SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Abstract– To better determine the effects of impact-related processes on radiometric chronometers in meteorites, we undertook an isotopic study of experimentally shocked and heated samples of lunar basalt 10017. Shock experiments at 55 GPa were completed on one subsample, and a second subsample was heated in an evacuated quartz tube at 1000 °C for 170 h. A third subsample was maintained as a control. Samarium-neodymium, Rb-Sr, 238U-206Pb, and 206Pb-207Pb isotopic analyses were completed on mineral fractions (leached and unleached), leached whole rocks, and complementary acid leachates. Disturbance in the shocked and heated samples was evaluated through comparison of their isochron diagrams with those of the control sample. The Sm-Nd isotope system was the least disturbed, the Rb-Sr isotope system was more disturbed, and the 238U-206Pb and 206Pb-207Pb isotope systems were the most disturbed by shock and annealing. Samples that experienced extended heating demonstrated greater isotopic disturbances than shocked samples. In some cases, the true crystallization age was preserved, and in others, age information was degraded or destroyed. In no case did the experiments generate isochrons that maintained linearity while being rotated or completely reset. Although our results show that neither experimental shock nor thermal metamorphism alone can account for the discordant ages represented by different isotope systems in some Martian meteorites, we postulate that shock metamorphism may render a meteorite more susceptible than its unshocked counterpart to subsequent disturbance during extended impact-related heating or aqueous alteration. The combination of these processes may result in the disparate chronometric information preserved in some meteorites.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

A principal goal of achondrite meteorite studies is to establish the geologic history of the meteorite’s parent body. Meteorites from differentiated bodies such as Mars or the Moon potentially record information on the timing and duration of geologic processes, including planetary differentiation, volcanism, and hydrothermal activity. Isotopic analyses of meteorites are used to establish the chronology of events and constrain the compositional characteristics of the source from which an individual meteorite is derived. The most commonly used method for determining the crystallization age and initial isotopic composition of a meteorite involves measuring parent–daughter radionuclide pairs, such as 147Sm-143Nd or 87Rb-87Sr, in various mineral fractions to produce an internal isochron. A requirement of the mineral isochron technique is that the isotope systematics of the meteorite have remained undisturbed since the time that the minerals crystallized from a melt. Other radiometric dating techniques such as 40Ar-39Ar may record the time of crystallization, but are also important for dating subsequent thermal events related to impacts that may disturb the isotope systematics in a meteorite and thus result in complete resetting of the chronometer.

In the case of some Martian and lunar samples, the isotopic compositions of various mineral and whole-rock fractions either fail to yield isochrons, or yield regression lines in different isotope systems from which calculated ages are discordant, indicating that these samples have experienced some disturbance since the time of crystallization (e.g., Lugmair et al. 1976; Papanastassiou and Wasserburg 1976; Shih et al. 1982; Chen and Wasserburg 1986; Borg et al. 1997). The discordant ages in Martian meteorites in particular have led to fundamentally contrasting interpretations of the timing of Martian magmatism (e.g., Bouvier et al. 2008; Nyquist et al. 2009). Among the shergottites, there are individual samples that yield internally well-defined isochrons with highly discordant ages. Specifically, Rb-Sr and Sm-Nd isotope systems yield concordant ages that have been interpreted to indicate relatively recent crystallization (165–575 Ma; Shih et al. 1982; Borg et al. 2001; Morikawa et al. 2001; Nyquist et al. 2001a; Brandon et al. 2004; Symes et al. 2008; Nyquist et al. 2009), whereas the 206Pb-207Pb isotope system has been interpreted to reflect crystallization at a relatively ancient time (∼4 Ga; Bouvier et al. 2005, 2008). Furthermore, many samples exhibit scatter of individual mineral fractions on internal isochron diagrams, providing clear evidence for postcrystallization disturbances (e.g., Chen and Wasserburg 1986; Borg et al. 1997; Shih et al. 2003). Thus, establishing the absolute chronology of the geologic history of Martian meteorites is hampered by disturbances occurring on the surface of Mars or Earth that modified or destroyed chronologic information in the sample. Deciphering the chronology of the Martian meteorites requires separating the effects of secondary disturbance from the isotopic record of primary crystallization.

There are a number of processes that could potentially disturb isotope chronometers in the Martian meteorites. One process is impact metamorphism. This includes shock with durations of ≤1 s (Beck et al. 2005), accompanied by transient heating, as well as postshock heating of longer duration (days to years or longer). Impact-related shock and heating both affect target rocks in a large impact event. Other possible processes that can disturb chronometers are alteration (i.e., reactions that decompose the minerals to form new ones) and contamination (i.e., addition of extraneous components to the rock). On Mars, in contrast to the Moon, water may play a significant role in disturbing isotope chronometers via alteration or contamination on the parent body. Petrologic and geochemical evidence indicates that alteration and contamination affect many of the Martian meteorites (e.g., Treiman et al. 1993; Crozaz et al. 2003), whereas every meteorite experiences at least one impact event, related to the ejection of the meteorite from Mars. Shock pressures experienced by the Martian meteorites during the ejection event are estimated from 5 to 55 GPa (Nyquist et al. 2001b, and references therein; Fritz et al. 2005). The range and extent of isotopic disturbance that can result from impact metamorphism and the modification that this can cause to isotope chronometers are not well established.

From studies on terrestrial meteorite impact sites, it is known that melting of target rock can “reset” isotope chronometers so that the time of the impact is recorded in this impact-related material (e.g., Sudbury: Davis 2008; Manicouagan: Hodych and Dunning 1992). Confidently dated materials include impact melt rocks, tektites, glass, suevite, and recrystallized minerals (summarized by Deutsch and Schärer 1994; Jourdan et al. 2009). Only rarely does a target rock record the impact age in the absence of complete melting (e.g., Sudbury: Deutsch et al. 1989). However, terrestrial impact deposits are rarely ideal for investigating the specific isotopic effects of different shock-related processes, such as pure thermal metamorphism, or shock accompanied only by transient shock-related heating, because it is difficult to control for variables such as initial lithologic heterogeneity in the target rock, duration of thermal metamorphism, and hydrothermal or other alteration. Furthermore, there are few examples of terrestrial impacts in target lithologies that are compositionally analogous to shergottites (e.g., Lonar crater).

Experimental studies provide an alternative approach for evaluating the isotopic effects of impact metamorphism in which it is possible to control many of the variables that are present in a natural setting. A preliminary experimental study on a lunar basalt found Rb loss as well as a minor degree of Sm-Nd isochron rotation in an experimentally heated sample (Nyquist et al. 1988, 1991). By contrast, no effect was observed in the Rb-Sr isotope systematics of another experimentally shocked sample (Nyquist et al. 1987). To expand this small database of experiments on basaltic materials and characterize the effects of impact metamorphism on Sm-Nd, Rb-Sr, U-Pb, and Pb-Pb chronometry, we have undertaken a study of experimentally shocked and heated basalts. We investigated two impact-related processes: (1) shock accompanied only by transient shock-related heating and (2) pure thermal metamorphism. A lunar basalt was selected for this study because this material is sufficiently old that it contains significant and highly variable amounts of radiogenic Sr, Nd, and Pb in the mineral fractions, and is essentially free of alteration. The particular sample used in this study, Apollo basalt 10017, is petrographically, geochemically, and chronologically well characterized (e.g., Beaty and Albee 1978; Gaffney et al. 2007a). One piece of the sample was used as an unshocked and unheated control for the experiment. A second piece was heated in a manner to simulate the effects of postimpact thermal metamorphism. A third sample of 10017 was shocked at 55 GPa in the course of another experiment (Nyquist et al. 1987), and we used this sample to investigate the effects of impact-related shock. In this article, we first present detailed results from these experiments. We then apply these results to the case of Martian meteorites, and evaluate the possible causes of discordant ages reflected by different chronometers in several Martian meteorites.

Experimental Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Our experiments investigated two endmember effects of an impact: (1) shock at 55 GPa, accompanied only by the transient shock heating; and (2) heating of unshocked material at 1000 °C for 1 week. Target material at the center of an impact may experience temperatures and pressures up to 1000 °C and 60 GPa (shock stage IV; Deutsch and Schärer 1994); at higher TP conditions the target rock may be melted or vaporized. Thus, the shock pressure of our experiment replicates the pressure conditions at the high end of the range expected in a natural impact setting. This pressure is also equal to the maximum pressure experienced by any of the Martian meteorites (Nyquist et al. 2001b, and references therein; Fritz et al. 2005).

Thermal metamorphism of target rock can result from contact with an impact melt sheet within the crater, or from heating by a continuous breccia deposit located within and outside of the crater, to a distance of two crater radii of the crater rim (Deutsch and Schärer 1994). Miller and Wagner (1979) used zircon and apatite fission tracks from basement rocks and crystalline blocks in the impact breccia at the Ries impact crater, Germany, to determine the thermal history. They found the maximum temperature experienced by basement rocks to be approximately 550–800 °C, with a duration of <1 h. The lowest allowable temperature was approximately 200 °C, with a duration of 1000 yr. Rates of diffusion and chemical reaction, and therefore potential isochron disturbance, are strongly dependent on the time as well as the temperature to which target rock is heated in an impact. However, it was not possible to run an experiment on a long (>year) time scale. In the attempt to maximize the possible extent of disturbance produced in our experiment, we conducted the experiment at a higher temperature than would generally be expected to affect the basement rock in a natural impact setting (1000 °C), and ran the experiment for 1 week. Therefore, the results of our heating experiment approach one extreme in the range of possible outcomes of thermal metamorphism resulting from an impact.

