Geochimicaet CosmochimicaActa, Vol. 61, No. 24, pp. 5417-5428, 1997 Copyright © 1997ElsevierScienceLtd Printed in the USA.All rights reserved 0016-7037/97 $17.00 + .00
Pergamon
PII S0016-7037(97)00315-3
Nitrogen and heavy noble gases in ALH 84001: Signatures of ancient Martian atmosphere S. V. S. MURTY and R. K. MOHAPATRA Physical Research Laboratory, Ahmedabad-380009, India (Received March 21, 1997; accepted in revised.form August 29, 1997) Abstract--Nitrogen and noble gases have been studied in a bulk sample and three density separates of the Martian orthopyroxenite ALH 84001. The 6~5N values which lie between 85%0 and -18%o (and after correcting for cosmogenic contribution, between 46%0 and -23%~), define a two component mixing trend in a plot of 6tSN vs. 1/N, with Chassigny as one endmember and another component with 6 ~SN -> 46%0. This trend is different from the one defined by the data from EET 79001,C and glass from Zagami. Most of the krypton and xenon are of trapped origin; the ratios ~29Xe/J32Xe and 136Xe/~32Xe being similar to the Martian atmospheric values as found in EET 79001 ,C. In addition, small contributions from in situ 238U fission and live J29I decay are evident in some high temperature steps, the later observation attesting to the antiquity of this Martian meteorite. Excesses at 80. 82Kr and ~2SXe due to neutron capture effects on bromine and iodine, respectively, are observed in all the samples. These neutron effects are not consistent with in situ production in the meteoroid during cosmic ray exposure and hence should be produced in the Martian atmosphere or surface and entered the meteorite as a trapped component. The lower 6 ~SN (->46%o) and 4°Ar/36Ar --- 1400 in the trapped component of ALH 84001, as compared to the values from EET 79001,C, together with the fact that radiogenic 4°Ar and trapped 36Ar, 84Kr, and 132Xe have similar release pattern, are strongly suggestive that the trapped component in ALH 84001 represents Martian atmosphere of - 4 G a ago. The noble gas elemental ratios 36Ar/mXe and 84Kr/~32Xe show an elemental fractionation trend, enriching the heavy noble gases, similar to what has been observed in Nakhla (Drake et al., 1994). Comparing the nitrogen and xenon isotopic records and the radiogenic and stable isotope ratios (4°Ar/ 1a9Xe and 36Ar/~4N) from ALH 84001 representing Martian atmospheric component of - 4 Ga ago, with those from EET 79001,C representing Martian atmospheric component of recent past, we infer the following on the evolution of the Martian atmosphere: (a) Xenon isotopic composition, as well as the amounts of xenon have been completely evolved at 4 Ga in Martian atmosphere and almost remained unchanged to the present; (b) The radiogenic 4°Ar has not been completely degassed into the atmosphere at 4 Ga; (c) Nitrogen has been lost in a continuous process, leading to an increase in the ratio of 3~'Ar/ 14N as well as the 6~5N in the present Martian atmosphere as compared to 4 Ga ago. These inferences are consistent with the model predictions (Pepin, 1994). Copyright © 1997 Elsevier Science Ltd 1. INTRODUCTION
(Bogard et al., 1984; Bogard and Johnson, 1983; Becker and Pepin, 1984; Wiens et al., 1986; Swindle et al., 1986; Ott, 1988; Ott and L6hr, 1992; Ott and Begemann, 1985), the Martian nitrogen signature of high 6 ~SN is seen only in the lithology C of EET 79001 (Becker and Pepin, 1984; Wiens et al., 1986) and the glass of Zagami (Marti et al., 1995), but not in the glass from shergottite LEW 88516 (Becker and Pepin, 1993). The objective of the present study is to decipher the noble gas and nitrogen components in ALH 84001 and to this extent we have studied a bulk sample and three density separates from ALH 84001. Cosmogenic neon and argon together with nuclear tracks have been discussed in Goswami et al. ( 1997); here we present our results on nitrogen and heavy noble gases (Xe, Kr, and Ar).
ALH 84001, recently classified as a Martian meteorite (Mittlefehldt, 1994a), is unique among the Martian meteorites known so far. Its long cosmic ray exposure age of 16 Ma (Eugster, 1994; Bogard, 1995; Miura et al., 1995; Swindle et al., 1995; Goswami et al., 1997) necessitates a separate impact event to deliver it from Mars. The presence of weathering products due to aqueous alteration processes on early Mars (Romaneck et al., 1994) together with its oldest gas retention age of 4 Ga (Ash et al., 1996; Goswami et al., 1997) makes it an appropriate sample to look for the early history of Mars. The recent finding of possible microfossils of ancient Martian life in ALH 84001 (McKay et al., 1996) made this meteorite the most valuable sample. The shergottite EET 79001, with a shock age of 180 Ma (Jones, 1985), gave the probable Martian atmospheric composition of nitrogen and noble gases of that era. If the composition of trapped gases can be characterised in ALH 84001, it may be possible to see the Martian atmospheric evolution over the eons. Though higher values of 4°Ar/36Ar and 129Xe/ ~32Xe, typical of Martian atmosphere (from Viking data, Owen et al., 1977) have been seen in several SNC meteorites
2. SAMPLES AND PROCEDURES A bulk sample and three density separates have been analysed for nitrogen and noble gases. While the bulk sample is made up of few pieces of several millimeters size, the density separation is carried out on the sample that has been crushed to < 100 #m size. The bulk sample is thus not an aliquot of the sample used for density separation. The density separates were prepared with the hope of getting the minerals of interest enriched in these separates. Since our main 5417
5418
S . V . S . Murty and R. K. Mohapatra
intent was to enrich maskelynite (p ~ 3 g/cc), carbonates ( < 3 g/ cc), and orthopyroxene ( > 3 g/cc), which is the principal mineral in ALH 84001, we simply used a solution of sodium polytungstate of density 3.0 g/cc for mineral separation. The meteorite sample crushed to about 100 #m size was allowed to settle for a day in the polytungstate solution after a thorough mixing, making sure that the density of the medium did not change during this period. The separated sinks, floats, and the suspended material are labelled, respectively, as >, <, and ~ 3 g/cc (density fractions). Each of these separates was cleaned thoroughly by ultrasonication in double distilled water, ethanol, and acetone and then air-dried before loading into the mass spectrometer. The three density fractions >, <, and ~3 g/cc, respectively, represent yields of 89%, 2%, and 9%. SEM analysis of the >3 g/cc indicated it to be predominantly orthopyroxene, while ~3 g/cc turned out to be a multimineral mixture. The amount of <3 g/cc was too small for characterisation. With 'the hope of further characterisation using the cosmogenic Ne and Ar as probes of major chemical elements, we carried out measurements without further sample characterisation.
