3. STUDIES OF METEORITES, AND LUNAR & MARTIAN ROCKS

a. ¹⁸⁷Re-¹⁸⁷Os & ¹⁹⁰Pt-¹⁸⁶Os Studies of Iron Meteorites & Pallasites.

Some iron meteorites (the so called “magmatic” irons) likely are pieces of asteroidal cores. As such their study can provide valuable insights to planetary core formation and crystallization processes. The highly siderophile elements (HSE: “iron-loving”, including Re, Os, Ir, Ru, Pt and Pd) are useful in elucidating crystallization sequence, so these elements have been the focus of much of our work on irons. We recently completed a ¹⁸⁷Re-¹⁸⁷Os and ¹⁹⁰Pt-¹⁸⁶Os isotopic and elemental study of the two largest magmatic iron meteorite groups, IIAB and IIIAB (Cook et al., 2004). That study revealed that the cores these meteorites sample crystallized very early in solar system history (approximately 4.5 billion years ago) and remained closed to the movement of HSE since crystallization. However, complex trace element behavior for Re, Pt and Os in these groups, particularly group IIIAB proved difficult to explain. During the past several years we have extended our study to the magmatic groups IVA and IVB (each of the groups presumably sample the cores of different asteroids). Although somewhat less complex than groups IIAB and IIIAB, the volatile depleted natures of IVA and IVB, together with their very different highly siderophile element abundances make them very interesting systems to study.

We have also examined several ungrouped irons that have previously been proposed to be linked to IVA (EET83230), IVB (Willow Grove, Tishomingo, Chinga), or other (Nedagolla) iron systems, based on certain chemical characteristics. As follows is a brief review of results from studies of these iron systems.

IVA Irons

A ¹⁸⁷Re-¹⁸⁷Os isochron regression for 12 IVA irons yielded an age of 4543±17 Ma and initial ¹⁸⁷Os/¹⁸⁸Os=0.09558±0.00017. This age is consistent with core crystallization relatively early in solar system history, and the goodness of fit implies that open system behavior subsequent to crystallization did not affect Re or Os abundances, or likely other HSE.

Chondrite normalized HSE data for the IVA irons form a continuous series of nested patterns consistent with Re, Os, Ir, Ru and Pt behaving as variably compatible trace elements during metal crystallization. Palladium is the only incompatible element among the IVA irons measured. Plots of log[Os] versus log[other HSE] yield straight lines, thus, retaining the relation: slope=(D_HSE-1)/(D_Os-1), where the D value is the solid metal-liquid metal bulk distribution coefficient for each element. This indicates that although absolute D values likely changed in response to changes in S and P contents within the melt, the relative D values evidently did not change significantly.

CI chondrite normalized abundances of HSE for IVA irons. Note that although EET 83230 is not classified as a IVA iron, its pattern is consistent with late formed solid-liquid mixtures in the crystallization sequence from low to high Ni. b. Log[Os] vs. log[other HSE] plots. Linear trends indicate that relative D values did not change significantly during the crystal-liquid fractionation sequence (from Walker et al., 2005).

A Re-Os model for IVA crystallization using initial DOs and DRe values of 2.5 and 2.3 (appropriate for the low assumed S content of this core) and assuming modest increases in S content and corresponding increases in D values as crystallization proceeded, can account for the Re-Os characteristics of this iron group. For this model, an initial Re concentration of 320 ng/g and chondritic ¹⁸⁷Re/¹⁸⁸Os of 0.41 was also assumed. On the plot of Re vs. ¹⁸⁷Re/¹⁸⁸Os this model can account for most IVA compositions as primary solids, or mixtures of solids and equilibrium liquids. Ungrouped iron EET 83230 (and IVA iron Fuzzy Creek) may be accounted for via mixing between early formed solids and a late-stage liquid. Nedagolla, an iron with some characteristics similar to IVA does not fit this model.

Crystallization model for IVA irons showing hypothetical liquid and solid tracks using internally consistent distribution coefficients for 0 to 80% fractional crystallization. Open squares along the liquid track correspond to 20% fractions of liquid. Open squares refer to mixtures of proportion of liquid and solid. The Re-Os elemental systematics of most IVA irons can easily be modeled as mixtures of equilibrium liquids and solids of the predicted trends. The IVA Fuzzy Creek (and ungrouped EET 83230) can be modeled as mixtures of early solids and evolved liquids.