Samples of lunar basalt 10017 were shocked at 54.6 GPa, using a 20 mm flat plate accelerator, by F. Hörz at NASA Johnson Space Center (Schaal and Hörz 1977; Nyquist et al. 1987; Hörz, personal communication). In preparation for the shock experiment, the sample was cored using gaseous nitrogen, and the core was then cut into wafers using a diamond blade saw. The wafers were polished with diamond abrasives. The final size of the wafers was 0.5 mm thick × 5 mm diameter. Rubber gloves were used for all sample handling. In the effort to avoid introducing water to the sample prior to the experiment, only ethyl alcohol was used to clean the core and wafers. The experiment was conducted at 5 mbar pressure of terrestrial atmosphere. The shock pulse duration was 0.8 μs, and the shock-induced temperature was estimated at approximately 800 °C. Following the experiment, the samples were pried loose from the tungsten sample holder with a stainless steel pick, with the samples breaking into 5–10 fragments during this process, and stored in a glass vial (in ambient atmosphere) until the time of isotopic analysis. Four sample wafers were combined for the isotopic analysis, and an additional wafer was used for the thin section.

The heating experiment was conducted using an unshocked subsample of 10017 that was flame-sealed in an evacuated (1 × 10−5 torr) quartz tube, and heated to 1000 °C for 170 h. Additional details of this experiment are given in Chaklader et al. (2006). The sample was heated in a vacuum to eliminate the possibility that atmospheric contamination could be added to the sample during heating (e.g., terrestrial Pb), and this temperature was chosen so that the results would illustrate the maximum extent of disturbance that can be attributed to heating, without initiating melting.

Sample Description and Petrography

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Lunar sample 10017,328 is a high-K, high-Ti lunar basalt collected on the southwest part of Mare Tranquilitatis. Adler et al. (1970), Brown et al. (1970), Dence et al. (1970), French et al. (1970), Kushiro and Nakamura (1970), Schmitt et al. (1970), Papike et al. (1976), McGee et al. (1977), and Beaty and Albee (1978) provided the earliest detailed petrographic descriptions of 10017. An extensive collection of petrographic and geochemical data for 10017 is available in the Lunar Sample Compendium (http://curator.jsc.nasa.gov/lunar/compendium.cfm). The basalt is fine grained with a vesicular to vuggy subophitic texture (Fig. 1). Vesicles are circular to irregular in shape, and are devoid of fillings or late-stage reaction products and have smooth inner walls. Vesicles range from 0.2 to 0.5 mm in diameter and constitute <5 vol% of the sample.

image

Figure 1.  Backscattered electron image of mare basalt 10017, from JEOL 8200 electron microprobe at the University of New Mexico. PL = plagioclase; PX = pyroxene; ILM = ilmenite; Mes = mesostasis.

Download figure to PowerPoint

Continuously zoned, subhedral clinopyroxene is the dominant phase, making up 45–60% of the basalt. The earliest-crystallized pyroxene have compositions of Wo40En44Fs16 and Wo10En65Fs25 and are zoned to Ca-poor, Fe-rich rims (Wo11En21Fs68). Pyroxene ranges in size from <50 to 400 μm. Ilmenite is lath-like to anhedral, has an average length of 100 μm and a modal abundance of 14–24%. Pyroxene and ilmenite are poikilitically enclosed by anhedral to lath-shaped plagioclase (An82–73). Plagioclase is 2–3 mm in length and has a modal abundance of 18–35%. The mesostasis is complex and composed of Fe-rich pyroxene, plagioclase, K-feldspar, Fe-metal, ilmenite, silica, K-rich glass, Fe-rich basaltic glass, apatite, and troilite. The mesostasis makes up approximately 8% of the basalt. Based on texture and mineral chemistry, the crystallization sequence is pyroxene[RIGHTWARDS ARROW]ilmenite[RIGHTWARDS ARROW]plagioclase. This crystallization sequence progressed under an oxygen fugacity of approximately iron-wüstite − 1. Basalt 10017 shows no significant petrographic evidence for shock, with pyroxene and plagioclase lacking significant fracturing and ilmenite remaining untwinned (Chaklader et al. 2006). The high ilmenite content of this sample is unique to lunar basalts, and therefore introduces a compositional variable to the experimental work that is not represented in the shergottites. However, its age, pristine nature, and the prior shock experiments on this sample made it otherwise ideal for this study (Schaal and Hörz 1977; Nyquist et al. 1987; Chaklader et al. 2006).

Textural and mineralogical responses to shock over the range of 34.8–78.5 GPa have been described by Schaal and Hörz (1977) and Chaklader et al. (2006). At 54.6 GPa, the subophitic texture is still recognizable (Fig. 2), although it is partially disrupted by remobilized mesostasis (Fig. 2b) and frothy glass (Fig. 2c). At this pressure, plagioclase undergoes conversion to maskelynite, pyroxene is fractured, ilmenite is shock twinned and fragmentary disjointed, and cristobalite is isotropic and diaplectic. The composition of the shocked pyroxene overlaps with that of the pyroxene in the unshocked sample (Chaklader et al. 2006).

image

Figure 2.  a) Backscattered electron image of mare basalt 10017 shocked at 56.4 GPa. PL = plagioclase; PX = pyroxene; ILM = ilmenite. b) Detail of area outlined in large rectangle in (a). Image illustrates remobilized mesostasis that consists of a variety of mineral fragments (including SiO2 rimmed with K-feldspar or plagioclase) and heterogeneous glass. c) Detail of area outlined in small rectangle in (a). Image illustrates heterogeneous, frothy glass.

Download figure to PowerPoint

The heated sample retained most of its subophitic texture (Fig. 3). This is probably due to the temperature of 1000 °C being approximately 100° below the solidus (Ringwood and Essene 1970; O’Hara et al. 1974). Chaklader et al. (2006) concluded that the major element compositions of the pyroxene and plagioclase in the heated sample are similar to those in the pristine sample. However, they observed a redistribution of Li and B among the mesostasis, pyroxene, and plagioclase in the heated sample.

image

Figure 3.  Backscattered electron image of mare basalt 10017 that was heated to 1000 °C. There appears to be limited evidence of mesostasis remobilization. PL = plagioclase; PX = pyroxene; ILM = ilmenite; Mes = mesostasis.

Download figure to PowerPoint

Analytical Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Because of the high percentage of incompatible element-enriched mesostasis present in this sample (∼8%), there is a potential for heterogeneous distribution of incompatible elements between the three subsamples. To minimize this concern, the subsamples were made as large as possible (739 mg for the control sample, 260 mg for the shocked sample, and 871 mg for the heated sample). Subsequently, all three subsamples were processed with the same initial sample preparation and mineral separation procedures, so that the resulting mineral fractions would be directly comparable among the three subsamples (Fig. 4; details following). Upon inspection under binocular microscope (98× magnification), no differences were observed among the magnetic separates made using the same magnetic field for the three different samples, indicating that, at least at this level, the separates were mineralogically comparable.

image

Figure 4.  Mineral separation schemes for a) shocked aliquot and b) heated aliquot. Shaded boxes indicate the fractions analyzed. WR = whole rock; Px = pyroxene; Plag = plagioclase; Ilm = ilmenite- and mesostasis-rich fraction; (R) = leached residue; (L) = leachate; rej = portion of magnetic separate rejected during handpicking.

Download figure to PowerPoint

The specific mineral separation techniques employed were chosen because they replicate those used to date basaltic meteorites (e.g., Gaffney et al. 2007b). Each sample was crushed with a sapphire mortar and pestle, and sieved to separate the 74–150 μm grain size fraction. These fractions were passed through a Frantz isodynamic magnetic separator to isolate plagioclase/maskelynite-rich, pyroxene-rich, and ilmenite-rich mineral separates. For ease of comparison among the three samples, throughout the remainder of the article we use the term “plagioclase” to refer to plagioclase in the heated and control samples, as well as maskelynite in the shocked sample. For the heated and control samples, purified plagioclase and pyroxene mineral separates were generated from the magnetic fractions by hand-picking. Portions of the mineral separates rejected during hand-picking are designated with the suffix “-rej.” The ilmenite-rich magnetic separates were not further purified by hand-picking. The ilmenite-rich fractions also contain a large proportion of mesostasis. There was insufficient material in the shocked sample magnetic separates to further purify with hand-picking. Separate whole-rock fractions were obtained from homogenized powders of each of the three samples.

As part of the analytical procedure, all whole-rock and most mineral fractions were leached in 2 N HCl, to remove terrestrial contamination from the grain surfaces. This leaching step is very important when analyzing meteorites that have undergone a period of residence on Earth’s surface. This procedure is also used to remove rare earth element (REE)-rich phosphate minerals from Martian meteorites and thereby help to generate compositional spread among Sm/Nd ratios of mineral fractions that would otherwise be dominated by REE from phosphates. Therefore, although the leaching procedure is not strictly necessary for Apollo lunar samples, we included this step in the experimental procedures to simulate the treatment that a Martian meteorite would receive in the course of geochemical analysis.

Prior to leaching, mineral and whole-rock fractions were washed in 0.5 N acetic acid, for 10 min in an ultrasonic bath, and rinsed with quartz-distilled water. These wash and rinse solutions were discarded. With the exception of the pyroxene-reject, plagioclase-reject, and ilmenite-rich fractions, all of the mineral and whole-rock fractions were leached in 2 N HCl, for 10 min, in an ultrasonic bath at room temperature. After the leachates were decanted from the solid fractions and reserved for analysis, the fractions were rinsed with quartz-distilled water. The water rinse for each fraction was combined with the corresponding HCl leachate, and these were processed and analyzed along with the residue mineral fractions. Fractions that have been leached are designated with the suffix “(R),” and leachates are designated with the suffix “(L).” In total, there are nine analyzed fractions for each of the heated and control samples, and seven analyzed fractions for the shocked sample.

