Understanding the Early Solar System through the Analysis of Meteorites: The Process of Maximizing Data while Minimizing Sample Destruction
Ellen J. Crapster-Pregont1,2
1.Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY, 10964, USA.
2.Dept. of Earth and Planetary Science, American Museum of Natural History, New York, NY, 10024, USA.
Every year approximately 40,000 meteorites make it to Earth’s surface. This value is based on camera network meteor data (Halliday et al., 1989; Bland, 2005) and weathering studies of hot desert meteorites (Bland et al., 1996a,b; Bland, 2005) for stones ranging from 10 to 106 g in mass. Of these, less than 10% are greater than 1 kg and less than 1% are collected, classified, and named (Fig. 1; based on values in Bland (2005) and the Meteoritical Society Bulletin Database). This small percentage is affected by our inability to retrieve many samples, such as ocean falls, and by surface survival rates. Figure 1 breaks down this small percentage a step further to highlight how valuable chondrites, or meteorites from undifferentiated parent bodies, are considering the information they hold about the highest temperature chemistry and processes in the protoplanetary disk as it started cooling and condensing, transitioning from gas and dust to crystalline solids, particularly the refractory (i.e. formed at high temperature) components observed in chondrites such as calcium- and aluminum-rich inclusions. These components preserve the most primitive information about our early solar system, and, as Figure 1 implies, only a small percent of a small percent of collected and classified meteorites contain this valuable information.
One of the greatest goals of a planetary scientist is to piece together chemical environments and physical processes that operated in our early solar system producing the planets and bodies that exist today. This is a Herculean task as much of the evidence lacks context or reflects a more recent, alteration or deformation history. Chondrites have bulk chemistries similar to that of the sun and preserve the history of the protoplanetary disk (i.e. gas and dust distributed in a disk-like fashion around a proto-sun prior to planet formation) and their parent body within their components. Similar to other valuable meteorite groups, the subsets of chondrites with highest scientific value are only available for research in small quantities. It is, therefore, important to maintain a delicate balance between sample preservation and contributing to the scientific knowledgebase from collection and curation through analysis. A combination of non-destructive techniques through the entire sequence from sample selection, preparation, and analysis maximizes scientific return while minimizing material loss.
So what exactly can chondrites tell us? To answer this, we first have to consider the variety of components that make up chondrites (Fig. 2). These components are categorized broadly as: calcium- and aluminum-rich inclusions (CAI), amoeboid olivine aggregates (AOA), chondrules, metal and sulfide nodules, and matrix. CAIs are highly refractory containing the greatest abundance of elements (i.e. Ca, Al, Ti) that condense from a vapor at high temperature, and minerals (e.g. corundum, hibonite, spinel, melilite) formed at high temperatures in the protoplanetary disk. CAIs exhibit a range of textures from primitive aggregates of tiny mineral grains to completely melted and recrystallized. AOAs typically have a core of highly refractory, CAI-like material but are then surrounded by olivine, a Mg-Si mineral that condenses at lower temperatures than the minerals in CAIs. As their name indicates, their texture is fragmented and clustered, with many of the olivine-rimmed refractory clumps arranged together at differing scales. Chondrules are diverse in composition but contain mainly Mg-Si minerals with varying Fe included to a varying degree. Many chondrules appear to have been completely melted prior to accretion while some may reflect a less melted, more agglomerated formation. Metal, generally Fe-Ni alloy, and sulfides can be common or rare depending on the conditions of chondrite formation. Matrix is the fine-grained material that holds all the chondrite components together. The percentage varies between types of chondrites and it is the most susceptible component, after metal and sulfides, to the effects of terrestrial alteration.