IVB Irons

A ¹⁸⁷Re-¹⁸⁷Os isochron regression for twelve IVB irons (and eight duplicate analyses) gives an age of 4575±56 Ma and an initial ¹⁸⁷Os/¹⁸⁸Os= 0.09536±0.00036. Precision is limited by the minimal range in ¹⁸⁷Re/¹⁸⁸Os from 0.322 to 0.398. Nonetheless, this age is similar to Re-Os ages of other iron groups and implies early crystallization and elemental closure.

Chondrite normalized HSE patterns are quite different from IVA irons, with low Ni samples having pronounced enrichments in Re, Os and Ir, and depletions in Pd. Our results are generally similar to those of previous studies (e.g. Perncika and Wasson, 1986; Campbell and Humayun, 2005), but of higher precision. Trace element behavior is similar to that of IVA irons, although for this system Pt appears to come as close to having a non-varying D value of 1 as nature likely permits. Over the entire range of IVB evolution, Pt abundances vary by only ±3%! As with the IVA irons, Pd behaves as an incompatible trace element.

CI chondrite normalized abundances of HSE for IVB irons. Tisomingo, which shares some chemical affinities with IVB irons, has a pattern consistent with it having a genetic relationship to IVB. Willow Grove has abundances that are less consistent with a IVB association. Chinga, a meteorite previously classified as an “anomalous” IVB appears to share little in common with IVB. b. Log[Os] vs. log[other HSE] plots. From Honesto et al. (2006).

A Re-Os model for IVB irons using initial DOs and DRe values of 2.0 and 1.8, satisfactorily accounts for the concentrations of all IVB irons analyzed. The ungrouped iron Tishomingo can be accounted for as a mixture between an early formed solid and a late-stage liquid. Chinga and Willow Grove have compositions that are inconsistent with the model presented.

Results of these types of models provide potentially valuable complements to the elemental modeling accomplished by previous studies. The previous studies have largely focused on other elements as key indicators of crystal-liquid fractionation, and indeed, used those elements as the chief discriminators as to whether or not a given iron is related to one group or another. For example, comparisons of Au vs. Ga, Au vs. Ge, and Au vs. Ir correlations have proven very difficult to explain for IVA irons by simple fractional crystallization models using an assumed uniform starting S content and associated D values believed appropriate to the system.

Crystallization model for IVB irons showing hypothetical liquid and solid tracks using internally consistent D values. The Re-Os systematics of most IVB irons can easily be modeled as mixtures of equilibrium liquids and solids of the predicted trends. Ungrouped Tishomingo can be modeled a mixture of an early-formed solid and highly-evolved liquid. Willow Grove and Chinga do not fit this model. Orange diamond is initial melt.

b. Late Accretionary History of the Inner Solar System via Studies of Lunar and Martian Meteorites.

Generally chondritic relative abundances and high absolute abundances of the highly siderophile elements (HSE: Ru, Rh, Pd, Re, Os, Ir, Pt, Au) in Earth’s upper mantle provide strong evidence that these elements were added to the Earth following the last major interaction between its metallic core and silicate fraction (see above). So called “late accretion” (see above) may have added materials comprising as much as 0.8% of the total mass of the Earth and possibly a similar proportion of mass to the Moon. One way we can study the chemical nature of late accreted materials to the inner solar system is by examining the HSE contained in lunar impact-melt rocks, rocks derived from the lunar mantle, and rocks derived from the Martian mantle. The highly siderophile element absolute and relative abundances in these rocks provide clues regarding the chemical nature of late accreted materials, the quantities of materials added to the rocky planets, and perhaps even the timing of these additions.

Lunar Impact Melt Breccias

This work is conducted as part of the Goddard Center for Astrobiology.

The HSE contained in lunar impact melt rocks were largely added to the Moon during the period of time from the origin of the lunar highlands crust (4.4-4.5 Ga) to the end of the late bombardment period (ca. 3.9 Ga). These materials provide the only direct chemical link to the late accretionary period of the Earth-Moon system. The chemical fingerprints of the HSE in late accreted materials may enable us to ascertain under what conditions and where in the solar system the late accreted materials formed. The ¹⁸⁷Os/¹⁸⁸Os ratios (reflecting long-term Re/Os), coupled with ratios of other HSE, can be diagnostic for identifying the nature of the impactor.