The mineral, whole-rock, and leachate fractions were spiked with mixed 87Rb-84Sr, 149Sm-150Nd, and 233U-236U-205Pb tracers, and digested with a combination of HF, HCl, and HNO3. Sequential separations of U, Pb, Rb, Sr, Sm, and Nd follow the procedures described by Borg et al. (2005). Isotope ratios were measured using the VG Sector 54 thermal ionization mass spectrometer at the University of New Mexico.

Isotope Systematics of the 10017 Control Sample

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

The Sm-Nd, Rb-Sr, 238U-206Pb, and 206Pb-207Pb data for the unshocked, unheated sample of 10017 are reported and discussed in detail by Gaffney et al. (2007a), and shown in Table 1 and Figs. 5–8. The relevant results are summarized here. The Rb-Sr, Sm-Nd, and 238U-206Pb isotope systems record concordant ages for 10017. The Rb-Sr isochron yields an age of 3.678 ± 0.069 Ga (MSWD = 4.8) and an initial isotopic composition of 87Sr/86Sri = 0.699410 ± 0.000065 (Gaffney et al. 2007a). Previously reported Rb-Sr ages of 10017 include (all corrected to λ87Rb = 0.01402 Ga−1; Minster et al. 1982; Begemann et al. 2001): 3.56 ± 0.05 Ga (Papanastassiou et al. 1970), 3.54 ± 0.22 Ga (Gopalan et al. 1970), and 3.67 ± 0.07 Ga (Nyquist et al. 1987). The Sm-Nd age of 10017 is 3.633 ± 0.057 Ga (MSWD = 0.55), and the initial Nd isotopic composition is εNdi = +3.24 ± 0.39. The 238U-206Pb isotope system preserves an age of 3.616 ± 0.098 Ma (MSWD = 86) with an initial isotopic composition of 206Pb/204Pbi = 31 ± 11. The 235U-207Pb isotope system yields an age of 3.80 ± 0.12 Ga (MSWD = 455). The 206Pb-207Pb isotope system does not yield a coherent age. Gaffney et al. (2007a) attribute the apparent disturbance of 207Pb, evident from the large uncertainty and MSWD of the 235U-207Pb age and the lack of a 206Pb-207Pb age, to contamination of glassy mesostasis in 10017 with ancient, radiogenic Pb derived from the lunar surface, via Pb volatilization and recondensation resulting from impacts. This contamination was not sufficient to disturb significantly the 238U-206Pb isotope systematics. Additional disturbance to the 206Pb-207Pb system by the heating and shock experiments can be at least partially assessed by comparison with the control sample.

Table 1.   Ages calculated from control, shocked, and heated samples of 10017.
 Age (Ga)MSWDInitial isotopic compositionaFractions used in age calculation
  1. aεNd = 104 × [((143Nd/144Ndtsample)/(143Nd/144NdtCHUR)) − 1]. CHUR parameters used in calculation of initial εNd values are: 143Nd/144Nd(0) = 0.512638 and 147Sm/144Nd(0) = 0.1967.

  2. bAges corrected to λ87Rb = 0.01402 Ga−1 (Minster et al. 1982; Begemann et al. 2001).

Control sample
 Sm-Nd3.633 ± 0.0570.55εNd = +3.24 ± 0.39All minerals and WR(R)
 Rb-Sr3.678 ± 0.0694.887Sr/86Sr = 0.699410 ± 65All minerals and WR(R)
 238U-206Pb3.616 ± 0.09886206Pb/204Pb = 31 ± 11All minerals and WR(R)
Shocked sample
 Sm-Nd3.65 ± 0.191.2εNd = +2.77 ± 0.64Plag(R), WR(R), Ilm
 Rb-Sr3.61 ± 0.198.887Sr/86Sr = 0.69945 ± 29All minerals and WR(R)
 Rb-Sr3.61 ± 0.145.287Sr/86Sr = 0.69946 ± 18Recombined fractions + Ilm
 238U-206Pb3.48 ± 0.6324206Pb/204Pb = 26 ± 21Px(R), WR(R), Ilm
Heated sample
 Sm-Nd3.51 ± 0.5253εNd = +1.6 ± 3.7All minerals and WR(R)
 Sm-Nd3.647 ± 0.0530.0077εNd = +2.38 ± 0.37All minerals
 Sm-Nd3.63 ± 0.121εNd = +2.57 ± 0.45Recombined fractions + Ilm
 Rb-Sr3.29 ± 0.417787Sr/86Sr = 0.69958 ± 32All minerals and WR(R)
 Rb-Sr3.59 ± 0.147.287Sr/86Sr = 0.699540 ± 68All minerals
 Rb-Sr3.76 ± 0.492887Sr/86Sr = 0.69953 ± 42Recombined fractions + Ilm
MethodAge (Ga)bReference  
Previously reported ages of 10017
 Rb-Sr3.56 ± 0.05Papanastassiou et al. (1970)  
 Rb-Sr3.54 ± 0.22Gopalan et al. (1970)  
 Rb-Sr3.67 ± 0.07Nyquist et al. (1987)  
image

Figure 5.  Sm-Nd isochron diagrams for 10017. a) Control aliquot (Gaffney et al. 2007a). b) Shocked aliquot. c) Heated aliquot. Ages and initial isotopic compositions are given in Table 1. (See Fig. 4 caption for abbreviations.)

Download figure to PowerPoint

image

Figure 6.  Rb-Sr isochron diagrams for 10017: a) Control aliquot (Gaffney et al. 2007a). b) Shocked aliquot. c) Heated aliquot. Ages and initial isotopic compositions are given in Table 1. (See Fig. 4 caption for abbreviations.) The radiogenic nature of the ilmenite-rich fraction reflects the abundance of incompatible element-enriched mesostasis in this fraction.

Download figure to PowerPoint

image

Figure 7. 238U-206Pb isochron diagrams for 10017. a) Control aliquot (Gaffney et al. 2007a). b) Shocked aliquot. c) Heated aliquot. Inset panel in (a) shows the position of the Ilm and WR(R) fractions relative to the rest of the fractions. Ages and initial isotopic compositions are given in Table 1. (See Fig. 4 caption for abbreviations.) The position of the WR(R) fraction at the radiogenic end of the control isochron reflects the fact the whole rock contains a highly radiogenic component that was not included in the mineral separates, which were obtained by tuning the magnetic field of the Frantz to optimize separation of plagioclase, pyroxene, and ilmenite only. The 235U-207Pb isochron diagram is not plotted because mineral fractions demonstrate the same relative relationships on this diagram as the on the 238U-206Pb isochron diagram.

Download figure to PowerPoint

image

Figure 8. 206U-207Pb isochron diagrams for 10017. a) Control aliquot (Gaffney et al. 2007a). b) Shocked aliquot. c) Heated aliquot. Inset panel in (a) shows the position of the Ilm and WR(R) fractions relative to the rest of the fractions. (See Fig. 4 caption for abbreviations.)

Download figure to PowerPoint

Effects of 55 GPa Shock on the Isotope Systematics of 10017

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Element Mobility in the Shocked Sample

The abundances of Rb, Sr, Sm, Nd, U, and Pb in 2 N HCl whole-rock leachates are not significantly different in the control and shocked samples, suggesting that shock does not cause a substantial change in mobility of these elements (Fig. 9). Rubidium, Sm, and Nd show slightly increased mobility, whereas Sr, U, and Pb show slightly decreased mobility. In all cases, these slight differences in mobility most likely reflect small degrees of heterogeneity among the subsamples.

image

Figure 9.  Relative element mobility during leaching of control, shocked, and heated samples of 10017. Percent of each element in leachate calculated in the basis of the whole-rock fraction, as: 100 × (ng element in WR(L))/(ng element in WR(L) + ng element in WR(R)).

Download figure to PowerPoint

Isotope Systematics of the Shocked Sample

Samarium-Neodymium

All of the analyzed fractions of the shocked sample, with the exception of Plag(L), plot within uncertainty of the Sm-Nd isochron defined by the control sample (Fig. 5b; Table 2). Unfortunately, the Nd isotopic analysis for the shocked Px(R) mineral separate failed. The recombined shocked WR(R + L) fraction lies on the control isochron, but the recombined shocked Plag(R + L + rej) fraction falls slightly to the left of the control isochron. The values for these “recombined” points were calculated from the isotopic compositions of the (R), (L), and rej fractions for each lithology (WR, Plag, etc.), weighted according to the total amount of each element in the fractions, to reconstruct the composition of the bulk fraction prior to hand-picking and leaching. The line regressed through the three mineral and whole-rock fractions (Plag(R), Ilm, and WR(R)) yields an age of 3.65 ± 0.19 Ma (MSWD = 1.2), with an initial εNd of +2.77 ± 0.64 (Table 1; Fig. 10). The large uncertainty associated with this age is in part due to the small spread in 147Sm/144Nd values among the fractions, resulting from the absence of a Px(R) analysis.

Table 2.   Sm-Nd data for 10017.
 wt. (mg)Sm (ppm)Sm (ng)Nd (ppm)Nd (ng)147Sm/144Nda143Nd/144Ndb
  1. Note: Plag = plagioclase; Px = pyroxene; WR = whole rock; Ilm = ilmenite; (L) = HCl leachate; (R) = HCl-leached residue; rej = portion of mineral fraction rejected during hand-picking. All samples and standards run as NdO+. The Px(R) fraction for the shocked sample failed to run on the mass spectrometer. Isochrons are calculated using either 2σp (from standard runs) or 2σm (from measured isotope ratios), whichever is larger. The regression method of Fletcher and Rosman (1982) is used to calculate initial isotopic composition.

