40Ar-39Ar DATING OF ALH 84001:  EVIDENCE FOR THE EARLY BOMBARDMENT OF MARS.  S. F. Knott, R. D. Ash, and G. Turner, Department of Earth Science, University of Manchester, Manchester M13 9PL, UK.

Published in Lunar and Planetary Science XXVI, LPI, Houston.

In a recent study of the diogenites, Mittlefehldt [1] demonstrated major differences between them and the Antarctic meteorite ALH 84001, principally, the presence in ALH 84001 of relatively sodic plagioclase present as isotropic shock-induced maskelynite, pyrite rather than troilite as the sulphide phase, chromite of distinct composition, and differences in trace-element chemistry. He argued that ALH 84001, formerly classified as a diogenite, is a member of the SNC group of meteorites, which are relatively well established as originating on the planet Mars. The distinctive composition of the oxygen isotopes in ALH 84001 confirmed their link with the SNC group [2]. Unlike the other martian meteorites, which have young crystallization ages, Sm-Nd systematics indicate that ALH 84001 formed early in the planet’s history [3]. In this report we present the first evidence from stepped heating and laser probe 40Ar-39Ar dating that it was subsequently involved in the early bombardment episode that generated the heavily cratered surfaces of the Moon, the southern uplands of Mars, and many of the minor bodies of the solar system.

A 21-mg sample of ALH 84001 was irradiated in the University of Michigan reactor, then step heated for periods of 30 minutes in 100C steps from 100C to 1600C. Variations in the relative proportions of 36Ar, 37Ar, 38Ar, 39Ar, and 40Ar released led us to infer the presence of six distinct and separable sources of these Ar isotopes:  radiogenic 40Ar from the in situ decay of K; cosmogenic 36Ar and 38Ar from spallation reactions induced by cosmic rays; 36Ar, 38Ar, and 40Ar from the terrestrial atmosphere; 37Ar, 38Ar and 39Ar from the reactor neutron irradiation of Ca, Cl, and K, respectively.

Sixty percent of the K-derived 39Ar is released below 700C from a mineral site with a comparatively low 37Ar/39Ar ratio and an inferred Ca/K (weight) ratio between 6 and 9 (Fig. 1). This release is dominated by maskelynite, with a contribution to Ca from carbonate. Above 700C the inferred Ca/K rises dramatically, exceeding 3000 by 1400C. Orthopyroxene, Ca = 1.2 wt%, is the major source of the high-temperature Ca-derived 37Ar. Because of the large amount of Ca-derived 37Ar, the cosmogenic component dominates the high-temperature release of 38Ar. This is reflected in the 38Ar/Ca ratio, which decreases by two orders of magnitude to level off above 1000C, where 80% of the 37Ar is evolved. The minimum value of 38Arcos/Ca in the 1300C and 1400C release is 4.3 × 10-7 ccSTP/gCa. Based on a production rate of 38Ar from Ca of 2.8 × 10-8 ccSTP/g/Ma we calculate a cosmic ray exposure age of 15 Ma, which compares well with previous estimates. 36Ar is a mixture of terrestrial atmosphere and a cosmogenic component. The proportion of cosmogenic 36Ar is calculated based on a production ratio, relative to 37Ar, 0.65 of the 38Ar production ratio. Cl-derived 38Ar released below 1300C was estimated by subtracting both the cosmogenic contribution and an atmospheric contribution (0.188× the atmospheric 36Ar). As a result of its residence in the Antarctic ice sheet ALH 84001 contains a substantial amount of terrestrial atmospheric Ar that is released along with Cl-derived 38Ar in decreasing amounts throughout the heating schedule. The Cl/36Ar ratio clusters in the interval (0.9-1.7) × 106 and compares with a value of around 107 for sea water. The observation of a Cl/36Ar ratio between fresh water and sea water appears qualitatively consistent with an origin of the Cl by wind-borne sea spray.

Following correction for cosmogenic 36Ar, an isochron plot of 40Ar/36Ar against K/36Ar indicates that the 40Ar is a mixture of K-correlated radiogenic Ar and terrestrial atmosphere (40Ar/36Ar = 295.5). The absence of any trend toward the composition of the martian atmosphere (40Ar/36Ar = 2400) argues against a significant contribution from that source. Based on this conclusion, the (atmospheric) 36Ar was used to apply a conventional correction to 40Ar for the presence of 40Ar from the terrestrial atmosphere. The remaining 40Ar was assumed to be from in situ decay of K and the 40Ar/K ratio used to calculate a K-Ar age for each temperature step which is plotted in the usual way in Fig. 1. The maskelynite shows some evidence of 40Ar loss in the low-temperature release but reaches a moderately well-defined plateau, 22%-64% 39Ar release, indicating an age of 4175 ± 25 Ma. The release pattern of the orthopyroxene is more complex possibly as a result of 39Ar recoil artifacts and/or shock degassing of 40Ar. The maximum age in this region of the release is 3.9 Ga. Laser probe and SEM analyses have confirmed the maskelynite as the major K-bearing phase and yielded ages consistent with the stepped heating data. Preliminary examination of a large (200 µm) carbonate grain gave an age of 3.6 Ga, which appears to be significantly younger than the maskelynite. No detectable gas was liberated from pyroxene during lasering.

We interpret the 4.17 Ga plateau age of ALH 84001 as the time of the shock event responsible for the production of the maskelynite and by implication the time of a major impact on the parent body. Given that the parent body may be Mars, this result could represent the first evidence for the timing of the extensive cratering of Noachian-age terrains typical of the southern hemisphere of Mars. Being 400 Ma younger than the formation age of the solar system, the age of ALH 84001 is comparable to the brecciation ages of lunar highland rocks [4] and to those of some meteorite regolith breccias [5]. It therefore provides further evidence of the timing of the widespread early bombardment of the planetary bodies. While the age of ALH 84001 is comparable to the oldest of the brecciation ages of the anorthositic lunar highland samples, it is significantly older than the majority of lunar highland breccias, which cluster in the interval 3.85 to 3.95 Ga. ALH 84001, a (monomict) cataclastic pyroxenite, is considerably less complex than the majority of the (polymict) lunar breccias. This may reflect a deeper plutonic source region, less affected by all but the largest impacts, particularly as the Sm-Nd age appears to demand survival of ALH 84001 from the earliest times. In this context we note some of the few lunar breccias that retain precataclysm ages are plutonic and monomict.

It is clearly not possible to draw more detailed conclusions regarding the bombardment history of Mars on the basis of a single sample. It is nevertheless tempting in making comparisons with the lunar bombardment chronology and the brecciation chronology of meteorites and to pose the question of whether, though broadly similar, the bombardment histories differed in important ways. Specifically, were the widespread late impacts on the Moon at 3.9 Ga part of the solar-system-wide accretionary bombardment or were they local to the Earth-Moon system? If the 3.9-Ga spike was widespread, what was its cause? Unless and until further ancient SNC meteorites are discovered, this and other questions raised may be answered only by directly sampling the martian highlands.

References:  [1] Mittlefehldt D. W. (1994) Meteoritics, 29, 214-221. [2] Clayton R. N. (1993) Antarctic Meteorite Newsletter, 16/(3), 4. [3] Jagoutz A. et al. (1994) Meteoritics, 29, 478-479. [4] Turner G. (1977) Phys. Chem. Earth, 10, 145-195. [5] Turner G. (1988) Meteoritics and the Early Solar System,, 276-288.