Different components and aspects of each component yield diverse information pertaining to characteristics of the early solar system and processes on parent bodies. Primitive, unmelted, CAI material holds the highest temperature record within its chemistry taking us back to a high-temperature protoplanetary disk (e.g. Ebel and Grossman, 2000; Grossman, 2010). Isotopic compositions can reveal the ages of components and define chemical reservoirs (e.g. Krot et al., 2005; Connelly et al., 2012; Holst et al., 2013). Composition, crystallization characteristics, and other features define the melting histories of components’ precursors including the temperature, duration, and extent of heating and potential formation processes (e.g. Jones and Rubie, 1991; Ebel et al., 2008; Krot and Bizzarro, 2009; Desch et al., 2010; Asphaug et al., 2011; Hubbard et al., 2012; Sanders and Scott, 2012; Johnson et al., 2015). Metal and iron content of certain minerals reflect how oxidizing or reducing conditions were and whether this condition varied in space or time (e.g. Connolly et al., 2001; Beck et al., 2012; Schrader et al., 2013). Size and abundance distributions of components among different chondrite types distinguish unique from shared histories among chondrite types (e.g. Cuzzi et al., 2001; Hezel and Palme, 2010; Friedrich et al., 2014). These are just a few examples of critical datasets and their potential implications. All in, chondritic components’, chemistry, proportions, textures, etc. provide constraints for the protoplanetary disk that astrophysicists try to model and for the pre-differentiation parent body processes in the early solar system.
Chondrites clearly contain a wealth of information that provides insight into the conditions of the protoplanetary disk and parent bodies even if only small percentages are recovered. A majority of meteorites studied today are collected through organized efforts, such as the Antarctic Search for Meteorites, which focus on sites in hot and cold deserts where meteorites are both preserved longer and can be concentrated. Terrestrial weathering essentially removes value from a sample as it alters much of the chemical and mineralogical information to reflect recent Earth surface rather than early solar system or chondrite parent body conditions. Falls are the most preferred specimens but are rare (Fig. 1), do not encompass all types of meteorites, and still suffer from terrestrial weather effects. To the planetary science research group at the American Museum of Natural History (AMNH) is guided by the thought that every meteorite sample should be handled and curated in a way that extracts as much information as possible about every aspect of the meteorite while preserving the sample for future use (Fig. 3), especially those samples that contain rare, valuable data about the early solar system.
At the AMNH, the analysis protocol for recent work begins with a trip to the in-house CT scanner (GE VtomeX-S x-ray computed tomography scanner) in the Microscopy and Imaging Facility. This instrument utilizes high-powered x-rays to produce data that is reconstructed into a 3-dimensional (3D) density map of the sample. Obtainable resolution, measured as the edge length of each cubic volume element, or ‘voxel’, depends on sample size, or distance from source, and size of focal spot; i.e. the best resolution for a sample 5x5x20 mm is ~4 micron/voxel on the scanner at AMNH (Fig. 3A and B). Resolution limits the types of analyses that can be conducted. Lower resolution allows virtual isolation (segmentation) and quantification of materials with significantly different densities (i.e. metals vs. silicates or chondrules vs. matrix) while higher resolution studies can differentiate different silicate and oxide minerals (e.g. Ebel et al., 2008; Friedrich and Rivers, 2013; Russell and Howard, 2013; Tsuchiyama et al., 2013). The 3D visualization permits analyses done in 2D to be placed into context (i.e. whether the mineral is in the core or rim of the chondrule) which could greatly affect interpretations. Component relationships and abundances can also be directly calculated from the CT data (e.g. Friedrich and Rivers, 2013; Russell and Howard, 2013; Goldman et al., 2014). While this technique can guide sample preparation, 2D analyses of surfaces are still required to address a majority of the component-based protoplanetary disk and parent body processes conundrums.