  2. aError limits include a minimum uncertainty of 0.1% plus 50% of the blank correction for Sm and Nd added quadratically. Reported values have been corrected for Nd blank of 9 ± 3 pg with 143Nd/144Nd = 0.512200 and Sm blank of 5 ± 3 pg with a natural isotopic composition.

  3. bNormalized to 146Nd/144Nd = 0.7219. Uncertainties are 2σm calculated from the measured isotope ratios. 2σm = [Σ(mi − μ)2/(n(− 1))]1/2 for n ratio measurements mi with mean value μ.

  4. cError limits are 2σp. 2σp = [Σ(Mi − π)2/(N − 1)]1/2 for N measurements Mi with mean value π.

Shocked sample
 Plag(R)3.071.0903.3453.2519.9790.20267 ± 260.512913 ± 10
 Plag(L)  0.7254 2.8500.15388 ± 790.512728 ± 20
 Px(L)  15.99 50.490.19147 ± 190.512672 ± 10
 WR(R)20.146.877138.517.23347.10.24121 ± 240.513860 ± 10
 WR(L)  183.3 577.80.19184 ± 190.512682 ± 10
 Ilm28.0032.72916.197.8827410.20209 ± 200.512930 ± 10
Heated sample
 Plag(R)4.870.61733.0062.40511.710.15519 ± 190.511752 ± 10
 Plag(L)  0.1991 0.71120.1692 ± 320.512602 ± 34
 Plag-rej7.810.90597.0753.00423.460.18228 ± 180.512410 ± 10
 Px(R)3.093.1559.7496.30819.490.30235 ± 300.515305 ± 10
 Px(L)  0.4872 1.6930.1740 ± 140.512475 ± 23
 Px-rej11.322.06323.355.51562.420.22616 ± 230.513468 ± 10
 WR(R)28.427.284207.017.08485.50.25780 ± 260.513890 ± 10
 WR(L)  285.4 938.40.18384 ± 180.512679 ± 10
 Ilm9.4612.80121.135.53336.10.21788 ± 220.513268 ± 10
La Jolla (n = 5)c    10 0.511843 ± 41
image

Figure 10.  Summary diagram of ages determined from Sm-Nd, Rb-Sr, and 238U-206Pb isotope systems, for control (Gaffney et al. 2007a), shocked, and heated samples of 10017. Also shown are previously published Rb-Sr ages for 10017, corrected to λ87Rb = 0.01402 Ga−1 (Minster et al. 1982; Begemann et al. 2001).

Download figure to PowerPoint

Nyquist et al. (1988) present Sm-Nd isotopic results for density and magnetic separates of a sample of basalt 10017 shocked at 35 GPa and a corresponding control sample. However, their data show a much smaller range of 147Sm/144Nd ratios than is evident in our shocked and control samples. This most likely reflects the different mineral separation procedures used in the two studies. The small range of 147Sm/144Nd ratios in the Nyquist et al. (1988) study precludes the calculation of an Sm-Nd age. However, the fractions analyzed by Nyquist et al. (1988) are shifted off the control isochron, reflecting a difference in the Sm/Nd ratios of the shocked fractions of up to approximately 3%. The disturbance is greatest in the most mesostasis-rich fractions, suggesting that the glassy mesostasis is more susceptible to disturbance.

Rubidium-Strontium

The shocked sample does not exhibit substantial disturbance in the Rb-Sr isotope system (Fig. 6b; Table 3). The slight displacement of the WR(L) fraction to the left of the control isochron is similar to that observed in the control sample, and therefore appears to reflect a disturbance that is not related to the experimental procedures. The one notable difference between the shocked and control sample is that the ilmenite-rich fraction in the shocked sample has a much lower Rb/Sr ratio, and a correspondingly lower 87Sr/86Sr. This effect probably results from a smaller proportion of incompatible element-enriched mesostasis in the shocked ilmenite-rich fraction (Gaffney et al. 2007a). The four mineral and whole-rock fractions from the shocked sample yield a regression line corresponding to an age of 3.61 ± 0.19 Ma (MSWD = 8.8), with an initial 87Sr/86Sr of 0.69945 ± 0.00029 (Table 1; Fig. 10). The Ilm + recombined mineral fractions yield similar results: age = 3.60 ± 0.14 Ma (MSWD = 5.2) and initial 87Sr/86Sr = 0.69946 ± 0.00018. These results are consistent with the results of Nyquist et al. (1987) for a sample of 10017 shocked to 35 GPa, who also found minimal disturbance in the Rb-Sr systematics.

Table 3.   Rb-Sr data for 10017.
 wt (mg)Rb (ppm)Rb (ng)Sr (ppm)Sr (ng)87Rb/86Sra87Sr/86Srb
  1. Note: Plag = plagioclase; Px = pyroxene; WR = whole rock; Ilm = ilmenite; (L) = HCl leachate; (R) = HCl-leached residue; rej = portion of mineral fraction rejected during hand-picking. Isochrons are calculated using either 2σp (from standard runs) or 2σm (from measured isotope ratios), whichever is larger.

  2. aError limits include a minimum uncertainty of 1% plus 50% of the blank correction for Rb and Sr added quadratically. Reported values have been corrected for Sr blank of 11 ± 4 pg with 87Sr/86Sr = 0.708200 and Rb blank of 6 ± 4 pg with a natural isotopic composition.

  3. bNormalized to 86Sr/88Sr = 0.1194. Uncertainties are 2σm calculated from the measured isotope ratios. 2σm = [Σ(mi − μ)2/(n(n − 1))]1/2 for n ratio measurements mi with mean value μ.

  4. cError limits are 2σp. 2σp = [Σ(Mi − π)2/(N − 1)]1/2 for N measurements Mi with mean value π.

Shocked sample
 Plag(R)3.070.65882.022418.312840.004557 ± 460.699693 ± 10
 Plag(L)  0.06841 59.290.003339 ± 150.699673 ± 10
 Px(R)8.771.66514.60244.321430.01971 ± 200.700515 ± 10
 Px(L)  0.1973 157.20.003632 ± 550.699777 ± 11
 WR(R)20.145.794116.7155.931400.1075 ± 110.704917 ± 10
 WR(L)  2.539 349.20.02104 ± 210.701004 ± 11
 Ilm28.0011.78329.8183.651400.1857 ± 190.709146 ± 10
Heated sample
 Plag(R)4.870.4287052.088501.224410.002475 ± 250.699624 ± 10
 Plag(L)  0.006371 4.3730.0042 ± 200.700147 ± 11
 Plag-rej7.810.40243.143461.836060.002521 ± 250.699652 ± 10
 Px(R)3.090.45011.391173.2535.20.007519 ± 750.699966 ± 10
 Px(L)  0.039851 2.8430.0406 ± 310.701613 ± 14
 Px-rej11.320.67527.644396.744900.004926 ± 490.699834 ± 10
 WR(R)28.424.484127.4126.035800.1030 ± 100.704192 ± 10
 WR(L)  1.132 688.10.004760 ± 480.705261 ± 10
 Ilm9.464.04038.22150.414230.07772 ± 780.703596 ± 10
NBS 987 (n = 7)c    500 0.710257 ± 30
Uranium-Lead

The shock experiments disturbed the 238U-206Pb systematics of 10017 (Fig. 7b; Table 4). With the exception of the WR(L) fraction, all of the measured fractions fall below the 238U-206Pb control isochron, and taken together, the fractions demonstrate much more scatter than is evident in the control sample (Fig. 7). Correspondingly, “ages” calculated from lines regressed through different subsets of mineral, whole-rock, and leachate fractions from the shocked sample have large uncertainties. The Px(R), WR(R), and Ilm fractions correspond to an “age” of 3.48 ± 0.63 Ga (MSWD = 24). The four mineral and whole-rock fractions together correspond to an “age” of 4.1 ± 2.7 Ga (MSWD = 161), and all seven fractions correspond to an “age” of 3.58 ± 0.92 Ga (MSWD = 1065). The “age” derived from the recombined fractions and ilmenite is 4.0 ± 1.6 Ga (MSWD = 1002). The disturbance apparent in this sample precludes the assignment of any significance to these calculated “ages.” Disturbance is most apparent in the Plag(R) and Plag(L) fractions, which fall farther below the control isochron than any of the other fractions, suggesting that the U-Pb systematics of plagioclase are particularly susceptible to disturbance during shock. The WR(R) and Ilm fractions in the shocked sample have much lower U/Pb ratios, and correspondingly less radiogenic 206Pb/204Pb, than in the control sample. This probably reflects heterogeneous distribution of a trace phase with high U/Pb (mesostasis) among the three initial whole-rock subsamples of 10017. The lower bulk U/Pb of the 10017 subsample used for the shock experiment results in a smaller overall spread in the U/Pb and 206Pb/204Pb values of the shocked mineral fractions. However, the scatter in the data is attributed to shock-related disturbance.

Table 4.   U-Pb data for 10017.
 wt (mg)U (ppm)U (ng)Pb (ppm)Pb (ng)238U/204Pba206Pb/204Pbb207Pb/204Pbb208Pb/204Pbb
  1. Note: Plag = plagioclase; Px = pyroxene; WR = whole rock; Ilm = ilmenite; (L) = HCl leachate; (R) = HCl-leached residue; rej = portion of mineral fraction rejected during hand-picking. Isochrons are calculated using either 2σp (from standard runs) or 2σm (from measured isotope ratios), whichever is larger.