2.1. Mass Spectrometry We have carried out a simultaneous analysis of nitrogen and noble gases by stepwise heating, using the VG 1200 mass spectrometer by standard procedures (Murty and Goswami, 1992). A 400°C combustion in 2 torr 02 is carried out for each sample to get rid of surficial contaminants. Subsequent gas extractions of the samples are done by RF heating in a Mo crucible. The sample is held at each temperature for about 45 min (about 30 min at 1600°C), and the evolved gases are collected on a stainless steel mesh (SSM) made of 2 #m S.S. powder at liquid nitrogen temperature. The He and Ne fraction is analysed first after a quick clean up on Ti-Zr and SAES getters. A portion of the total extracted gas ( 1 - 1 0 % ) is valved off for N2 analysis, while the rest is cleaned on Ti-Zr and SAES getters and separated into Ar, Kr, and Xe fractions by differential adsorption on charcoal fingers. Each fraction is let into the mass spectrometer and analysed. The volatile contaminants in the N_, fraction (chiefly, C,H) are converted into their oxides by exposing it to CuO at 750°C for 15 rain, reabsorbing the excess 02 back on to CuO at 400°C. The condensible gases (CO2 and H20) are separated from N2 by keeping liquid nitrogen on a cold finger while Nz is being transferred to another SSM in the inlet volume. Nitrogen is analysed in the molecular form on the Faraday cup, and the mass peaks 28 ( ~4N~4N), 29 (~4N tSN), and 30 (~SN tSN) are followed, the mass 30 peak being used as a monitor for possible CO contamination. For smaller samples, the masses 29 and 30 are also analysed on the multiplier, after analysing 28, 29 on Faraday cup, so as to precisely estimate the CO interference correction. Blanks interspersed before and after the sample are carried out at each temperature in identical fashion. Nitrogen blanks are in the range of few hundred picograms and are always -<5% of sample gas. A typical 1400°C blank gave 36Ar = 4 × 10-~2 S4Kr = 2 × 10- ~3, and 132Xe = 3 × l 0 -14 (in ccSTP units). Noble gas blanks are < 10% for all the temperature steps. Air standards are run to assess the sensitivity and the instrumental mass discrimination. The data reported here have been corrected for blanks and mass discrimination, and the associated errors have been propagated. In addition to these normal corrections the ~°Kr signal has been corrected for the 4~lAr~ interference by measuring the 4°Ar in the Kr fraction and a pre-determined 4 ° A r + / 4 ° A r + ratio. The nitrogen isotopic data have been corrected for CO interference, as detailed in Murty and Goswami (1992) and Murty (1997), by assigning all excess at mass 30 to be due to CO contribution. The correction to 615N due to CO is always <1%o. Data for 78Kr is not reported due to unresolvable benzene interference. 3. RESULTS The data for N, A r (Table 1), krypton (Table 2), and xenon ( T a b l e 3) are given in tables. The elemental ratios and the 6 ~SN corrected for cosmogenic contributions have been compiled in Table 4. T h o u g h the 400°C combustion is carried out mainly to remove surficial contaminants, looking
at the 6 ~SN of this fraction ( and also the isotopic composition of argon for bulk and > 3 g / c c fraction), it is evident that the gases mainly came from the sample, and hence they have been added to the totals. The concentrations and the isotopic compositions of the bulk sample and those calculated from the mass balance of the various density fractions do not match for all the species. While the concentrations of S4Kr and ~3~Xe match within 10%, those of 36Ar differ by 25%, and for nitrogen the difference is by 66%. This type of discrepancy can result from one of the three possibilities: material loss during density separation, contamination, and sample heterogeneity. Material loss is ruled out as the mass of individual density fractions recovered approximates the starting sample mass. The shifts in the isotopic composition of the noble gases as well as in the 6~5N observed in the density separates, as compared to the bulk sample, cannot be generated by atmospheric or organic contaminants. Considering the fact that the bulk sample is not a true aliquot of the powdered sample used for density separation, sample heterogeneity is the most likely explanation for the observed discrepancy, and our data interpretation is under this assumption.
3.1. Nitrogen The nitrogen content varies from 0.751 ppm in the bulk sample to 4.58 p p m in the ~ 3 g / c c fraction. The release pattern of 615N for all the samples has been shown in Fig. 1. The bulk sample and the > 3 g / c c fraction have similar release pattern, showing positive /515N in all temperature fractions with a m a x i m a in 615N at 1200°C. The other two samples have higher N contents but show entirely different release pattern for 6~5N. The ~ 3 g / c c sample starts with a small positive 6 ~5N in the 400°C combustion step and then changes to negative 6~5N in subsequent temperatures. The < 3 g / c c sample on the other hand starts with a sharp negative 6~5N in 400°C combustion, with subsequent temperatures exhibiting positive 6~5N. This is due to the presence of more than one type of nitrogen c o m p o n e n t in A L H 84001 with opposite 6 ~N signatures (in addition to the omnipresent comogenic nitrogen) and their differing proportions in various samples. The 400°C fraction of < 3 g / c c sample with 6~5N - - 4 0 % c could be yet another component, a most likely candidate being a Martian weathering product like a nitrate (Grady et al., 1993). Over all, the 6~5N varies from a high value of 85%e in the bulk to a lower value of - 17.6%~ in the --3 g / c c fraction. A m o n g the individual temperature fractions 6 ~SN varies between 201%c in the 1200°C fraction of the bulk sample and -21.3%c in the 1000°C fraction of ~ 3 g / c c (ignoring the 400°C fraction of < 3 g / c c ) . The lowest value of 6 ~5N found in A L H 84001 is similar to that found in Chassigny ( W r i g h t et al., 1992). Higher 6 ~SN values similar to the values observed in A L H 84001 have only been found in the glassy lithologies of EET 79001 and Zagami, a m o n g the SNCs ( B e c k e r and Pepin, 1984, 1986, 1993; Wiens et al., 1986; Marti et al., 1995; Wright et al., 1992). Because of the higher cosmic ray exposure age of A L H 84001, part of the reason for high 6~5N (particularly in the case of bulk sample where the nitrogen content is low) could be a contribution from cosmogenic nitrogen. But
5419
N and noble gases in ALH 84001 Table 1. Nitrogen and argon in ALH 84001 samples. (10 10 ccSTP/g) Temp. (°C)
N '~ (ppm)
615N (%0)
36ARM~*:
36Arcor."