During sample preparation, cutting is the step that results in the most unrecoverable sample loss. Typical diamond embedded rock-cutting blades lose a >100 micron thick slice of material. Use of a 20, 30 or 50 micron tungsten (W) wire saw (Princeton Instruments) minimizes the thickness of material lost. This effectively minimizes sample loss and maximizes the number of surfaces that can be analyzed within a given piece of meteorite, a method called ‘serial sectioning’ (Fig. 3C; ps1B and ps2A are cut surfaces). This technique permits >100 micron diameter components to be exposed on two mirrored cut surfaces while larger components can be sectioned in more than two adjacent sets of surfaces (e.g. Ebel et al., 2008). The wire saw also produces a smooth surface requiring minimal grinding when the sample is polished for analysis.
Polishing is necessary to reduce surface topography which negatively affects most analysis techniques. Some techniques, such as electron microprobe (EMP) analysis, require a polish finished with 1 or 0.25 micron diamond solutions, while others, such as electron backscatter diffraction (EBSD), require extremely good polishes adding a chemical etching component with the use of colloidal silica. Diamond is a preferred polishing compound because it does not contaminate the sample with aluminum (Al) or silicon (Si) both which are of interest for components in chondrites. Alcohol or mineral spirits are preferred over water for polishing and rinsing because water may cause oxidation, reactions, or dissolution of some minerals. Successfully prepared samples can be coated with a thin layer of carbon to make them conductive, a necessity for use in most electron beam instruments.
The Cameca SX100 EMP at AMNH uses two types of spectrometers: wavelength dispersive and energy dispersive, to generate elemental concentration data for individual points or regions. Pixel-by-pixel element intensity maps (Fig. 3D) can be combined into red-green-blue (RGB) composites allowing differentiation between types of inclusions and minerals in meteorites over large, region maps (>1 micron/pixel) or individual inclusions (1 micron/pixel). Figure 2 illustrates zoom in of different components and figure 3E shows a component of interest outlined in white. Element intensities measured by the EMP are converted to oxide weight percent (wt%) via calibration against standards analyzed with the same instrument settings as the samples. A variety of software, either customized or packaged, can be used to evaluate each inclusion pixel-by-pixel using element intensity maps and combinations of ratios and cation formulas. A phase map is produced with each pixel assigned a false color indicating mineralogy as determined by element intensities (Fig. 3F). Bulk chemistry, mineralogy, modal abundance, texture, and area are quantifiable from either region or individual inclusion maps (Fig. 3G). The choice of analysis software will affect the time required for sample preparation, calibration, data acquisition, and image analysis and this choice is made based on the scale and focus of the study.
Up to this stage of analysis the techniques (CT, wire saw, polishing, EMP) are minimally destructive to the valuable samples. The data are used to evaluate and compare chondrite component characteristics including chemical composition, mineralogy, and textures. So, with minimal sample loss many scientific questions (major element chemical environments, abundances of components in chondrites, the mineralogy of chondrite components etc.) regarding the protoplanetary disk and parent body processes can begin to be addressed. Our group uses this protocol to build databases that provide contextual and quantifiable information about the early solar system that is accessible for reevaluation in the future. This preserves maximum data for the limited primitive, pristine chondritic samples that have been collected and catalogued.
This database serves as a resource for directing further analyses which can be more destructive or cost prohibitive (Fig. 3H). Electron backscattered diffraction (EBSD) provides crystal orientation information which is used to understand crystallization and deformation of mineral grains. Figure 4 depicts preliminary EBSD data that highlights twinning in metal found in a chondrule which will provide formation constraints (e.g. Crapster-Pregont et al., 2015). Secondary ion mass spectrometry (SIMS) is minimally destructive but may require travel and analysis costs as not all institutions maintain these instruments. Inductively coupled mass spectrometry (ICP-MS) requires either dissolving the sample into solution or blasting the point of interest with a laser while both can be done with minimal sample loss, this method may also require extra cost. Both SIMS and ICP-MS yield information about trace element abundances and even isotopic information yielding constraints on reservoirs and ages (e.g. Stracke et al., 2012; McCubbin et al., 2014) for SIMS and ICP-MS respectively). Focused ion beam liftout for transmission electron microscopy (FIB-TEM) can be used for extremely high-resolution chemical and orientation analysis of relationships within and among minerals within a chondrite component (e.g. Stroud et al., 2002; Stroud et al., 2003). EBSD, SIMS, ICP-MS, and FIB-TEM are just a few techniques implemented by planetary scientists to obtain detailed data from chondrites to continue addressing questions about the early solar system. However, unlike the protocol described above, each of these techniques requires sample consumption to produce data. While valuable sample is lost, the initial context and basic information from the chondrite is preserved in datasets from the minimally destructive analysis techniques.