Photos of an Apollo 17 lunar impact melt breccia fragment.

A critical issue, however, will be deconvolving the exogenous from indigenous components. Towards this end we have obtained a variety of Apollo 14 and 17 melt breccias, along with a lunar meteorite, NWA482 for analysis. The A17 and A14 melt rocks are believed to have formed ~ 3.9 Ga ago during the generations of the Serenitatis and Imbrium basin forming events (and perhaps the lunar far side for NWA482), respectively, and likely sample the impactors that generated these late formed basins. We have studied multiple sub-pieces of each rock. All rocks studied are analyzed for Os isotopes and highly siderophile element abundances (Pt, Pd, Ir, Ru, Re and Os).

Removal of the effects of indigenous contributions from the HSE of impact-melt rocks is critical to accurately fingerprinting the HSE of the impactors. In the A17 poikilitic rocks and 14321 subsamples (below), Ir shows good linear correlations with all other HSE, consistent with two-component mixing of a single indigenous component and a single meteoritic component. Results for poikilitic rock 72395 show evidence for indigenous Pd and Ru, whereas there is no evidence of any indigenous HSE in the 14321 data.

 
Plot of Ir (in ng/g) versus Re, Os, Ru, Pt and Pd for Apollo 17 and Apollo 14 impact melt rocks. If it is assumed that the lunar target rocks contained very low Ir, the near 0 y-axis intercepts on both plots for Re, Os and Pt suggest that these elements were also present in very low abundance in the lunar target rocks. Non-zero intercepts for Pd and Ru in the A-17 rocks, however, suggest that these two elements were present in significant abundance in the target rocks, and the indigenous abundances must be subtracted from the estimate for the impactor.

When corrected for indigenous components, the putative impactor HSE compositions differ from diagnostic characteristics of the main chondrite groups (below), possibly implicating impactors with different nebular histories from anything currently in our sample collections.

Plots of ¹⁸⁷Os/¹⁸⁸Os vs. Ru/Ir vs. Pd/Ir (corrected for indigenous components), as compared with carbonaceous, enstatite and ordinary chondrites. The presumed impactor compositions for A17 poikilitic rocks and the A-14 rocks do not match any known chondrite group. This could reflect impactors with HSE compositions that are different from our museum samples, or reflect fractionation processes that occurred during the generation of these complex rocks. To resolve some aspects of this question we have begun analysis of “pristine” lunar highlands rocks.

Goddard Center for Astrobiology 2007 summer intern Lorne Loudin (Keene State College), accompanied by Igor Puchtel, analyzes Apollo 17 lunar impact melt rocks using ICP-MS.

HSE Characteristics of the Lunar and Martian Mantles

As compared with surface breccias, the HSE abundances of the lunar mantle must reflect concentrations present in the accreting Moon, or additions to the Moon prior to complete formation of the lunar crust. Low abundances might indicate relatively limited additions shortly after the formation of the Moon, or extraction of most HSE into the diminutive lunar core. Relatively high abundances might reflect an early spike in late accretion. Similarly, HSE abundances of the Martian mantle can potentially reveal whether the late accretionary history of Mars was disparate from that of the Earth-Moon system. Unfortunately, unlike for Earth, we do not have bona fide samples of either the lunar or Martian mantles and must currently be content to work with derivative materials (mafic and ultramafic rocks generated by partial melting of the mantles).

Towards this end, we recently completed a study of the Os isotopic and HSE concentration systematics of lunar picritic orange and green glasses (Walker et al., 2004). We discovered that residues of leached glass spherules contained more radiogenic Os (higher ¹⁸⁷Os/¹⁸⁸Os at present) than the leachates, but lower Os abundances, suggesting the presence of at least two Os (and likely other HSE) components. The presumed radiogenic indigenous component has much lower Os and HSE concentrations than had been previously presumed for lunar glasses. This may be a reflection of the mantle sources of the glasses containing quite low concentrations of these elements.