38Ar/3*Ar (4°Ar/36Ar)M
(4°Ar/3eAr)c,,r
Bulk (692.38mg)
400*
0.075
1000
0.555
1200
0.057
1400
0.041
1600
0.023
Total
0.751
34.20 ±2.01 76.93 .37 201.7 .7 108.5 1.1 112.4 1.9 84.93 .65
0.2
0.2
10.9
7.4
22.3
4.2
20.6
2.2
0.6
0.2
54.7
14.2
0.1
0.1
15.3
11.2
33. l
6.0
16.0
2.3
1.3
0.7
65.8
20.3
0.2275 .0032 0.6493 .0012 1.258 .003 1.356 .002 0.9723 .0074 1.166 .002
466.3 2.5 5535 18 1131 4 247.3 .8 174.4 2.5 1662 5
466.3 2.5 8198 134 6037 245 2273 101 419 12 6366 158
0.2937 .0025 0.5793 .0020 1.267 .004 1.311 .002 0.8023 .0025 1.108 .003
2736 38 4629 15 495.9 1.6 204.5 .6 195.6 .9 1385 4
2849 40 6330 86 2753 113 1412 60 354 8 4503 88
>3 g/cc (670.07 mg) 400*
0.046
1000
0.478
12/)0
0.182
1400
0.112
1600
0.048
Total
0.866
27.92 .42 45.16 .72 80.39 2.59 58.72 .75 79.17 1.65 55.27 1.16
~3 g/cc (42.15mg) 400*
0.465
1000
3.794
1600
0.322
Total
4.581
2.57 .87 -21.35 1.01 -3.04 1.25 -17.64 1.01
<0.1
.
27.9
21.7
74.6
43.5
102.5
65.2
.
. 0.5205 .0043 0.7641 .0019 0.6977 .0025
. 10106 33 1985 6 4198 13
12980 147 3404 71 11545 96
<3 g~c (20.70mg) 400*
0.123
800
0.216
1600
2.899
Total
3.238
-39.71 .65 33.60 2.40 14.54 .62 13.75 .74
<0.1
.
5.7
4.6
42.2
20.5
47.9
25.1
.
. 0.4786 .0023 0.8854 .0019 0.8372 .0020
. 8583 47 5018 14 5441 18
10570 110 10325 265 10371 236
* Combustion in 2 torr 0 2. Errors in the concentrations are ±10%, and in the isotopic ratios are at 95% confidence level. Subscripts M and c,,,-, respectively, mean measured and cosmogenic corrected.
1200°C is not a high enough temperature for the major release of cosmogenic nitrogen. This is also born out by the cosmogenic argon data. At 1200°C about < 5 0 % o f cosmogenic 38Ar is released, whereas in the 1400°C fraction about equal amount o f cosmogenic 38Ar is released. With the reasonable assumption that cosmogenic nitrogen parallels cosmogenic 3BAr in release pattern, the 1400°C fraction should have about equal effect due to cosmogenic nitrogen on the
measured 6 ~SN. The distinctly lower 6 ~SN o f 1400°C fraction despite having similar nitrogen content, indicates that the higher 6 ~SN in 1200°C fraction is not entirely due to cosmogenic effect, but rather due to a trapped component. A similar trend is displayed by the 1200°C and 1400°C fractions of the > 3 g / c c , but due to the higher amount o f nitrogen in these steps as compared to the bulk sample, the relative increase in 6 ~SN is less. The other two density fractions ( - 3
5420
S.V.S. Murty and R. K. Mohapatra Table 2. Krypton data for ALH 84001. 84Kr = 100
Temp. (°C)
84Kr+ (10 12ccSTP/g)
8°Kr
8ZKr
83Kr
86Kr
4.565 ±.075 5.142 .075 5.024 .067 6.542 .345 4.868 .076
21.525 .112 22.032 .091 21.398 .076 23.302 .523 21.671 .101
20.832 .161 21.575 .019 21.222 .095 23.975 .128 21.182 .101
29.706 .067 29.973 .068 29.514 .207 28.278 .599 29.733 .104
5.208 .022 5.010 .051 5.385 .094 10.154 .194 5.233 .052
21.459 .087 21.051 .105 21.046 .098 25.989 .358 21.261 .101
20.638 .076 20.812 .104 21.139 .015 21.692 .189 20.832 .077
29.963 .047 29.851 .073 29.802 .365 30.239 .231 29.885 .126
Bulk 1000
83.52
1200
57.88
1400
42.11
1600
1.71
Total
185.2
>3 g/cc 1000
55.94
1200
70.53
1400
32.96
1600
2.44
Total
161.9
~3 g/cc 1000
90.51
1600
153.0
Total
243.5
6.178 .147 3.900 .108 4.747 .122
22.704 21.930 30.477 .151 .470 .350 20.469 20.048 30.039 .328 .127 .116 21.299 20.747 30.202 .262 .255 .203
5.888 .074
22.876 21.466 30.504 .445 .423 .201
<3 g/cc 1600
250.8
~Errors in the concentrations are ± 15% and in the isotopic ratios are at 95% confidence level.