Even though fewer meteorite samples exist in a catalogued database than are predicted to fall in a year, it is possible to optimize the analysis process with respect to the value of the chondrite and the information it contains. When combined these techniques (Fig. 3) reduce the amount of material lost and maximize the information obtained from a single meteorite sample. The larger set of data preserves contextual and quantifiable data for each CAI, AOA, chondrule, metal nodule, and matrix while guiding future, destructive analysis. By using a series of instruments, visualizations, and software protocols it is possible to begin to better understand the complexity of the protoplanetary disk, and planet formation processes preserved in meteorites with maximum conservation of these precious samples.
Acknowledgments: The Brian Mason Travel Award is sponsored by the International Meteorite Collectors Association for the 2015 78th Annual Meteoritical Society Meeting. Research is supported by the National Science Foundation Graduate Research Fellowship Program grant DGE-11-44155 and NASA Cosmochemistry grant NNX10AI42G.
Fig. 1: Percentage bar representations demonstrate the rarity of and necessity to fully analyze chondrites and their components. Percentages for predicted annual impacts of meteorites (Bland, 2005) are far greater than those collected, classified, and catalogued (top bar: 2014 data from the Meteoritical Society Bulletin Database). When all meteorites in the Database are considered the remaining percentage bar comparison show: iron, achondrite, or chondrite; whether a find or the much less common observed fall; exhibiting parent body alteration or pristine; and whether the component is composed of minerals predicted to condense at highest temperature (refractory) in the protoplanetary disk. All percentages are based on number not mass.
Fig. 2: Backscattered electron image (BSE) of Moss (CO3.6) AMNH #5185 with examples of different components boxed with corresponding outset false color, 3-element red-green-blue composite images. The 3-element combination Mg-Ca-Al clearly distinguishes calcium- and aluminum-rich inclusions (CAI; primarily blue and green), chondrules (primarily red), and amoeboid olivine aggregates (AOA; blue and green core with red surrounding) from each other. While metal appears black in the Mg-Ca-Al images the combination of Fe-Ni-S permits chemical variation observation for the metal nodules and metal in the AOA. Matrix is a darker red color highlighted in the white box within the corresponding image of a different component.
Fig. 3: Preparation and analysis protocol for minimizing sample loss and maximizing data. (A) Photo of Moss (CO3.6) AMNH #5185; (B) single CT slice, high density is whitest; (C) post-wire saw sections; (D) EMP element intensity maps for aluminum (Al), calcium (Ca) and magnesium (Mg) with inclusion outlined; (E) RGB composite, note ease of distinguishing inclusion; (F) false color mineral map output: purple-spinel, red-olivine (olv), green-clinopyroxene (cpx); (G) quantitative data produced; (H) further destructive techniques possible using a high level of prior contextual knowledge (A-G).
Fig. 4: Electron backscatter diffraction (EBSD) generated false color, reverse pole figures maps for a whole metal nodule (a; 2 μm/pixel) and higher resolution portion of a different nodule (b; 0.5 μm/pixel) in the second metal layer in the Acfer 139 layered chondrule. Color represents the orientation of the metal at each pixel described by the mixing chart in the center of the figure where each apex is a different crystal axis. Small wire-frame cubes highlight the orientation of various regions. Lamellar-like features are twinning not artifacts of the polishing process. Image unmodified from (Crapster-Pregont et al., 2015) with permission.