  2. aError limits include a minimum uncertainty of 0.1% plus 50% of the blank correction for U and Pb added quadratically. Reported values have been corrected for Pb blank of 12 ± 6 pg with 208Pb/204Pb = 38.400, 207Pb/204Pb = 15.310, and 206Pb/204Pb = 16.890, and U blank of 4 ± 3 pg with a natural isotopic composition.

  3. bUncertainties are 2σm calculated from the measured isotope ratios. 2σm = [Σ(mi − μ)2/(n(n − 1))]1/2 for n ratio measurements mi with mean value μ.

  4. cError limits are 2σp. 2σp = [Σ(Mi − π)2/(N − 1)]1/2 for N measurements Mi with mean value π. The greater uncertainty on the Faraday–Daly standard runs reflects the added uncertainty from the Faraday–Daly gain calibration.

Shocked sample
 Plag(R) 3.070.08755 0.26880.16240.498673.4 ± 3.852.43 ± 0.9239.39 ± 0.6764.96 ± 0.79
 Plag(L)   0.06121 0.374213.4 ± 3.226.277 ± 0.09720.692 ± 0.07947.10 ± 0.15
 Px(R) 8.770.1690 1.4820.38943.41587.43 ± 0.5088.77 ± 0.2651.50 ± 0.1592.19 ± 0.27
 Px(L)   0.3162 2.23217.19 ± 0.1540.763 ± 0.07028.502 ± 0.05570.13 ± 0.15
 WR(R)20.140.689013.881.26225.42164.78 ± 0.44143.86 ± 0.2569.13 ± 0.13135.68 ± 0.29
 WR(L)   2.922 10.3069.33 ± 0.2684.98 ± 0.1541.609 ± 0.082155.29 ± 0.36
 Ilm28.001.61845.293.629101.6103.57 ± 0.1999.301 ± 0.09049.47 ± 0.061119.30 ± 0.19
Heated sample
 Plag(R) 4.870.01432 0.069720.10870.529630.6 ± 3.090.7 ± 2.770.2 ± 2.1107.2 ± 2.6
 Plag(L)   0.01335 0.50482.9 ± 4.840.53 ± 0.2325.95 ± 0.1361.52 ± 0.30
 Plag-rej 7.810.01533 0.11970.22811.78219.3 ± 1.4119.0 ± 1.473.42 ± 0.83139.0 ± 1.5
 Px(R) 3.090.03670 0.11340.14350.443635.6 ± 3.051.6 ± 1.333.83 ± 0.7874.8 ± 1.7
 Px(L)   0.01037 0.84011.45 ± 0.6243.56 ± 0.1527.18 ± 0.1364.24 ± 0.24
 Px-rej11.320.05266 0.59620.45265.12430.63 ± 0.25109.94 ± 0.3862.19 ± 0.22131.76 ± 0.49
 WR(R)28.420.4714613.400.449012.76437.2 ± 4.4184.36 ± 0.4692.34 ± 0.25204.60 ± 0.62
 WR(L)   3.727 27.4139.04 ± 0.24122.70 ± 0.1463.226 ± 0.092145.65 ± 0.25
 Ilm 9.460.4793 4.5341.22111.55171.66 ± 0.69192.71 ± 0.4896.05 ± 0.26216.79 ± 0.64
All Faraday runs: NBS 981 (n = 10)c2 16.902 ± 0.03415.440 ± 0.03836.54 ± 0.11
Faraday–Daly runs: NBS 981 (n = 17) 0.5–2 16.889 ± 0.08115.436 ± 0.09136.53 ± 0.24
Lead-Lead

Unlike in the U-Pb system, there are no apparent systematic effects of disturbance in the 206Pb-207Pb system (Fig. 8b). The relative positions of the shocked fractions mimic the positions in the control sample. However, the overall spread of the shocked data defines a more narrow range than in the control sample. This can most likely be attributed to heterogeneity of the bulk sample, as suggested by the U-Pb systematics. As in the control sample, the shocked WR(R) fraction is more radiogenic than the Ilm fraction. However, in contrast to the control sample, the recombined shocked WR fraction is more radiogenic than the Ilm fraction. This is an additional indication of heterogeneous distribution of radiogenic Pb-rich mesostasis among the three 10017 subsamples.

Effects of Heating on the Isotope Systematics of 10017

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Element Mobility in the Heated Sample

During leaching of the heated sample, Rb, Sr, Sm, Nd, and U show extents of mobility that are similar to those in the control sample (Fig. 9). By contrast, the mobility of Pb during leaching is substantially increased in the heated sample. Approximately 30% of the Pb is removed from the control whole rock by leaching, whereas nearly 70% of the Pb is removed from the heated whole rock by leaching. The positions of the WR(L) and WR(R) fractions above and below, respectively, the 238U-206Pb control isochron are consistent with loss of radiogenic Pb from the WR(R) fraction, and incorporation of excess radiogenic Pb in the leachate. This Pb mobility is probably driven by the volatility of this element. This greatly increased mobilization of Pb in the heated aliquot corresponds to loss of age information in the 238U-206Pb isotope system, as well as increased scatter of the mineral fractions in the 206Pb-207Pb system. This demonstrates that the standard practice of leaching meteorites prior to isotopic analysis may be detrimental to the preservation of U-Pb isotopic systematics in a thermally metamorphosed sample.

Isotope Systematics of the Heated Sample

Samarium-Neodymium

In the Sm-Nd system for the heated sample, the leachate and WR(R) fractions are most clearly disturbed (Fig. 5c; Table 2). The three leachate fractions fall above the control isochron, indicating either Sm-Nd fractionation during leaching or preferential incorporation of radiogenic Nd, mobilized from crystal lattice sites damaged by radioactive decay, in the leachates. However, neither of these possible processes appears to have had a significant effect on the position of the corresponding Plag(R) or Px(R) fractions relative to the control isochron, because the Plag(L) and Px(L) fractions contain modest amounts of Sm and Nd relative to the corresponding mineral fractions. By contrast, the WR(R) fraction falls below the control isochron. However, the recombined WR(R) + WR(L) fraction falls directly on the control isochron, indicating that the whole rock remained a closed system through the heating event. All of the mineral fractions fall on the control isochron, and yield a concordant age of 3.647 ± 0.053 Ga (MSWD = 0.0077) and an initial εNd of +2.38 ± 0.37. The calculated age including the WR(R) fraction has a much larger uncertainty: 3.52 ± 0.52 Ga (MSWD = 53). The recombined fractions (R + L + rej) all fall on the control isochron, and preserve a concordant crystallization age, 3.63 ± 0.12 Ga (MSWD = 1.0), and an initial εNd of +2.57 ± 0.45 (Table 1; Fig. 10). Despite the apparent disturbance of the WR(R) fraction, the Sm-Nd system appears to be extremely resilient to disturbance by extensive heating.

Rubidium-Strontium

The Rb-Sr systematics of the whole-rock fraction in the heated sample are disturbed (Fig. 6c; Table 3). The WR(L) fraction falls far above the control isochron, and the WR(R) fraction falls a small distance below the control isochron. However, 87Rb/86Sr of the heated WR(L) and WR(R) fractions are similar to the control WR(L) and WR(R), respectively, indicating that the offsets of the heated WR(L) and WR(R) fractions from the control isochron reflect preferential removal of radiogenic Sr by the HCl leachate. Although the WR(R) and WR(L) fractions fall to opposite sides of the control isochron, the recombined WR fraction falls slightly above the control isochron. This suggests that heating resulted in open-system behavior in the Rb-Sr isotope systematics of the whole rock. This open-system behavior must occur during the heating event and subsequent sample processing. The remainder of the heated leachate and residue fractions fall on or very slightly to the left of the control isochron. Although the ilmenite-rich fraction falls on the control isochron, it has lower 87Rb/86Sr and lower 87Sr/86Sr than the ilmenite fraction from either the control or shocked samples. This again suggests that incompatible element-enriched mesostasis is heterogeneously distributed among the three samples used in this study. In this sample, the WR fraction is the most radiogenic fraction analyzed. This indicates that the whole rock contains a radiogenic component that is not represented in any of the mineral fractions, which were obtained by selectively tuning the magnetic field on a Frantz isodynamic magnetic separator to isolate high-purity plagioclase, pyroxene, and ilmenite mineral separates. The best age for the heated sample, 3.59 ± 0.14 Ga (MSWD = 7.2), is derived from the mineral fractions (Table 1). This age corresponds to an initial 87Sr/86Sr of 0.699540 ± 68, which is within uncertainty of the initial derived from the control isochron. The age calculated using the WR(R) and all mineral fractions is 3.29 ± 0.41 Ga (MSWD = 77), reflecting the disturbed WR(R) fraction. The age calculated from the recombined fractions + Ilm likewise yields a large uncertainty: 3.76 ± 0.49 Ga (MSWD = 28). Thus, the Rb-Sr system is more easily disturbed by heating than the Sm-Nd system, but can still yield reliable ages if clean mineral separates are analyzed.