g/cc and <3 g/cc), having relatively much higher nitrogen contents reveal less positive to negative 6'5N. To explain these 615N variations, at least two trapped nitrogen components, one with a positive 615N and another with negative 6~5N are needed (in addition to the cosmogenic component). The 6~5N of the individual temperature steps cannot be corrected for cosmogenic contribution, as there is no independent way of assessing the cosmogenic nitrogen released in each temperature fraction. But the total 6'~N for each sample can be corrected for cosmogenic contribution. From cosmogenic 2~Ne and the relation ~SNc = (4.0 ___0.5)2~Nec, valid for most silicate minerals and in particular for orthopyroxene (Mathew and Murty, 1993), we correct the measured ~SN for cosmogenic contribution. The corrected values are given in Table 4, and they range from -23%0 to +46%0. Such higher 6 ~SN has not been reported in the SNCs except in the glassy lithologies of EET 79001 (Becker and Pepin, 1984; Wiens et al., 1986) and Zagami (Marti et al., 1995)
and is most likely the indication for the presence of trapped Martian atmospheric nitrogen in ALH 84001. Wiens et al. (1986) have shown that the nitrogen data of the lithology C samples of EET 79001 lie on a straight line in a 6~5N vs. Ar/N plot, with the datum for Mars (from Viking data) falling on this line, meaning that the lithology C of EET 79001 has a nitrogen component of Martian atmospheric origin. In Fig. 2, the nitrogen data of ALH 84001, along with the data for glassy lithologies of EET 79001 and Zagami, are shown in a 6JSN vs. I / N plot. A least square line through the data of EET 79001,C, Zagami and ALH 84001 is dominated by the EET 79001,C and Zagami data, and the ALH 84001 data points scatter along this line. Instead, if we fit the ALH 84001 data points separately, they define a straight line (correlation coefficient = 0.78) with different slope than that defined by the data of EET 79001,C and Zagami. Apparently both these lines seem to pass through the Chassigny point (not included in the regression) signifying that both the trends can be explained by one common endmember, while the other endmember differs. If the other endmember is taken to be the Martian atmospheric component tbr ALH 84001, similar to the case of EET 79001,C, it is immediately obvious that the atmospheric component in ALH 84001 has a much lighter 615N than for the case of EET 79001,C. The Martian atmospheric 615N from ALH 84001 can be constrained to be ---46%~.(the highest value of the bulk sample) but certainly less than the value obtained from EET 79001,C. This is a clear indication that the 615N of the Martian atmosphere in the past is less than the present value of ~620%o (Viking data) and that 615N of Martian atmosphere has evolved with time. In Fig. 3 where 6 JSN is plotted against 36Ar/14N, the ALH 84001 data points define a distinctly different trend compared to the glassy samples of Zagami and EET 79001. Since no simultaneous measurements exist for argon and nitrogen for Chassigny, we derived the 36Ar/14N for the Chassigny point in Fig. 3, using the argon data from Ott (1988) and nitrogen data from Wright et al. (1992). The ALH 84001 data fall very much off the reference line which is a least squares fit through the EET 79001 ,C, Zagami and Chassigny points. A fit through ALH 84001 and Chassigny points (not shown in Fig. 3) has a poor correlation coefficient of 0.2. This could be understood as due to variable change in the 36Ar]lgN of each sample of ALH 84001 brought about by elemental fractionation and consequently disturbing the two component structure of Chassigny type and ancient Martian atmospheric components. Hence the different trend displayed by ALH 84001 in Fig. 3 can be due to one or a combination of the following reasons: (a) It could indicate an elemental fractionation of the trapped Martian atmospheric gases in ALH 84001, leading to a different 36Ar/t4N ratio, an effect similar to the one observed for argon, krypton, xenon (Drake et al., 1994; Swindle et al., 1995; Miura et al., 1995); (b) The Martian atmospheric 36Ar/~4N ratio is different when the gasses are trapped in ALH 84001 and in glasses of Zagami and EET 79001. We will return to this point later. 3.2. Noble Gases
The noble gases in ALH 84001 are a mixture of cosmogenic, radiogenic, and trapped components. The argon iso-
5421
N and noble gases in ALH 84001 Table 3. Xenon data for ALH 84001. 132Xe = 100
Temp. (°C)
~32Xe* (10 t : ccSTP/g)
I'-~Xe
126Xe
128Xe
129Xe
13°Xe
131Xe
134Xe
1~6Xe
0.5916 ±.0192 0.6622 .0120 I).7012 .0829 0.8578 .1711 0.6573 .0359
0.6784 .0317 1.0215 .0814 0.7089 .0990 1.9961 .156 0.8553 .0744
9.189 .031 8.640 .093 8,431 .194 10.129 .260 8.741 .108
206,08 1.12 221.24 .58 221.20 .47 223.20 1.09 217.32 .70
16.381 .316 16.126 .065 15,623 .166 18.870 .661 16.082 .166
79.488 .378 80.947 .231 79.971 .051 83.679 .660 80.324 .223
39.724 .390 39.867 .399 40.463 .236 41.612 .232 4(t.020 ~348
34.007 .147 34.687 .104 34.085 .145 36.176 .562 34.358 .132
0.6845 .0304 0.5785 .0162 0.6365 .0219 1.3876 .2083 0.6295 .0242
0.7869 .0764 0.7225 .0439 0.7246 .0767 2.8333 .1063 0.7789 .0591
8.833 .368 8.455 .056 8.715 .030 12.519 .195 8.672 .116
201.75 1.80 201.27 .71 214.33 .50 183.17 5.12 203.95 .97
15.944 .043 15.925 .087 16.475 .097 19.237 .087 16.121 .080
78.318 .634 80.638 .271 81.107 .055 84.200 4.006 80.376 .371
40.201 .570 39.267 .157 40.405 .082 40.578 1.149 39.739 .249
33.547 .371 34.319 .114 35.673 .159 33.116 .521 34.446 .184
0.9255 .1743 0.3200 .0926 0.4555 .1109
1.2528 .1495 0.7521 .0274 0.8642 .0548
10.479 .473 8.269 .169 8.764 .237
152.27 2.29 166.46 1.83 163.29 1.94
16.273 .339 17.166 .143 16.966 .187
78,885 1,194 80,504 .430 80,142 ,601
38.329 .665 38.490 .610 38.454 .623
31.968 .568 33.018 .093 32.783 .199
0.8992 .0376
1.8941 .1869
8,949 .289
215.88 8.84
17.530 .324
87,426 2,943
40.056 .272
33.665 .319
Bulk 1000
7.3(/
1200
12.53
1400
8.00
1600
1/.34
Total
28.17
>3 g~c 1000
5.81
1200
15.97
1400
6.55
1600
0.59
Total
28.92
- 3 g/cc 10110
10.76
160(/
37.31
Total
48.07
< 3 g/cc
1600
43.70
~ Errors in the concentrations are _+15% and in the isotopic ratios are at 95% confidence level.