Asphaug E., Jutzi M. and Movshovitz N. (2011) Earth and Planetary Science Letters. Earth and Planetary Science Letters 308, 369–379.
Beck P., De Andrade V., Orthous-Daunay F. R., Veronesi G., Cotte M., Quirico E. and Schmitt B. (2012) The redox state of iron in the matrix of CI, CM and metamorphosed CM chondrites by XANES spectroscopy. Geochimica et Cosmochimica Acta 99, 305–316.
Bland P. A. (2005) The impact rate on Earth. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 363, 2793–2810.
Bland P. A., Berry F. J., Smith T. B., Skinner S. J. and Pillinger C. T. (1996a) The flux of meteorites to the Earth and weathering in hot desert ordinary chondrite finds. Geochimica et Cosmochimica Acta 60, 2053–2059.
Bland P. A., Smith T. B., Jull A. J. T., Berry F. J., Bevan A., Cloudt S. and Pillinger C. T. (1996b) The flux of meteorites to the Earth over the last 50 000 years. Monthly Notices of the Royal Astronomical Society 283, 551–565.
Connelly J. N., Bizzarro M., Krot A. N. and Nordlund Å. (2012) The absolute chronology and thermal processing of solids in the solar protoplanetary disk. Science 338, 651–655.
Connolly H. C. Jr, Huss G. R. and Wasserburg G. J. (2001) On the formation of Fe-Ni metal in Renazzo-like carbonaceous chondrites. Geochimica et Cosmochimica Acta 65, 4567–4588.
Crapster-Pregont E. J., Towbin W. H. and Ebel D. S. (2015) Metal-Centric Perspective of a Layered Chondrule in the CR Chondrite Acfer 139: Insights from Electron Backscattered Diffraction. Lunar and Planetary Science Conference Proceedings. Abstract #1561.
Cuzzi J. N., Hogan R. C., Paque J. M. and Dobrovolskis A. R. (2001) Size-selective concentration of chondrules and other small particles in protoplanetary nebula turbulence. ApJ 546, 496–508.
Desch S. J., Morris M. A., Connolly H. C. and Boss A. P. (2010) A Critical Examination of the X-Wind Model for Chondrule and Calcium-Rich, Aluminum-Rich Inclusion Formation and Radionuclide Production. ApJ 725, 692–711.
Ebel D. S. and Grossman L. (2000) Condensation in dust-enriched systems. Geochimica et Cosmochimica Acta 64, 339–366.
Ebel D. S., Weisberg M. K., Hertz J. and Campbell A. J. (2008) Shape, metal abundance, chemistry, and origin of chondrules in the Renazzo (CR) chondrite. Meteoritics and Planetary Science 43, 1725–1740.
Friedrich J. M. and Rivers M. L. (2013) Three-dimensional imaging of ordinary chondrite microporosity at 2.6 micron resolution. Geochimica et Cosmochimica Acta 116, 63–70.
Friedrich J. M., Weisberg M. K., Ebel D. S., Biltz A. E., Corbett B. M., Iotzov I. V., Khan W. S. and Wolman M. D. (2014) Chondrule size and related physical properties: A compilation and evaluation of current data across all meteorite groups. Chemie der Erde – Geochemistry, 1–25.
Goldman R. T., Crapster-Pregont E. J., and Ebel D. S. (2014) Comparison of Chondrule and CAI Size Measured by Electron Microprobe (2D) and Computed Tomography (3D). Lunar and Planetary Science Conference Proceedings. Abstract #2263.
Grossman L. (2010) Vapor-condensed phase processes in the early solar system. Meteoritics and Planetary Science 45, 7–20.
Halliday I., Blackwell A. T. and Griffin A. A. (1989) The flux of meteorites on the Earth’s surface. Meteoritics 24, 173–178.
Hezel D. C. and Palme H. (2010) The chemical relationship between chondrules and matrix and the chondrule matrix complementarity. Earth and Planetary Science Letters 294, 85–93.