Upper photo is of the Apollo 17 orange glass sampling site. Lower photo is PPL view of the Apollo 17 orange glass spherules (FOV ~0.5 mm across).

Highly siderophile element abundance patterns (normalized to chondritic abundances x1000) for etchates and residues of 2 size fractions of green glass 15426. The pattern for the etchate is similar to chondritic meteorites and is suggestive of modest micrometeorite contamination. The fractionated residue is likely more representative of the indigenous lunar glass composition.

Mixing calculations suggest that all green glasses analyzed contain some proportion of a chondritic contaminant. Several samples with quite radiogenic ¹⁸⁷Os/¹⁸⁸Os may approach the concentration of the indigenous orange glass component and provide limiting constraints on the concentration of Os in the pure glass.

Relative to terrestrial rocks with comparable MgO contents we concluded that the lunar mantle sources of the orange and green glasses were depleted in the HSE by at least a factor of 20 relative to the terrestrial mantle. This observation indicates that the lunar mantle did not receive a late accretionary component like that required to explain the HSE budget of Earth's mantle. The "missing" HSE could reside in the lunar crust, which is both ancient and thick, and may have protected the lunar mantle from this putative late influx of material (in which case, the late accreted materials must partly reside within the lunar crust, see above). Alternately, the missing HSE could have been extracted into a small lunar core at the time of its formation, or may even continue to reside in the lunar mantle in residual metallic iron species. The latter two hypotheses can be tested, because metal would likely lead to strong fractionation of Re from Os in the silicate mantle. This is because metal has less affinity for Re than Os. Consequently, other materials derived from the lunar mantle, such as basalts, would likely show supra-chondritic Os isotopic compositions at the time of their formation, if metal was responsible for the apparent depletion of HSE in the lunar mantle. For this reason, lunar basalts will be a major target of our future investigations of the Moon.

For reasons similar to those for the Moon, determining the absolute and relative abundances of HSE in the Martian mantle are of great importance. As an example, the abundances of Os we have measured in most Martian meteorites lie within, but at the low end of the range of data defined by terrestrial volcanic rocks (see figure below). This may suggest that the abundances of the HSE in the Martian mantle are similar to, or only a factor of 2-5 lower than in the terrestrial mantle.

These results, therefore, suggest a veneer of approximately the same proportion as was added to Earth. Of note, the extremely low concentrations of Re and Os present in the ca. 4.5 billion year old ALH 84001 may reflect derivation from a mantle reservoir that had not yet accumulated much of its late veneer prior to its very early formation age of the meteorite.

MgO (wt. %) vs. Os (ng/g) for typical terrestrial rocks (gray and black symbols), lunar orange and green glasses (orange and green symbols), and Martian SNC meteorites (red circles). Concentrations of etchates and residues of green glass 15426,164 (>200 µm) and orange glass 74220 (74-150 µm) are shown. The Os concentrations of the indigenous components in the lunar glasses (residues) are considerably less than the concentrations in the bulk glasses.

c. Efficiency of Nebular Mixing.

The extent of isotopic heterogeneity of the solar nebula is still much debated. It is an important topic because the degree of heterogeneity provides important clues regarding material injection processes and nebular mixing rates. There is no question that isotopic heterogeneities for many elements are clearly delineated in components within primitive meteorites. Such heterogeneities reflect the presence of presolar minerals that may have wildly different isotopic compositions compared to the solar system average. The issue of heterogeneity among larger bodies, such as meteorite parent bodies and planets is equally complicated and more contentious.