topes 36Ar and 3SAr are mostly dominated by cosmogenic contributions, but a fair a m o u n t of trapped 36Ar is clearly present. The K-Ar age calculated by using K = 108 + 16 ppm (Mittlefehldt, 1994b) and assuming all 4°Ar to be radiogenic yields an age > 5 Ga, clearly indicating that part of 4°Ar also belongs to a trapped component. If we assume that the trapped argon is of terrestrial atmospheric origin and accordingly correct for trapped 4°Ar, the K-Ar age will be still 4.6 Ga. The measured values 36Ar and the ratio 4°Ar/ 36At for each temparature step have been corrected for cosmogenic contribution, and the corrected values are listed in Table I. For this correction, we assumed the (36Ar/38Ar) = 4.1 _+ 0.2 for the trapped component, as found in EET 79001,C ( W i e n s et al., 19861 and (38Ar/36Ar) = 1.5 for cosmogenic component. The corrected values of 4°Ar/36Ar represent a two c o m p o n e n t mixture of trapped and radiogenic 4°Ar, and the m i n i m u m value of (4°Ar/36Ar) .... a m o n g the various temperature steps of all samples should more closely correspond to the trapped ratio. The m i n i m u m value
for (4°Ar/36Ar) ..... occurs in the 400°C and 1600°C fractions of bulk sample and the 1600°C fraction of > 3 g / c c fraction and correspond to - 4 0 0 . But these fractions have very small amount of argon released, and the corrected values may be influenced heavily by small terrestrial atmospheric contribution or cosmogenic c o m p o n e n t correction uncertainties. Even ignoring these, the next lowest (4°Ar/36Ar) .... = 1412 in the 1400°C step of > 3 g / c c fraction. Hence the trapped value of (4°Ar/36Ar) in A L H 84001 -< 1412, a value m u c h smaller than 2400, as found in EET 79001,C. Trapped argon also hence suggests that the Martian atmosphere trapped in A L H 84001 is m u c h older than the one found in EET 79001,C. Taking the trapped (4°Ar/36Ar) = 1400 (the minim u m value found in A L H 84001 samples), a K-At age of 4.2 Ga will be obtained, in good agreement with the Ar-Ar age of 4 Ga ( A s h et al., 19961. Argon of Martian atmospheric origin is thus clearly indicated in A L H 84001. Krypton and xenon on the other hand are mostly dominated by trapped component, and the contribution to 84Kr and 132Xe
5422
S.V.S. Murty and R. K. Mohapatra
Table 4. Elemental ratios of trapped noble gases and nitrogen (corrected for spallation) in ALH 84001 samples.
300
36Ar Temp• (°C)
36Ar
84Kr
]32Xe
L32Xe
100 33.5 28.2 74.5 50.4
11.4 4.6 5.2 5.1 6.6
250
'~N (X 107)
615N " (%c)
200
C.C.C.C.C~ J-
I
Bulk 1000 1200 1400 1600 Total
7o 100
50
11.8
45.9 _+4.9
ALH 84001 B Bulk G > 3glcc E - 3g/cc L < 3glcc
0 ~.~ C E
>3 g/cc
-SO
1000 1200 1400 1600 Total
9.6 4.4 5.0 4, l 5.6
,
,
J
14.6
20.7 4.5
202 116 136
8.5 4.1 5.1
8.9
-23.2 1.2
144 50.6 57.4
4.9 5.8 5.7
l
,
,
,
,
1
I
. . . .
I
2
. . . .
I
3
. . . .
I
4
,
,
i
i
5
t I N ( p p m ) "t
~ 3 g/cc 1000 1600 Total
,
0
193 37.6 35.0 118 70.2
Fig. 2. Cosmogenic corrected 6 ]~N has been plotted against 1/N for ALH 84001 samples. Data for the glassy lithologies of EET 79001 [C1, C2 (Becker and Pepin, 1984); C3 (Wiens et al., 1986)] and Zagami [ZA, ZB (Marti et al., 1995)] and for the bulk sample of Chassigny (Wright et al., 1992) are also included in the plot. The regression through the data points (excluding Chassigny) (...) defines a mixing line between Chassigny type and EET 79001,C type components, but merges with the line through only data of EET 7900l,C and Zagami. The regression line through only ALH 84001 data defines a line ( _ _ ) with different slope.
>3 g/cc 800 1600 Total
4.8
8.8 1.0
'~ Corrected for spallation.
and ]36Xe/132Xe (up to 0.36) are also clearly indicative of Martian atmospheric signatures. The higher 136Xe/132Xe cannot be explained as due to in situ fission component, as the uranium content is only - 1 2 ppb (Dreibus et al., 1994), though a minor fission contribution cannot be ruled out. 3.3. R e l e a s e P a t t e r n
by cosmogenic c o m p o n e n t is less than one percent. In Table 4, the trapped ratios (36Ar/132Xe) and (84Kr/132Xe) for each temperature fraction of the four samples have been compiled. The higher values of the ratios ]29Xe/t32Xe ( u p to 2.23)
In Fig. 4, the release pattern of the trapped components 3~'Ar, 8nKr. and ~32Xe as well as the (4°Ar/36Ar) .... for the
300 12
ALH 84001 • Bulk • > ~cc • - $ gl©¢
200
160
•
<
200
~CC O
E
120
z
z
m¢<~ lOO
80 4o
B IIG ~L
lO
ALH 84001 B Bulk G > 3g/cc E ~ 3g/cc L < 3glcc
Vl(
0
•i'r.
I E
,4~
1o I
0
20
n
I
40
~
I
60
~
I
80
~
0 I
,
,
,
I
2
i
i
i
I
i
i
i
4
I
i
i
i
6
(36Ar / 14 N )ATOM,105
1 O0
Cumulative N release (%)
Fig. 1. Release pattern of c5'SN for the four ALH 84001 samples. Temperature in units of 100°C are marked against each point.
Fig. 3 . 6 'SN is plotted against the elemental ratio 36Ar/HN. Data sources from literature are same as for Fig. 2 and Fig. 5. The best fit line through the data of EET 79001,C, Zagami and Chassigny is shown for reference. The ALH 84001 data fall off this line.
N and noble gases in ALH 84001
5423
10 s 36Artrap
50 _11
132Xetrap
/'
40
(4°Ar/36Ar)~o~./.,~" /
6000
104
7 /
t~
~.
/ /
?
