials uses higher X-ray energies than medical XCT to produce a three-dimensional image assembled (computed) from multiple X-ray scans, or slices (tomography), through an opaque object. XCT images are produced using either “X-ray tube” based laboratory cabinet instruments or using a synchrotron source. In laboratory “X-ray tube” instruments, one of which was used for this study, the X-rays are generated from the bombardment of a target material (e.g. tungsten) by high energy electrons in an evacuated tube. The resulting X-ray radiation that interacts with the object is polychromatic, meaning a wide spectrum of photons that represent a wide range of X-ray energies. This range of energies will interact with the material in various ways based on the atomic makeup of the material. We assume the nature of that interaction will be either X-ray absorption or X-ray scattering (elastic and inelastic scattering), both of which are how we create a CT image of the material. The result of these interactions with the material is called attenuation, or a decrease in the number of x-ray photons transmitted through the material. The degree of attenuation for the various materials is shown as varying shades of gray in the final assembled CT image.
Among the various ways X-rays interact with the materials, X-ray absorption is of concern for the Mars 2020 cache samples. X-ray absorption can drive chemical reactions and has been shown to alter materials, e.g. by breaking chemical bonds in compounds (Moini et al., 2014; Bertrand et al., 2015), changing the natural radiation signature of astromaterials such as thermoluminescence in meteorites (Sears et al., 2016), and can even heat and alter volatile rich samples in an insulating medium (e.g. Stardust samples in aerogel) (Ebel et al., 2009). There is also a low probability of lower energy X-rays re-emitting following X-ray absorption. It is these lower energy X-rays that may interact with and potentially alter organic materials as certain low-energy photons are readily absorbed by organic compounds. These concerns are the basis for this and similar studies that were conducted for the OSIRIS-Rex returned samples (Glavin et al.,2024).
This study was designed to simulate the preliminary XCT examination of a fully enclosed Mars sample cache-tube using X-ray energies and fluxes designed (see section 3.1.3) to adequately penetrate the cache tube and characterize a range of Mars-like materials. These parameters would normally be considered ‘non-destructive’ by virtue of the fact that the X-ray energy exceeds that which carbon bonds would absorb. However, the energies we identified to properly characterize the diversity of materials that comprise the interaction volume may drive generation of secondary X-rays that can interact with the organic species. There is a certain irony present in that high-energy photons are harmless to organics but low-energy photons can drive chemical damage, but fundamentally the mechanism here is that the low-energy photons are absorbed by organics and the absorbed energy can drive chemical changes.
The experimental samples were designed to best approximate not only the materials and cache containers from Mars 2020, but also to account for Mars-relevant organics for both life detection and possible terrestrial contaminants. The possibility exists that terrestrial contaminants could be altered by X-ray radiation to produce compounds that may produce “false positive” results that mimic potential martian life. Should alteration occur in organic compounds that are terrestrial contaminants, we would have knowledge of known by-products that would not be considered evidence of life. We were able to acquire analogue Mars 2020 cache tube material of the same composition, wall thickness, diameter and length. Analogue Mars regolith, developed by the University of Central Florida (Cannon et al., 2019) uses pure and well-characterized terrestrial materials and rocks that approximate the Rocknest-B regolith composition measured by MSL Curiosity rover, and serve to host Mars-relevant organic compounds for our experiments.
When crafting these experiments we were faced with the dilemma of which compounds among so many that we would examine. The selection of Mars 2020 relevant organic compounds leverages the extensive work that was done by the Organic Contamination Panel (OCP) (Summons et al., 2014) prior to Mars 2020 launch. While the primary charge for the OCP was to evaluate and recommend contamination requirements for Mars 2020, they also provided an assessment of approaches that would be employed in recognizing and interpreting a native martian organic signal while reliably differentiating that from terrestrial contamination. Based on the “unknowable nature of what constitutes a martian organic signature” (Summons et al., 2014), the OCP produced two lists of high-priority compounds to search for in martian materials called the Tier 1 and Tier 2 compounds. These target organics would be both a Mars relevant signal and potential contaminants that could be measurable by current analytical techniques. The OCP’s criteria for assembling these lists included: “…molecules known to be important to terrestrial life; molecules known to be present in carbonaceous meteorites and/or important to prebiotic chemistry; molecules that have already been tentatively detected in martian materials and are thus likely to be measured in returned samples (e.g. chlorobenzene, PAH’s).” The “Tier 1” list of compounds are described as those “…likely to be most important to the science goals of the mission” and in which the mission is recommended to control to within 1 ppb concentration per compound, with the “Tier 2” list controlled to less than 10 ppb (Summons et al., 2014), to prevent obfuscation in post-return laboratory analyses.