Our most recent efforts have been focused on the Os isotopic compositions of primitive meteorites, especially with respect to acid resistant components (in some cases including presolar materials)(see Yokoyama et al., 2007). Osmium is one of the most refractory elements. It consists of seven stable isotopes produced by stellar nucleosynthesis via the p-process (¹⁸⁴Os), s-process (¹⁸⁷Os), p- and s-processes (¹⁸⁶Os), and s- and r-processes (¹⁸⁸Os, ¹⁸⁹Os, ¹⁹⁰Os and ¹⁹²Os). In addition to these processes, ¹⁸⁶Os and ¹⁸⁷Os are produced by radioactive decay of ¹⁹⁰Pt (t_1/2 = 488 Gyr) and ¹⁸⁷Re (t_1/2 = 41.5 Gyr), respectively. To explore nucleosynthetic effects in Os we recently, precisely measured Os isotopic ratios in bulk samples of five carbonaceous, two enstatite and two ordinary chondrites, as well as the acid-resistant residues of three carbonaceous chondrites. We found that all bulk meteorite samples have uniform ¹⁸⁶Os/¹⁸⁹Os, ¹⁸⁸Os/¹⁸⁹Os and ¹⁹⁰Os/¹⁸⁹Os ratios, when decomposed by an alkaline fusion total digestion technique. These ratios are also identical to estimates for Os in the bulk silicate Earth. Despite Os isotopic homogeneity at the bulk meteorite scale, acid insoluble residues of three carbonaceous chondrites are enriched in ¹⁸⁶Os, ¹⁸⁸Os and ¹⁹⁰Os, isotopes with major contributions from stellar s-process nucleosynthesis. Conversely, these isotopes are depleted in acid soluble portions of the same meteorites.

The complementary enriched and depleted fractions indicate the presence of at least two types of Os-rich components in these meteorites, one enriched in Os isotopes produced by s-process nucleosynthesis, the other enriched in isotopes produced by the r-process. Presolar silicon carbide is the most probable host for the s-process-enriched Os present in the acid insoluble residues. Because the enriched and depleted components present in these meteorites are combined in proportions resulting in a uniform chondritic/terrestrial composition, it requires that disparate components were thoroughly mixed within the solar nebula at the time of the initiation of planetesimal accretion. This conclusion contrasts with evidence from the isotopic compositions of some other elements (e.g., Sm, Nd, Ru, Mo) that suggests heterogeneous distribution of matter with disparate nucleosynthetic sources within the nebula.

Plots of ε¹⁸⁶Osi versus ε¹⁸⁸Os (A) and ε¹⁹⁰Os versus ε¹⁸⁸Os (B) for bulk samples of carbonaceous chondrites processed by Carius tube digestion, and acid-resistant phases. The hatched area represents the 2σ reproducibility of nine bulk chondrites decomposed by alkaline fusion. Bold lines indicate the regressions results for data from all chondrite samples measured in this study and four bulk Tagish Lake samples of Brandon et al. (2005).

To learn more about our research in meteoritics and planetary science, please refer to:

Morgan J.W., Horan M.F., Walker R.J. and Grossman J.N. (1995) Rhenium-osmium concentration and isotope systematics in group IIAB iron meteorites. Geochim. Cosmochim. Acta 59, 2331-2344.

Smoliar M.I., Walker R.J. and Morgan J.W. (1996) Re-Os ages of group IIA, IIIA, IVA and IVB iron meteorites. Science 271, 1099-1102.

Meisel T., Walker R.J. and Morgan J.W. (1996) The osmium isotopic composition of the Earth’s primitive upper mantle. Nature 383, 517-520.

Walker R.J., Morgan J.W., Beary E., Smoliar M.I., Czamanske G.K. and Horan M.F. (1997) Applications of the ¹⁹⁰Pt-¹⁸⁶Os isotope system to geochemistry and cosmochemistry. Geochim. Cosmochim. Acta 61, 4799-4808.

Shirey S.B. and Walker R.J. (1998) Re-Os isotopes in cosmochemistry and high-temperature geochemistry. Annual Reviews of the Earth and Planetary Sciences 26, 423-500.

Horan M.F., Smoliar M.I. and Walker R.J. (1998) ¹⁸²W and ¹⁸⁷Re-¹⁸⁷Os systematics of iron meteorites: chronology for melting, differentiation and crystallization in asteroids. Geochim. Cosmochim. Acta 62, 545-554.

Brandon A.D., Walker R.J., Morgan J.W. and Goles G.G. (2000) Re-Os isotopic evidence for early differentiation of the Martian mantle. Geochim. Cosmochim Acta. 64, 4083-4095.

Borisov A. and Walker R.J. (2000) Os solubility in silicate melts: new efforts and results. American Mineralogist 85, 912-918.

Righter K., Walker R.J. and Warren P.H. (2000) Significance of highly siderophile elements and Os isotopes in the lunar and terrestrial mantles. in Origin of the Earth and Moon, R. Canup and K. Righter eds., Univ. of Arizona Press, Tucson, 291-322.