30
i,.. ;
/
4000
20
x
O4 O3
•"
2000
10
103
CD ~/"
I
~/ 10
I 400
800 Temperature
~ 1200
I 0 1600
Fig. 4. Release pattern of the spallation corrected ~Ar/36Ar(which reflects the release pattern of radiogenic 4°Ar) and those of trapped 36Ar, 84Kr, and 132Xefor the bulk sample of ALH 84001.
bulk sample have been shown. It can be seen that the ratio (4°Ar/36Ar) .... which represents radiogenic 4°Ar, and the trapped component of argon, krypton, and xenon have similar release pattern. The slight displacement of 132Xe peak to higher temperature is a reflection of the relatively higher retentivity of Xe. The >3 g/cc fraction also shows similar trend. This will be possible only under one of the following two scenarios. Either that the trapped component got trapped at the same time when radiogenic 4°Ar retention started, implying that the trapped component is as old as the K-Ar age; or that the mineral or phase wherein the gases are trapped has later became a part of the sample, and the matching release pattern is just coincidental. Gilmour et al. (1996) have suggested that the carrier phase of trapped xenon in ALH 84001 is orthopyroxene, the major mineral composing >90% of bulk and not any minor phase. It is, therefore, most likely that the atmospheric component in ALH 84001 is as old as the K-Ar age. The lower 6~5N and the lower (4°Ar/36Ar)t~,p. in ALH 84001, as compared to that in EET 79001,C, also strengthen this suggestion.
......
"
/
EET 79001, C I
,
~lTi[~ ~ ' ~
/
Earth
~/"~
~
2
(°C)
S°lar
••T
8000
S4Krtrap
ALH84001
~
Chassigny
• • •
Bulk > 3g/cc ~ 3g/cc
•
< 3g/cc
101
5
10
50
100
84Kr / 132Xe
Fig. 5. The noble gas elemental ratios ~Ar/]~2Xe and S4Kr/'~2Xe for the temperature fractions of ALH 84001 samples are plotted here. Line joining the Chassigny and EET 79001,C and the trend for chondrites are shown for reference. Literature sources: Chassigny (Ott, 1988); EET 79001,C (Swindle et al., 1986); Mars (Owen et al., 1977); Solar, Earth (Ozima and Podosek, 1983).
plot off the trend defined by the other SNCs, consistent with similar observations made earlier by Swindle et al. (1995) and Miura et al. (1995), again clearly indicating that the elemental ratios 36Ar/J32Xe and 84Kr/~32Xe in ALH 84001 are less than expected for one of the endmembers defining the two component mixing trend for the SNCs. A Martian
2.0
3.4. Elemental Ratios The elemental ratios of the trapped argon, krypton, and xenon have been summarised for the various temperature fractions of the samples in Table 4. The trapped components have been obtained by correcting the measured amounts for cosmogenic contributions. These ratios are similar to the ones derived by Miura et al. (1995). Figure 5 is a plot of the elemental ratios 36Ar/132Xe vs. ~4Kr/132Xe for the temperature steps of ALH 84001 samples. The trends defined by chondrites and the other SNC meteorites are also shown in the figure. Most of the ALH 84001 data points fall below the trend line defined by the SNCs, indicating lower than expected 36Ar/132Xe in ALH 84001, as compared to the other SNCs. The plots of the isotopic ratio ]29Xe/132Xe against the elemental ratios 84Kr/132Xe (Fig. 6) and 36Ar/132Xe (Fig. 7) also show that the ALH 84001 data
x 1.5 /
ALH 84001 • Bulk • > 3glcc • ~ 3g/cc • < 3glcc
/ / B Chassigny
1.0 I
,
,
,
,
I
5
,
.
.
.
I
10
.
,
,
,
I
,
,
15
,
,
I
20
.
,
.
.
I
,
i
25
84Kr / 132Xe Fig. 6. The xenon isotopic ratio t29Xe/~32Xe has been plotted against the elemental ratio 84Kr!132Xefor the temperature fractions of ALH 84001 samples from this study. Line joining Chasigny and EET 79001,C points is shown for reference. Literature sources same as in Fig. 5.
5424
S.V.S. Murty and R. K. Mohapatra of the trapped xenon (Drake et al., 1994) no such mineral has been so far identified in ALH 84001. The exact mechanism that led to elemental fractionation in ALH 84001, is not clear at the moment, but if it is similar to the case of Nakhla, no carrier phase of weathering origin has so far been identified in ALH 84001.
EET 79001~;~
2.0
A~
3.5. Krypton Components in ALH 84001 ~
t.5 ~
"
ALH 84001
/
•
Bulk
/
•
~ Sglcc
~]Chanlgny , I 200
•
< 3~JCC
J
1,0
I 400
~
,
I
600
, 800
36Ar 1132Xe
Fig. 7. The xenon isotopic ratio 129Xe/132Xe has been plotted against the elemental ratio 36Ar/]32Xe for the temperature fractions of ALH 84001 samples from this study. Line joining Chassigny and EET 79001,C is shown for reference. Literature sources same as in Fig. 5.
atmospheric component that is elementally fractioned, enriching the heavier noble gases, similar to that found in Nakhla (Drake et al., 1994) can explain the A L H 84001 data. While in the case of Nakhla the clay mineral iddingsite, a weathering product has been suggested to be the carrier
Krypton isotopic data for all the four samples shows excess at S°Kr over and above the spallation component. Swindle et al. (1995) have also reported similar excesses at so. 82Kr" In Fig. 8, the ratios 8°Kr/8~Kr and 82Kr/84Kr have been plotted for the temperature fractions of all the samples. Trends expected for mixing of spallation and neutron capture produced krypton from Br(n, y ) reaction, with air like krypton have also been shown in the plot. The spallation krypton is assumed to be similar to that of diogenites, and the ratio (8°Kr/82Kr)c = 0.71 _+ 0.15 has been adopted from Eugster and Michel ( 1995 ). The (8°Kr/82Kr)Br depends on the energy spectrum of the neutrons and is about 2.5 and 3.3 for epithermal and thermal neutrons, respectively (Marti et al., 1966). Figure 8 clearly indicates that an admixture of neutron capture produced krypton can account for the shifts in the data points from simple two component mixture of trapped and cosmogenic components. Though in principle the shifts can also be accounted for by a mass fractionated trapped component, it requires a variable extent of fractionation tbr each datum which is not feasible. So neutron capture effects from Br(n, y ) are clearly present in ALH 84001, similar to earlier such findings in EET 79001,C and other SNC meteorites
0.10
0.08
Y 0 ¢0
0.06
0.04
0.20
0.24
0.28
0.32
82Kr 184Kr Fig. 8. Plot of krypton isotopic ratios 8°Kr/S~Kr and 82Kr/84Kr for the temperature fractions of ALH 84001 samples from this study. Trends for mixing trapped component (assumed to be air like) with cosmogenic and Br(n, 3') components are also shown. The cosmogenic component is taken to be similar to diogenites with (8°Kr/82Kr)~ = 0.71 _+ 0.15 (Eugster and Michel, 1995). The outer and inner envelopes of the Br(n, 3') mixing trend represent the (8°Kr/ 82Kr)B, values of 3.3 and 2.5 respectively for the thermal and epithermal neutrons (Marti et al.. 1966).