Uranium-Lead

The heated sample shows extensive scatter on the 238U-206Pb isochron diagram (Fig. 7c). The majority of the fractions are well outside of analytical uncertainty of the control isochron. The WR(R) fraction lies far to the right of the control isochron, and the remainder of the fractions lie to the left of the control isochron. The only fractions with compositions close to the control isochron are the recombined WR fraction, as well as the Px(R), Px(L), and Plag(L) fractions. The location of the recombined WR(R + L) fraction on the control isochron points to closed system behavior of this sample during heating. The less radiogenic Pb isotopic composition of the heated sample compared to the control sample is attributed to heterogeneous distribution of incompatible element-enriched mesostasis among the 10017 subsamples. The recombined Plag and Px fractions are nearly identical in their U/Pb and Pb isotopic compositions, indicating that heating effectively homogenized the Pb isotopic and U/Pb compositions of the bulk plagioclase and pyroxene portions of the sample. Thus, hand-picking and leaching are required to generate compositional spread among the (R) and -rej fractions. The recombined plagioclase and pyroxene fractions as well as the ilmenite-rich fraction fall to the left of the control isochron, indicating that these fractions have been affected by other disturbance, in addition to homogenization. Taken together, these observations indicate that while the bulk plagioclase and pyroxene fractions were affected during heating by either loss of U or gain of radiogenic Pb, this disturbance is localized in the residual mineral fractions. The acid-soluble portion of these fractions remained undisturbed, as illustrated by the observation that the leachates remain on the control isochron.

Lead-Lead

It is challenging to interpret experimental disturbances in the 206Pb-207Pb system because the control sample itself is disturbed by impact processes on the Moon. Nevertheless, disturbance of the heated sample is identified by the changes in the relative positions of the fractions in comparison to the control sample. Disturbance mirrors many of the characteristics of the 238U-206Pb system in the heated sample. Notably, the Plag(L) and Px(L) fractions have nearly identical, unradiogenic Pb isotopic compositions. In comparison, these two leachates in the control sample have compositions that, although at the unradiogenic end of the array, are nonetheless different by approximately 50%. Thus, heating appears to have homogenized the leachable Pb component in the bulk plagioclase and pyroxene fractions. In the control sample, the Plag(R), Plag-rej, and recombined Plag fractions are less radiogenic than the Px(R), Px-rej, and recombined Px fractions, respectively. By contrast, the heated sample has Plag(R) and Plag-rej fractions that are more radiogenic than the Px(R) and Px-rej fractions, respectively, and the recombined Plag and Px fractions have nearly identical Pb isotopic compositions. These changes in the plagioclase and pyroxene fractions are notable effects of the heating procedure and demonstrate that this level of thermal metamorphism can disturb a 206Pb-207Pb isochron in a basalt. The Px(R) fraction is the least radiogenic of all the heated mineral or rock fractions. This indicates that leached pyroxene, rather than leached plagioclase, is the most reliable recorder of initial (unradiogenic) Pb in a thermally metamorphosed sample.

General Observations on Heating-Related Disturbance

The heated whole-rock sample shows similar characteristics of disturbance in all three isotope systems. In each case, the leached whole-rock fraction lies below the control isochron and the whole-rock leachate lies above. Thus, thermal metamorphism appears to result in preferential diffusion of the radiogenic daughter isotope to cracks or grain boundaries from where it is preferentially removed by leaching. In the mineral separates, however, loss of the radiogenic daughter to the leachate in proportions large enough to affect the composition of the mineral residues is not consistently observed. Therefore, we infer that the primary source of the leachable radiogenic component is trace phases and glass in the incompatible element-enriched mesostasis contained primarily in the whole-rock fraction. It is evident that it is the combination of leaching with heating that generates this disturbance because, in general, the reconstructed whole-rock fractions ((R) + (L)) lie on or close to the control isochrons. Although it may be possible to minimize this disturbance by reducing the strength of acid used for leaching or eliminating the leaching step altogether, leaching is a necessary analytical step required to remove terrestrial contamination from meteorite samples, especially for those samples that have a long terrestrial residence time or were collected from desert environments.

Our results show that thermal metamorphism of a basalt under the conditions investigated here does not result in a resetting of Sm-Nd, Rb-Sr, U-Pb, or Pb-Pb isotope systems. In the absence of complete melting, resetting an isochron to reflect the time of metamorphism rather than the time of crystallization requires subsolidus diffusional exchange and complete elemental and isotopic reequilibration among the phases from which the isochron is derived. In addition to isotopic changes, this process will also result in other chemical (e.g., compositional zoning) and textural changes in the phases. Cherniak (2001, 2003) has calculated the conditions required for isotopic resetting of Sr and Nd in plagioclase and Pb in pyroxene via diffusion. At 1000 °C, plagioclase (An67) will retain its isotopic composition at the 100 μm scale for 10 yr for Sr and approximately 104 yr for Nd (Cherniak 2003). Under the same temperature conditions, 1 mm pyroxene crystals will retain their Pb isotopic characteristics for approximately 104 yr or longer, depending on the pyroxene composition (Cherniak 2001). Our results are consistent with these calculations. Therefore, in an impact deposit with formation temperatures on the order of 1000 °C or lower, and annealing times <∼104 yr, diffusive exchange and equilibration are unlikely to result in an age representing the time of impact. This is consistent with observations from terrestrial impact settings where most basement rocks and impact deposits devoid of melt lithologies do not record the impact age (Jourdan et al. 2009).

Summary of Experimental Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

There are three primary conclusions from our experimental results. Their implications for meteorite chronology are discussed below.

  • 1
     Neither shock at 55 GPa nor annealing at 1000 °C for 170 h resulted in regression lines through the mineral and whole-rock data that represent a spurious younger (or older) age. In some cases, the true crystallization age of the sample is preserved, although with larger errors, and in other cases age information is degraded or destroyed.
  • 2
     During shock and heating, the Sm-Nd system is the most resistant, the Rb-Sr system is less resistant, and the 238U-206Pb is the least resistant to disturbance. Although the control sample did not yield a 206Pb-207Pb age, the position of the fractions in the heated sample relative to their positions in the control sample suggests that the 206Pb-207Pb system is highly susceptible to disturbance.
  • 3
     Mild HCl leaching enhances disturbance in both shocked and heated samples. This appears to reflect nonproportional removal of parent and daughter nuclides by the leaching process.

Significance of Results for Martian Meteorite Chronology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Martian rocks record a complex range of chronometric information that has been interpreted as a record of numerous geologic events, including crystallization, impact events, aqueous alteration, and ejection as a meteorite from Mars (e.g., Shih et al. 1982; Chen and Wasserburg 1986; Jones 1986; Borg et al. 1999). Whereas some Martian meteorites yield concordant ages with multiple chronometers (e.g., Northwest Africa 1460: Rb-Sr, Sm-Nd, and Ar-Ar crystallization age is 346 ± 17 Ma; Nyquist et al. 2009), Rb-Sr and Sm-Nd isotope systems in other meteorites yield different ages (e.g., Shergotty: Jagoutz and Wänke 1986). A number of shergottites yield young (<∼350 Ma), concordant Rb-Sr and Sm-Nd (and, for some samples, Lu-Hf and U-Pb) ages, and ancient (>4 Ga) Pb-Pb ages (Queen Alexandra Range [QUE] 94201: Borg et al. 1997; Gaffney et al. 2007b; Zagami: Borg et al. 2005; Bouvier et al. 2005). Despite these complicated relationships among multiple chronometers, a general consensus has emerged that the Rb-Sr and Sm-Nd arrays in the shergottites represent young (<∼575 Ma) crystallization ages for these samples (summarized by Borg and Drake 2005). However, linear arrays in 206Pb-207Pb isotope space have recently been interpreted as representing an ancient age of crystallization (∼4 Ga) for some shergottites (Bouvier et al. 2005, 2008). This interpretation requires a mechanism of disturbance that yields young and concordant, but erroneous, ages in the Rb-Sr, Sm-Nd, Lu-Hf, 238U-206Pb isotope systems, while preserving an ancient 206Pb-207Pb crystallization age. The results of our experiments offer insights to possible disturbance mechanisms.

All of the shergottites show evidence for shock, with peak shock pressures ranging from 5 to 55 GPa (Nyquist et al. 2001b; Fritz et al. 2005). Evidence for shock includes plagioclase converted to maskelynite, pyroxene with strong mosaicism, mechanical twinning and undulatory extinction, and olivine with strong mosaicism, undulatory extinction, and planar deformation features (e.g., Nyquist et al. 2001b; Fritz et al. 2005). Also present in many shergottites are veins and pockets of melt glass that may contain vugs or vesicles (Fritz et al. 2005). These shock effects are similar to those observed in the shocked sample of 10017 (Fig. 2). Although Martian meteorites experienced shock with a duration of up to 10 ms (Beck et al. 2005), compared to 0.8 μs in our experiment, we do not observe petrographic differences that can be attributed solely to the difference in duration of shock.

The most significant result from this study is that in no case does simple thermal or shock metamorphism generate a linear array in any isotope system that corresponds to any age other than the crystallization age recorded in the control sample. In other words, shock or heating did not result in reset or rotated isochrons. Instead, shock and heating either had no effect, or resulted in the degradation or destruction of age information. It is therefore highly unlikely that the Rb-Sr, Sm-Nd, or 206Pb-207Pb ages of Martian meteorites suggested by linear arrays of isotopic data are the result of simple shock or thermal metamorphism.

Although our results strongly suggest that neither pure shock nor pure thermal metamorphism can generate the chronometric disturbances observed in these meteorites, other processes related to impacts may be capable of producing these effects. One possibility is that extended heating of a shocked sample with damaged crystal lattice structures may cause isotopic disturbance that is not evident in a sample that has experienced only shock (with transient shock-related heating) or only extended heating. A second possibility is that a shocked sample may be more susceptible to chemical disturbance during subsequent aqueous alteration on Mars or Earth. Impacts may initiate hydrothermal activity that results in aqueous alteration of target rocks. Alternatively or in addition, meteorites may be altered during exposure to the terrestrial environment. Although our experiments do not address these possibilities directly, other experiments on the properties of shocked materials indicate that these ideas warrant further exploration (Ostertag and Stöffler 1982; Boslough and Cygan 1988; Cygan et al. 1989; Boslough 1991). Shock produces crystal defects and fractures within silicate material, which then serve as sites for chemical reaction either as a diffusion interface during heating or through interaction with an aqueous fluid (Boslough 1991). Aqueous dissolution experiments show that shocked silicate minerals have a higher rate of reaction in an aqueous fluid than the unshocked counterpart (Boslough and Cygan 1988; Cygan et al. 1989). Shocked feldspar is also much more reactive than unshocked feldspar, as shown by experimental results in which Na in shocked feldspar exchanged completely with K derived from a surrounding melt (Ostertag and Stöffler 1982). As Rb is similar to K in charge and ionic radius, it is expected that Rb would likewise be susceptible to disturbance in shocked feldspar.