Walker R.J. (2000) News & Views: The extraterrestrial wedding ring. Nature 406, 359-360.

Becker H., Walker R.J., MacPherson G.J., Morgan J.W. and Grossman J.N. (2001) Rhenium-osmium systematics of calcium-aluminum-rich inclusions in carbonaceous chondrites. Geochim. Cosmochim. Acta 65, 3379-3390.

Morgan J.W., Walker R.J., Brandon A.D. and Horan M.F. (2001) Siderophile elements in Earth's upper mantle and lunar breccias: Data synthesis suggests manifestations of the same late influx. Meteoritics and Planetary Science 36, 1257-1275.

Walker R.J., Horan M.F., Morgan J.W., Becker H., Grossman J.N. and Rubin A. (2002) Comparative ¹⁸⁷Re-¹⁸⁷Os systematics of chondrites: Implications regarding early solar system processes. Geochim. Cosmochim. Acta 66, 4187-4201.

Horan M.F., Walker R.J., Morgan J.W., Grossman J.N. and Rubin A. (2003) Highly siderophile elements in chondrites. Chem. Geol. 196, 5-20.

Becker H. and Walker R.J. (2003) The ⁹⁸Tc-⁹⁸Ru and ⁹⁹Tc-⁹⁹Ru chronometers: new results on iron meteorites and chondrites. Chem. Geol. 196, 43-56.

Becker H. and Walker R.J. (2003) Efficient mixing of the solar nebula from uniform Mo isotopic composition of meteorites. Nature 425, 152-155.

Lazar G.C., Walker D. and Walker R.J. (2004) Experimental partitioning of Tc, Mo, Ru and Re between solid and liquid during crystallization in Fe-Ni-S. Geochim. Cosmochim. Acta 68, 643-652.

Cook D.L. Walker R.J., Horan M.F., Wasson J.T. and Morgan J.W. (2004) Pt-Re-Os systematics of group IIAB and IIIAB iron meteorites. Geochim. Cosmochim. Acta 68, 1413-1431.

Gelinas A., Kring D.A., Zurcher L., Urrutia-Fucugauchi J., O. Morton and Walker R.J. (2004) Osmium isotope constraints on the proportion of bolide component in Chicxulub impact melt rocks. Meteoritics Planet. Sci .39, 1003-1008.

Walker R. J., Horan M.F., Shearer C.K. and Papike J.J. (2004) Depletion of highly siderophile elements in the lunar mantle: evidence for prolonged late accretion. Earth Planet. Sci. Lett. 224, 399-413.

Brandon A.D., Walker R.J. and Puchtel I.S. (2006) Platinum-Os isotope evolution of the Earth’s mantle: constraints from chondrites and Os-rich alloys. Geochim. Cosmochim. Acta 70, 2093-2103.

Becker H., Horan M.F., Walker R.J., Gao S., Lorand J.-P. and Rudnick R.L. (2006) Highly siderophile element composition of the Earth’s primitive upper mantle: Constraints from new data on peridotite massifs and xenoliths. Geochim. Cosmochim. Acta 70, 4528-4550.

Lee S. R. and Walker R. J. (2006) Re-Os isotope systematics of mantle xenoliths from South Korea: evidence for complex growth and loss of lithospheric mantle beneath East Asia. Chem. Geol. 231, 90-101.

Yokoyama T., Rai V.K., Alexander C.M.O’D., Lewis R.S., Carlson R.W., Shirey S.B., Thiemens M.H. and Walker R.J. (2007) Osmium isotope evidence for uniform distribution of s- and r-process components in the early solar system. Earth and Planet. Sci. Lett. 259, 567-580.

Walker R.J., McDonough W.F., Honesto J., Chabot N.L., McCoy T.J., Ash R.D. and Bellucci J.J. Origin and chemical evolution of group IVB iron meteorites. Geochim. Cosmochim. Acta, in revision.

Puchtel I.S., Walker R.J., James O.B. and Kring D.A. Osmium isotope and highly siderophile element systematics of lunar impact melt rocks: Implications for the late accretion history of the Moon and Earth. Submitted to Geochim. Cosmochim. Acta.

Last Revised July 2007