N and noble gases in ALH 84001 (Becker and Pepin, 1984; Bogard et al., 1984; Swindle et al., 1986; Ott, 1988). Whether these neutron effects are produced in situ in the meteorite during its recent cosmic ray exposure, or they represent some process that took place in the Martian atmosphere, will be of interest to know. In Table 5, the calculated excess S°Kr* due to 79Br(n, T) contribution for our samples, as well as those of literature (Swindle et al., 1995; Miura et al., 1995) have been given. The excess ~°Kr* in our bulk sample agrees with those of literature except for the sample 84001, 28, where an order of magnitude higher excess has been observed (Swindle et al., 1995). The 8°Kr* roughly parallels the excess 129Xe* (excess over the Chassigny value of I29Xe/132Xe = 1.03) which is a measure of the amount of Martian atmospheric component present in ALH 84001. For sample 28 of Swindle et al. (1995) such is not the case. The 8°Kr* for #28 could be erroneous, and so we do not consider this datum in further discussion. Halogen data for any of these samples is not available to make an estimate of the neutron fluence needed to produce the 8°Kr* in situ in the meteoroid. For three bulk samples of ALH 84001, Ash et al. (1995) report a C1 content in the range 3.1 - 11.8 ppm, with an average of 8 ppm. Dreibus and W~inke (1985) report C I / B r ~ 120 for a set of SNCs. Assuming that this ratio is valid for ALH 84001, we estimate a Br content of ~ 6 6 ppb for the bulk samples of ALH 84001. For thermal neutrons a fluence of 6 × 10 ~5 n/ cm 2 is needed to produce the observed 8°Kr* = 0.6 × 10 -12 ccSTP/g. For epithermal case the fluence needed will be ten times smaller (6 × 1014 n/cm2). This neutron fluence is an order of magnitude higher than that derived for the case of Nakhla, to explain the 8°Kr* observed in it (Ott, 1988). Considering the high exposure age of 16 Ma for A L H 84001 and its pre-atmospheric radius of ~ 10 cm, based on nuclear tracks (Goswami et al., 1997), it is quite possible to generate a neutron fluence of the order of 6 × 1015 n / c m 2 in the meteoroid, based on the calculation of Spergel et al. (1986) (assuming L-chondrite composition for A L H 84001). But for a meteoroid of ~ 10 cm size a negligible fraction of these neutrons, will be in the energy range of interest (Spergel et al., 1986) to produce the observed 8°Kr*. The lower
Table 5. Neutron capture produced 8°Kr* and (4°Ar/t29Xe*)in the ALH 84001 samples.
4°Ar
4°ArTotal40mFTrap.
~29Xe* 8°Kr* Total Trap.* 129Xe* 129Xe* References Sample Bulk >3 g/cc - 3 g/cc <3 g/cc , 28 ,29 #1 #2
(10 ~2 ccSTP/g) 32.2 29.2 29.0 49.3 6.1 25.8 23. l 24.8
(10 8 ccSTP/g)
0.62 909 197 1.5 765 283 1.l 4300 913 3.1 2600 351 7.3 820 240 0.49 . . . 0.78 778 175 0.38 1050 228
(X 106) 0.28 0.26 1.5 0.53 1.35 . 0.34 0.42
0.06 0.10 0.31 0.07 0.39 0.07 0.09
*4~)ArTrap. = 1400 × 36ArTrap.. [11 Swindle et al., 1995; [2] Miura et al., 1995.
This work This work This work This work [1] [1] [2] [2]
5425
2.5 EET 79001, C
2.0
X ,t-
t.5
1.0
~ Chassigny i
0.30
i
i
I 0.32
• n
n
i
I
L
0.34
L
<3g/cc a
l
i
0.36
136Xe 1132Xe
Fig. 9. Plot of the xenon isotopic ratios 129Xe/132Xeand 136Xe/ ~32Xe for the temperature fractions of the ALH 84001 samples from this study. Line joining the Chassigny and EET 79001,C points is shown for reference. Literature sources same as in Fig. 5.