We chose eight compounds from the “Tier 1” list as a condensed set of chemicals that includes one representative from each class that are both scientifically important contaminants and potential biosignatures. The compounds are also reasonable approximations of major contributors from carbonaceous chondrite meteorites, either as infall contribution onto Mars or as material detected in asteroidal material by OSIRIS REx and Hayabusa2. Our final group included a sugar (glucose), an amino acid (glycine), a nucleic acid (adenine), a lipid (palmitic acid), a hydrocarbon (heptacosane), a nitrogenous compound (urea), a hydroxy carboxylic acid (pyruvic acid), and a polycyclic aromatic hydrocarbon (PAH, napthalene). All compounds are isotopically labeled (D, 15N, 13C) so that they are easily discernible from contaminants that may be introduced during the experiment and so reaction products produced as a by-product of radiation exposure can be definitively identified.
The XCT parameters, which include the X-ray accelerating voltage (the targeted X-ray energy that accounts for all material compositions being imaged), intensity (x-ray flux or number of X-rays produced in the beam) and exposure duration (how long the sample remains in the X-ray beam) were selected based on the composition of the analogue regolith and cache tube composition. The University of Texas at Austin’s X-ray Computed Tomography facility provided a tool called MuCalc (Mu is for µ which is the estimated attenuation of a material) we could use to calculate the correct accelerating voltage needed to differentiate all the various materials, including the highly attenuating material of the cache tube. Attenuation is based on the highest atomic number present in a material’s composition, which presents a complication since the cache tube itself absorbs most the X-rays thereby reducing the number of X-rays remaining to interact with the regolith analogue. The advantage is that the cache tube actually acts as a filter for much lower energy X-rays generated by the tube-based source, thus eliminating the lower energy original X-rays as a source for organic alteration and leaving only secondary X-rays generated from host materials as the culprits of alteration.
As additional insurance for contamination that might mask X-ray mitigated alteration, samples were assembled as clean as possible at NASA Johnson Space Center (NASA-JSC) in a dry-nitrogen glove box. This included making controls (unexposed samples from which to make comparisons against the exposed samples) and blanks to record any contamination introduced during sample assembly. Completed in one session, the 42 samples were exposed to X-rays at NASA-JSC using the Nikon XTH 320LC XCT instrument and then sent to the Open University for analyses to look for evidence of alteration.
All the isotopically labelled compounds were mixed as powders with the analogue regolith. The compounds were added in equal amounts across all samples, the amounts of which were approximately three orders of magnitude above the maximum expected native Mars signal in order to identify the smallest additions or changes. Similar to the assembly process, the extraction process was completed cleanly and in one session so as to reduce any introduced contamination. The goal was to measure the concentration of each compound between the control and exposed samples to identify any changes or trends with dose through triplicate liquid and gas chromatography-mass spectrometry (LC-MS, GC-MS) analyses.
Of the eight extracted organic compounds measured, three (naphthalene, heptacosane, and glucose) yielded concentrations reasonably close to the expected values, within 2.0±1.1 mg/mL of the original 0.2 g aliquot added to the experimental samples. Urea and glycine had the largest variance at 6.5±2.6 mg/mL and 7.4±1.7 mg/mL respectively. Palmitic acid was divided in half for both GC-MS and LC-MS as nice check of our methods and instrument response, with measured concentrations that varied more than expected, between 0.456±0.2 mg/mL and 0.650±0.2 mg/mL respectively. These concentrations were lower overall between the two techniques and is thought to be due to either a heterogenous distribution of the analyte between the splits or the relative sensitivity difference between the techniques.
Adenine and pyruvic acid measured concentrations were significantly lower in all the samples. During pyruvic acid extraction and processing, I observed a reaction between the solvent and regolith that may have impacted the complete extraction and final measured concentration. This included coloration of the pyruvic acid supernatant liquid and precipitation of small crystals from ultrapure water extracts during dilution. I was able to repeat the process with laboratory materials and analogue regolith back at NASA-JSC to see if we could achieve the same results. The resulting precipitate was analysed by X-ray diffraction at NASA-JSC, revealing a number of salts: 34.8 wt.% Bloedite (Na2Mg(SO4)2*4H2O), 13.9 wt.% Gypsum (CaSO4*2H2O), 11.4 wt.% Nitratine (NaNO3), 1.3 wt.% Siderite(Fe(II)CO3), 0.8 wt.% Rozenite (Fe(II)SO4*4H2O), 1.1 wt.% Magnesite (MgCO3), 0.8 wt.% Talc/rhomboclase (Mg3Si4O10(OH)2/HFe(III)(SO4)2*4H2O), and a balance of 36 wt.% X-ray amorphous materials. This suggests that the analogue regolith may present some complexity associated with complete extraction of some of our analytes. Adenine is notoriously difficult to extract in water, something we had hoped we had solved during method development.