Our new experimental results, together with previous experimental shock studies (Ostertag and Stöffler 1982; Boslough and Cygan 1988; Cygan et al. 1989), support the possibility that shocked Martian rocks are more susceptible to chemical modification during postimpact heating or aqueous alteration than unshocked rocks. Furthermore, some Martian meteorites contain minerals that precipitated from fluid, thus indicating that these meteorites experienced aqueous alteration at some point in their history (e.g., QUE: Wentworth and Gooding 1996). Isotopic disturbances of some Martian meteorites may therefore reflect a combination of shock with heating or/and aqueous alteration. Our experiments demonstrate that the Pb isotope systems are the most susceptible to disturbance during heat and shock. The very low concentrations of Pb in the shergottites (0.6–1.3 ppm Pb in Zagami maskelynite: Borg et al. 2005; Bouvier et al. 2005, 0.05 ppm in QUE 94201 maskelynite: Gaffney et al. 2007b) makes this isotope system further susceptible to disturbance. In the shergottites, maskelynite is the primary host for Pb, and the ancient (4.1 Ga) 206Pb-207Pb ages derived from some shergottites are largely based on maskelynite analyses (Bouvier et al. 2005, 2008). The increased propensity for chemical exchange of diaplectic plagioclase glass (maskelynite) in which the lattice bonds have been broken (Ostertag and Stöffler 1982) may render the maskelynite more susceptible to disturbance due to heating or interaction with fluids than other silicate phases in the rock. Chemical exchange between glass and fluid may precede visible devitrification of the glass (Mungall and Martin 1994). Additionally, sequential leaching and dissolution experiments conducted on plagioclase and pyroxene mineral separates of lunar anorthosite 60025 indicate that, in contrast to shocked pyroxene, shocked plagioclase does not retain its initial Pb isotopic composition (Connelly and Borg 2010). Rather, plagioclase in 60025 has taken up highly radiogenic Pb derived from elsewhere on the lunar surface. Collectively, these observations indicate that the combination of shock with heat or interaction with fluids may cause an even greater disturbance to the Pb isotope system than is observed in our experiments. Thus, the apparent age (∼4.1 Ga) represented in maskelynite-based 206Pb-207Pb isochrons may be an artifact of exchange of Martian Pb hosted in the shock-damaged maskelynite with Pb derived from the terrestrial environments during interaction with fluid, rather than a true crystallization age of the shergottites. The apparently ancient ages represented in the Pb-Pb isochrons may therefore result from mixing between Martian Pb, which is characterized by relatively unradiogenic isotopic compositions, with more radiogenic terrestrial Pb (Gaffney et al. 2007b).

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Experimental shock metamorphism of lunar basalt 10017 results in Rb-Sr and Sm-Nd ages that are concordant with control ages, although the uncertainties on the ages are two to three times larger than for the control ages. In the experimentally heated sample, Rb-Sr and Sm-Nd isochrons based on mineral fractions alone are concordant with the control ages, and uncertainties on the ages are comparable to (Sm-Nd) or slightly larger than (Rb-Sr) those for the control ages. However, in the heated sample, Rb-Sr and Sm-Nd ages derived from lines regressed through the mineral and whole-rock data together, though concordant with the control ages, have uncertainties that are >10% of the calculated age. No age information is preserved by the 238U-206Pb system in either the shocked or heated system. These results show that the Sm-Nd system is the most resistant, the Rb-Sr system is less resistant, and the 238U-206Pb and 207Pb-206Pb systems are the least resistant to disturbance during shock and thermal metamorphism. In no case do our experiments produce rotated or reset isochrons that reflect a spurious age. Instead, either the true crystallization age of the sample is preserved, or age information is degraded or destroyed. For all isotope systems in the heated sample, the leached whole-rock fraction falls to the right of the control isochrons. This suggests that one common effect of thermal metamorphism is preferential diffusion of the radiogenic daughter isotope to grain boundaries, from where it is preferentially removed during leaching. Thus, the combination of heating with leaching introduces disturbance to the whole-rock fractions. The apparent gain of radiogenic Pb by the low-μ (238U/204Pb) mineral components during heating indicates that these components may not be reliable indicators of initial Pb isotopic compositions in planetary materials.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References

Acknowledgments— We thank Fred Hörz for providing the shocked sample of 10017. Audrey Bouvier, Alex Deutsch, and an anonymous reviewer provided detailed and constructive reviews. This work was supported by NASA Mars Fundamental Research Program and Cosmochemistry Program. This work utilized the SAO/NASA Astrophysics Data System. A portion of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Publication number: LLNL-JRNL-409386.