Holst J. C., Olsen M. B., Paton C., Nagashima K., Schiller M., Wielandt D., Larsen K. K., Connelly J. N., Jorgensen J. K., Krot A. N., Nordlund Å. and Bizzarro M. (2013) 182Hf–182W age dating of a 26Al-poor inclusion and implications for the origin of short-lived radioisotopes in the early Solar System. In Proceedings of the National Academy of Sciences of the United States of America. pp. 8819–8823.
Hubbard A., McNally C. P. and Mac Low M.-M. (2012) Short Circuits in Thermally Ionized Plasmas: A Mechanism for Intermittent Heating of Protoplanetary Disks. ApJ 761, 58.
Johnson B. C., Minton D. A., Melosh H. J. and Zuber M. T. (2015) Impact jetting as the origin of chondrules. Nature 517, 339–341.
Jones R. H. and Rubie D. C. (1991) Thermal histories of CO3 chondrites: application of olivine diffusion modelling to parent body metamorphism. Earth and Planetary Science Letters 106, 73–86.
Krot A. N. and Bizzarro M. (2009) Chronology of meteorites and the early solar system. Geochimica et Cosmochimica Acta 73, 4919–4921.
Krot A. N., Yurimoto H., Hutcheon I. D. and MacPherson G. J. (2005) Chronology of the early Solar System from chondrule-bearing calcium-aluminium-rich inclusions. Nature 434, 998–1001.
McCubbin F. M., Shearer C. K., Burger P. V., Hauri E. H., Wang J., Elardo S. M. and Papike J. J. (2014) Volatile abundances of coexisting merrillite and apatite in the martian meteorite Shergotty: Implications for merrillite in hydrous magmas. American Mineralogist 99, 1347–1354.
Russell S. S. and Howard L. (2013) The texture of a fine-grained calcium-aluminium-rich inclusion (CAI) in three dimensions and implications for early solar system condensation. Geochimica et Cosmochimica Acta 116, 52–62.
Sanders I. S. and Scott E. R. D. (2012) The origin of chondrules and chondrites: Debris from low-velocity impacts between molten planetesimals? Meteoritics and Planetary Science 47, 2170–2192.
Schrader D. L., Connolly H. C. Jr, Lauretta D. S., Nagashima K., Huss G. R., Davidson J. and Domanik K. J. (2013) The formation and alteration of the Renazzo-like carbonaceous chondrites II: Linking O-isotope composition and oxidation stateof chondrule olivine. Geochimica et Cosmochimica Acta 101, 302–327.
Stracke A., Palme H., Gellissen M., Münker C., Kleine T., Birbaum K., Günther D., Bourdon B. and Zipfel J. (2012) Refractory element fractionation in the Allende meteorite: Implications for solar nebula condensation and thechondritic composition of planetary bodies. Geochimica et Cosmochimica Acta 85, 114–141.
Stroud R. M., Long J. W., Swider-Lyons K. E. and Rolison D. R. (2002) Transmission electron microscopy studies of the nanoscale structure and chemistry of Pt50Ru50 electrocatalysts. Microscopy and Microanalysis 8, 50–57.
Stroud R. M., Nittler L. R., Alexander C. M., Bernatowicz T. J. and Messenger S. R. (2003) Transmission electron microscopy of non-etched presolar silicon carbide. Lunar and Planetary Science Conference XXXIV. Abstract #1755.
Tsuchiyama A., Nakano T., Uesugi K., Uesugi M., Takeuchi A., Suzuki Y., Noguchi R., Matsumoto T., Matsuno J., Nagano T., Imai Y., Nakamura T., Ogami T., Noguchi T., Abe M., Yada T. and Fujimura A. (2013) Analytical dual-energy microtomography: A new method for obtaining three-dimensional mineral phase images and its application to Hayabusa samples. Geochimica et Cosmochimica Acta 116, 5–16.