Since Urea and Glycine show greatest signal loss relative to the control, I am looking for radiolytic by-products through non-targeted, non-quantitative comparisons of full MS scans of the blanks, controls and the highest dose samples.
Collectively, the range in post-exposure concentration data show variability that could be caused by sources of error and/or possible changes by ionizing radiation. Deconvolving one from the other is complicated by a combination of measured loss of material due to sample handling, variable extraction efficiencies and some unexpected analyte/solvent interaction with the regolith simulant as indicated in XRD analyses of residues. The results, while not currently conclusive, do provide the appropriate parameters needed to characterize the Mars 2020 cached samples, along with the sample processing methods, designed to be first order processing similar to how organics might be extracted from actual Mars-returned samples, thus providing information about the possible complexity involved with identifying a native Martian organic signal.
Acknowledgment: This research was supported by NASA Johson Space Center and the Open University. Also, many thanks go to Scott Eckley at NASA, and Matt Kershaw and Simona Nicoara at the Open University for their expert support of the XCT, LCMS and GCMS analyses. Lastly, I am grateful to the Meteoritical Society and the Brian Mason Travel Award (sponsored by the International Meteorite Collectors Association) which provided support for my attendance at the 2024 86th Annual Meeting of the Meteoritical Society held in Brussels, Belgium, enabling me to present my PhD research.
References:
Beaulieu, K. R., et al., (2017) Structure from Motion Photogrammetry and Micro X-ray Computed Tomography 3-D Reconstruction Data Fusion for Non-Destructive Conservation Documentation of Lunar Samples, 48th LPSC, LPI #2649
Bertrand, L., et al., (2015) Mitigation strategies for radiation damage in the analysis of ancient materials, Trends in Analytical Chemistry 66: p. 128-145.
Cannon, K.M., et al. (2019) Mars Global Simulant MGS-1: A Rocknest-based open standard for basaltic martian regolith simulants. Icarus 317, p.47-478.
Ebel, D., et al., (2009) Three-dimensional textural and compositional analysis of particle tracks and fragmentation history in aerogel, MAPS 44 (10): p. 1445-1463.
Eigenbrode, J. L., et al., (2018) Organic matter preserved in 3-billion-year-old mudstones at Gale crater, Mars, Science 360: p. 1096–1101.
Flynn, G., et al., (2000) X-Ray Computed Microtomography (Cmt): A non-invasive screening tool for characterization of returned rock cores from mars and other solar system bodies, 31st LPSC, LPI #1893.
Freissinet, C., et al., (2016), Organic molecules in the Sheepbed Mudstone, Gale Crater, Mars, Journal of Geophysical Research: Planets, 120:3 p.495-514.
Friedrich, J.M., et al. (2019) Effect of polychromatic X-ray microtomography imaging on the amino acid content of the Murchison CM chondrite, MAPS, 54, p.220-228.
Glavin, D. P., et al. (2024), Investigating the impact of x-ray computed tomography imaging on soluble organic matter in the Murchison meteorite: Implications for Bennu sample analyses
MAPS, 23pp. https://doi.org/10.1111/maps.14111
Harrington et al., iMARS-II Report 2018
Hezel, D., Friedrich, J.M. and Uesugi, M., eds. (2013) Looking inside: 3D structures of meteorites, Special Issue Geochimica et Cosmochimica Acta 116: p. 1-110.
Kminek, G., et al., (2014) Report of the workshop for life detection in samples from Mars Life Sciences in Space Research 2: p. 1-5.
Moini, M., et al., (2014) Assessing the Impact of Synchrotron X-ray Irradiation on Proteinaceous Specimens at Macro and Molecular Levels, Analytical Chemistry, 2014, 86 (19): p. 9417–9422.
Rummel, J. D., et al., (2002) A draft protocol for detecting possible biohazards in martian samples returned to Earth. NASA/CP-2002-211842.
Schoonen, M., et al., (2018) Examining Returned Samples in their Collection Tubes using Synchrotron Radiation-based Techniques, 2nd International Mars Sample Return, LPI #6014
Sears, D. W. G., Sears, et al., (2016) Report X-ray computed tomography imaging: A not-so-nondestructive technique, MAPS 51.4: p. 833-838.
Summons, R. E., et al., (2014) Planning Considerations Related to the Organic Contamination of Martian Samples and Implications for the Mars 2020 Rover, Astrobiology 14.12: p. 969-1027.
Tait, K.T. et al., (2022) The Mars Sample Return Planning Group 2 (MSPG2) Astrobiology, 22, p57-80.
Zeigler, R. A., et al., (2013) Nondestructive Analysis of Astromaterials by Micro-CT and Micro-XRF Analysis for PET Examination, 76th Annual Meeting of the Meteoritical Society, LPI #5350.