Editorial Handling— Dr. Alexander Deutsh

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedures
  5. Sample Description and Petrography
  6. Analytical Procedures
  7. Isotope Systematics of the 10017 Control Sample
  8. Effects of 55 GPa Shock on the Isotope Systematics of 10017
  9. Effects of Heating on the Isotope Systematics of 10017
  10. Summary of Experimental Results
  11. Significance of Results for Martian Meteorite Chronology
  12. Conclusions
  13. Acknowledgments
  14. References
  • Adler I., Walter L. S., Lowman P. D., Glass B. P., French B. M., Philpotts J. A., Heinrich K. J. F., and Goldstein J. I. 1970. Electron microprobe analysis of Apollo 11 lunar samples. Proceedings, Apollo 11 Lunar Science Conference. pp. 8792.
  • Beaty D. W. and Albee A. L. 1978. Comparative petrology and possible genetic relations among the Apollo 11 basalts. Proceedings, 9th Lunar and Planetary Science Conference. pp. 359463.
  • Beck P., Gillet Ph., El Goresey A., and Mostefaoui S. 2005. Timescales of shock processes in chondritic and Martian meteorites. Nature 435:10711074.
  • Begemann F., Ludwig K. R., Lugmair G. W., Min K., Nyquist L. E., Patchett P. J., Renne P. R., Shih C. Y., Villa I. M., and Walker R. J. 2001. Call for an improved set of decay constants for geochronological use. Geochimica et Cosmochimica Acta 65:111121.
  • Borg L. and Drake M. J. 2005. A review of meteorite evidence for the timing of magmatism and of surface or near-surface liquid water on Mars. Journal of Geophysical Research (Planets) 110, doi:10.1029/2005JE002402.
  • Borg L. E., Nyquist L. E., Taylor L. A., Wiesmann H., and Shih C.-Y. 1997. Constraints on Martian differentiation processes from Rb-Sr and Sm-Nd isotopic analyses of the basaltic shergottite QUE 94201. Geochimica et Cosmochimica Acta 61:49154931.
  • Borg L. E., Connelly J. N., Nyquist L. E., Shih C.-Y., Wiesmann H., and Reese Y. 1999. The age of carbonates in Martian meteorite ALH 84001. Science 286:9094.
  • Borg L. E., Nyquist L. E., Reese Y., Wiesmann H., Shih C. Y., Ivanova M., Nazarov M. A., and Taylor L. A. 2001. The age of Dhofar 019 and its relationship to the other Martian meteorites (abstract #1144). 32nd Lunar and Planetary Science Conference. CD-ROM.
  • Borg L. E., Edmunson J. E., and Asmerom Y. 2005. Constraints on the U-Pb isotopic systematics of Mars inferred from a combined U-Pb, Rb-Sr, and Sm-Nd isotopic study of the Martian meteorite Zagami. Geochimica et Cosmochimica Acta 69:58195830.
  • Boslough M. B. 1991. Shock modification and chemistry and planetary geologic processes. Annual Review of Earth and Planetary Sciences 19:101130.
  • Boslough M. B. and Cygan R. T. 1988. Shock-enhanced dissolution of silicate minerals and chemical weathering on planetary surfaces. Proceedings, 18th Lunar and Planetary Science Conference. pp. 443453.
  • Bouvier A., Blichert-Toft J., Vervoort J. D., and Albarède F. 2005. The age of SNC meteorites and the antiquity of the Martian surface. Earth and Planetary Science Letters 240:221233.
  • Bouvier A., Blichert-Toft J., Vervoort J. D., Gillet P., and Albarède F. 2008. The case for old shergottites. Earth and Planetary Science Letters 266:105124.
  • Brandon A. D., Nyquist L. E., Shih C.-Y., and Wiesmann H. 2004. Rb-Sr and Sm-Nd isotope systematics of shergottite NWA 856: Crystallization age and implications for alteration of hot desert SNC meteorites (abstract #1931). 35th Lunar and Planetary Science Conference. CD-ROM.
  • Brown G. M., Emeleus C. H., Holland J. G., and Phillips R. 1970. Mineralogical, chemical, and petrological features of Apollo 11 rocks and their relationship to igneous processes. Proceedings, Apollo 11 Lunar Science Conference. pp. 195219.
  • Chaklader J., Shearer C. K., and Borg L. E. 2006. The behavior of Li and B in lunar mare basalts during crystallization, shock, and thermal metamorphism: Implications for volatile element contents of Martian basalts. American Mineralogist 91:15531564.
  • Chen J. H. and Wasserburg G. J. 1986. Formation ages and evolution of Shergotty and its parent planet from U-Th-Pb systematics. Geochimica et Cosmochimica Acta 50:955968.
  • Cherniak D. J. 2001. Pb diffusion in Cr diopside, augite, and enstatite, and consideration of the dependence of cation diffusion in pyroxene on oxygen fugacity. Chemical Geology 177:381397.
  • Cherniak D. J. 2003. REE diffusion in feldspar. Chemical Geology 193:2441.
  • Connelly J. N. and Borg L. E. 2010. Revisiting the Pb isotopic system in lunar ferroan anorthosite 60025 (abstract #1966). 41st Lunar and Planetary Science Conference. CD-ROM.
  • Crozaz G., Floss C., and Wadhwa M. 2003. Chemical alteration and REE mobilization in meteorites from hot and cold deserts. Geochimica et Cosmochimica Acta 67:47274741.
  • Cygan R. T., Casey W. H., Boslough M. B., Westrich H. R., Carr M. J., and Holdren G. R. Jr. 1989. Dissolution kinetics of experimentally shocked materials. Chemical Geology 78:229244.
  • Davis D. W. 2008. Sub-million-year age resolution of Precambrian igneous events by thermal extraction-thermal ionization mass spectrometer Pb dating of zircon: Application to crystallization of the Sudbury impact melt sheet. Geology 36:383386.
  • Dence M. R., Douglas J. A. V., Plant A. G., and Traill R. J. 1970. Petrology, mineralogy and deformation of Apollo 11 samples. Proceedings, Apollo 11 Lunar Science Conference. pp. 315340.
  • Deutsch A. and Schärer U. 1994. Dating terrestrial impact events. Meteoritics 29:301322.
  • Deutsch A., Lakomy R., and Buhl D. 1989. Strontium- and neodymium-isotopic characteristics of a heterolithic breccia in the basement of the Sudbury impact structure, Canada. Earth and Planetary Science Letters 93:359370.
  • Fletcher I. R. and Rosman K. J. R. 1982. Precise determination of initial Nd from Sm-Nd isochron data. Geochimica et Cosmochimica Acta 46:19831987.
  • French B. M., Walter L. S., and Heinrich K. F. J. 1970. Quantitative mineralogy of an Apollo 11 lunar sample. Proceedings, Apollo 11 Lunar Science Conference. pp. 433444.
  • Fritz J., Artemieva N., and Greshake A. 2005. Ejection of Martian meteorites. Meteoritics & Planetary Science 40:13931411.
  • Gaffney A. M., Borg L. E., and Asmerom Y. 2007a. The origin of geochemical diversity of lunar mantle sources inferred from the combined U-Pb, Rb-Sr and Sm-Nd isotope systematics of mare basalt 10017. Geochimica et Cosmochimica Acta 71:36563671.
  • Gaffney A. M., Borg L. E., and Connelly J. N. 2007b. Uranium-lead systematics of Mars inferred from the basaltic shergottite QUE 94201. Geochimica et Cosmochimica Acta 71:50165031.
  • Gopalan K., Kaushal S., Lee-Hu C., and Wetherill G. W. 1970. Rb-Sr and U, Th-Pb ages of lunar materials. Geochimica et Cosmochimica Acta Supplement 1:11951205.
  • Hodych J. P. and Dunning G. R. 1992. Did the Manicouagan impact trigger end-of-Triassic mass extinction? Geology 20:5154.
  • Jagoutz E. and Wänke H. 1986. Sr and Nd isotopic systematics of Shergotty meteorite. Geochimica et Cosmochimica Acta 50:939953.
  • Jones J. H. 1986. A discussion of isotopic systematics and mineral zoning in the shergottites—Evidence for a 180 m.y. igneous crystallization age. Geochimica et Cosmochimica Acta 50:969977.
  • Jourdan F., Renne P. R., and Reimold W. U. 2009. An appraisal of the ages of terrestrial impact structure. Earth and Planetary Science Letters 286:113.
  • Kushiro I. and Nakamura Y. 1970. Petrology of some lunar crystalline rocks. Proceedings, Apollo 11 Lunar Science Conference. pp. 607626.
  • Lugmair G. W., Marti K., Kurtz J. P., and Scheinin N. B. 1976. History and genesis of lunar troctolite 76535 or: How old is old? Proceedings, 7th Lunar Science Conference. pp. 20092033.
  • McGee P. E., Warner J. L., and Simonds C. H. 1977. Introduction to the Apollo collections. Part I: Lunar igneous rocks. Houston, TX: Curators Office, Johnson Space Center.
  • Miller D. S. and Wagner G. A. 1979. Age and intensity of thermal events by fission track analysis: The Ries impact crater. Earth and Planetary Science Letters 43:351358.
  • Minster J. F., Birck J. L., and Allegre C. J. 1982. Absolute age of formation of chondrites studied by the 87Rb-87Sr method. Nature 300:414419.
  • Morikawa N., Misawa K., Kondorosi G., Premo W. R., Tatsumoto M., and Nakamura N. 2001. Rb-Sr isotopic systematics of lherzolitic shergottite Yamato-793605. Antarctic Meteorite Research 14:4760.
  • Mungall J. E. and Martin R. F. 1994. Severe leaching of trachytic glass without devitrification, Terceira, Azores. Geochimica et Cosmochimica Acta 58:7583.
  • Nyquist L. E., Hörz F., Wiesmann H., Shih C.-Y., and Bansal B. 1987. Isotopic studies of shergottite chronology: I. Effect of shock metamorphism on the Rb-Sr system (abstract). 18th Lunar and Planetary Science Conference. pp. 732733.
  • Nyquist L. E., Bansal B. M., Wiesmann H., Shih C.-Y., and Hörz F. 1988. Isotopic studies of shock metamorphism: II. Sm-Nd (abstract). 19th Lunar and Planetary Science Conference. pp. 875876.
  • Nyquist L. E., Bogard D. D., Garrison D. H., Bansal B. M., Wiesmann H., and Shih C.-Y. 1991. Thermal resetting of radiometric rges. I: Experimental investigation (abstract). 22nd Lunar and Planetary Science Conference. pp. 985986.
  • Nyquist L. E., Reese Y., Wiesmann H., and Shih C.-Y. 2001a. Age of EET 79001B and implications for shergottite origins (abstract #1407). 32nd Lunar and Planetary Science Conference. CD-ROM.
  • Nyquist L. E., Bogard D. D., Shih C.-Y., Greshake A., Stöffler D., and Eugster O. 2001b. Ages and geologic histories of Martian meteorites. Chronology and Evolution of Mars 96:105164.
  • Nyquist L. E., Bogard D. D., Shih C.-Y., Park J., Reese Y. D., and Irving A. J. 2009. Concordant Rb-Sr, Sm-Nd and Ar-Ar ages for Northwest Africa 1460: A 346 Ma old basaltic shergottite related to “lherzolitic” shergottites. Earth and Planetary Science Letters 73:42884309.
  • O’Hara M. J., Biggar G. M., Hill P. G., Jefferies B., and Humphries D. J. 1974. Plagioclase saturation in lunar high-titanium basalt. Earth and Planetary Science Letters 21:253268.
  • Ostertag R. and Stöffler D. 1982. Thermal annealing of experimentally shocked feldspar crystals. Proceedings, 13th Lunar and Planetary Science Conference. pp. 457463.
  • Papanastassiou D. A. and Wasserburg G. J. 1976. Rb-Sr age of troctolite 76535. Proceedings, 7th Lunar Science Conference. pp. 20352054.
  • Papike J. J., Hodges F. N., Bence A. E., Cameron M., and Rhodes J. M. 1976. Mare basalts—Crystal chemistry, mineralogy, and petrology. Reviews of Geophysics and Space Physics 14:475540.
  • Ringwood A. E. and Essene E. 1970. Petrogenesis of Apollo 11 basalts, internal constitution and origin of the moon. Proceedings, Apollo 11 Lunar Science Conference. pp. 769799.
  • Schaal R. B. and Hörz F. 1977. Shock metamorphism of lunar and terrestrial basalts. Proceedings, 8th Lunar Science Conference. pp. 16971729.
  • Schmitt H. H., Lofgren G., Swann G. A., and Simmons G. 1970. The Apollo 11 samples: Introduction. Proceedings, Apollo 11 Lunar Science Conference. pp. 154.
  • Shih C.-Y., Wooden J. L., Bansal B. M., Wiesmann H., Nyquist L. E., Bogard D. D., and McKay G. A. 1982. Chronology and petrogenesis of young achondrites, Shergotty, Zagami, and ALHA77005—Late magmatism on a geologically active planet. Geochimica et Cosmochimica Acta 46:23232344.
  • Shih C.-Y., Nyquist L. E., Weismann H., and Barrat J. A. 2003. Age and petrogenesis of picritic shergottite NWA 1068: Sm-Nd and Rb-Sr isotopic studies (abstract #1439). 34th Lunar and Planetary Science Conference. CD-ROM.
  • Symes S. J. K., Borg L. E., Shearer C. K., and Irving A. J. 2008. The age of the Martian meteorite Northwest Africa 1195 and the differentiation history of the shergottites. Geochimica et Cosmochimica Acta 72:16961710.
  • Treiman A. H., Barrett R. A., and Gooding J. L. 1993. Preterrestrial aqueous alteration of the Lafayette (SNC) meteorite. Meteoritics 28:8697.
  • Wentworth S. J. and Gooding J. L. 1996. Water-based alteration of the Martian meteorite, QUE 94201, by sulfate-dominated solutions (abstract). 27th Lunar and Planetary Science Conference. p. 1421.