(131Xe/t26Xe) -- 2.5 _+ 1.0 in our A L H 84001 samples also rules out significant neutron effects (Kaiser, 1977). The neutron effects observed in krypton are, therefore, not likely to be in situ produced in the meteoroid, but rather are likely produced in the Martian atmosphere or on Martian surface and are a trapped component, as has been also suggested for EET 79001,C (Becker and Pepin, 1984). The parallel increase of 129Xe* and 8°Kr* also supports this interpretation. 3.6. Xenon Components in ALH 84001 The xenon isotopic ratios 129Xe]132Xe and ~36Xe[132Xe are typical of the Martian atmospheric component as found in EET 79001,C (Becker and Pepin, 1984; Swindle et al., 1986). In Fig. 9, the xenon isotopic ratios 129Xe/132Xe have been plotted against 136Xe/132Xe for the temperature fractions of all the samples in this study. The two component mixing line between the Chassigny component (Ott, 1988 ) and EET 79001,C (Swindle et al., 1986) has also been shown in Fig. 9. Most data points lie along this mixing line, confirming that the trapped xenon in ALH 84001 is a mixture of these two components. However, there are four data points that significantly deviate from the mixing trend. Two of these points ( 1600°C of bulk and 1400°C of > 3 g / c c ) show excess 136Xe with respect to the mixing trend, while the other two points (1400°C of bulk, 1600°C of < 3 g / c c ) show excess 129Xe with respect to the mixing trend. The excess 136Xe* corresponds to ~0.1 × 10-J3 and ~0.7 × l0 -13 (in c c S T P / g units) for the 1600°C of bulk and 1400°C of > 3 g / c c fraction, respectively. For an uranium content of 12 ppb in ALH 84001 (Dreibus et al., 1994), a fission 136Xe of about 0.4 × 10 -13 c c S T P / g can be generated over the 4.56 Ga crystallisation age of the meteorite (Jagoutz et al., 1994). Hence the '36Xe* can be accounted for by the 136Xefrom 238U. In principle, the excess 136Xe could also be
5426
S.V.S. Murty and R. K. Mohapatra
partly or entirely due to 244pu fission, considering the antiquity of the sample. Excess t29Xe* with respect to the mixing trend amounts to 0.3 × 10 -12 and 12 × 10 12 (in ccSTP/g units), respectively, in the 1400°C of bulk and 1600°C of <3 g/cc. This 129Xe* can only be explained as due to the in situ decay of live 129I. Such excess ~29Xe* has also been observed in ALH 84001 earlier (Swindle et al., 1995). In a neutron irradiated sample of ALH 84001 Gilmour et al. (1995) have observed a correlated high temperature release of 129Xe* and ~2SXe*, suggesting that 129Xe* could be really due to in situ decay of t~9I. Considering that the Sm-Nd crystallisation age of ALH 84001 is 4.56 Ga (Jagoutz et al., 1994), it is quite conceivable that some refractory phases have retained their 129Iproduced 129Xe intact to reveal excess 129Xe * in high temperature fractions. From the excess 129Xe* ( = 129I) and the canonical value of 1291/127I = 10 -4, we estimate an I content of about 0.02 ppb ( 1400°C bulk) and 0.7 ppb (1600°C, <3 g/cc), while the measured ! contents of SNCs are much higher than 12 ppb (Dreibus and W~inke, 1985). Though I data of Antarctic meteorites are often prone to contamination, the I content of Shergotty (an observed fall) is also 36 ppb (Dreibus and W~inke, 1985). The 129Xe * correlated I of course need not represent the total I content of the meteorite, since most I could be uncorrelated with 129Xe*. The point that we would like to make is that the 129Xe* correlated I derived here is not absurd. Depending on the choice of the trapped ratio ~28Xe/ ~S2Xe for ALH 84001 (Air or AVCC) a maximum excess 'eSXe* (over and above the normal spallation xenon) of up to 0.4 × 10 ~2 ccSTP/ g (in 3 g/cc) can be found in the samples analysed. This excess could be due to neutron effects on I. In the absence of I contents it will be difficult to assess whether this I related ~28Xe* is produced in situ in the meteoroid or in Martian atmosphere and also the neutron fluence required. However, if we assume that the C1/I = 3000 for Shergotty is valid for ALH 84001, we can estimate an I content of 2.7 ppb in the bulk sample of ALH 84001, using C1 = 8 ppm (Ash et al., 1995). From the neutron fluence derived using S°Kr*, we can estimate ~2SXe* to be 1.6 and 3.7 (in 10-14 ccSTP/g units) for thermal and epithermal neutrons, respectively, which is a factor of ten less than the maximum 12SXe* calculated for the bulk sample (26 × ]0 -14 ccSTP/g). Using arguments similar to the accounting of 8°Kr*, one might say that the neutron effects on iodine are also of Martian atmospheric or surficial origin, and ~2~Xe* is incorporated into ALH 84001 as a trapped component. Since the spallation correction is negligible, at least for the ratios 129Xe/132Xe and 136Xe/132Xe, the measured values directly correspond to the trapped component. The ratios 129Xe/132Xe and 136Xe/132Xe are less than the SPB values (Swindle et al., 1986), but most of these data fall on the mixing line joining the Chassigny and EET 79001,C data points in the plot of 129Xe/132Xe Vs. 136Xe/132Xe. The lower values for these ratios in ALH 84001 are hence due to the presence of Chassigny type xenon in addition to Martian atmospheric component. About 15% Chassigny type-Xe will explain the observed xenon composition of the bulk sample. Similarly the ~ 3 g/cc fraction shows the lowest 129Xe/132Xe and 136Xe/132Xe ratios among all the samples of ALH 84001. Presence of about 56% of Chassigny type Xe, in this fraction
will explain the lower values. Lighter nitrogen in this fraction also signifies the presence of a non-atmospheric component in this fraction, and Chassigny type N-component would also explain the occurance of lighter N in this fraction. From the L29Xe/136Xe data of ALH 84001, vis-a-vis Zagami and EET 79001, Mathew et al., 1997 have also suggested that either xenon in ALH 84001 shows evolution of the Martian atmosphere or the presence of an indigenous xenon component, in addition to the atmospheric component. But the later possibility is more likely, considering the short half-life of 129I (16 Ma) and consequent rapid evolution of 129Xe/136Xe early in Martian history.
3.7. Characteristics of Trapped Martian Atmosphere from ALH 84001 From the xenon and nitrogen isotopic signatures and the isotopic correlations of xenon (Fig. 9) and nitrogen (Fig. 2), it is very clear that ALH 84001 has trapped a Martian atmospheric component. Since the gas retention age of ALH 84001 is 4 Ga (Ash et al., 1996; Goswami et al., 1997), and the release patterns of radiogenic 4°Ar, and trapped 3~Ar (S4Kr and 132Xe) are similar, the Martian atmospheric component in this meteorite should correspond to that era. Comparing the characteristics of the trapped gases in ALH 84001, with those of EET 79001,C (that represents Martian atmospheric signatures of more recent past) will be helpful in understanding the evolution of Martian atmosphere. In Table 5 we have compiled the values of 129Xe*, trapped 4°Ar*, and the ratio (4°Ar/~29Xe) for all our samples as well as those of earlier workers (Swindle et al., 1995; Miura et al., 1995). ~29Xe* can be taken as a measure of the amount of Martian atmospheric component present in the sample. With the exception of the glassy lithologies in EET 79001 and Zagami, ALH 84001 seems to have the highest proportion of Martian atmospheric gases among the SNCs. Since ALH 84001 lacks shock produced glass, it seems feasible to trap gases by some alternate mechanisms as well, though the exact trapping machanism and the carrier phase have not been identified unambiguously. The ratio (4°Ar/~2')Xe*) in EET 79001 ,C as well as Shergotty is - 1 . 4 × 106 (Becker and Pepin, 1984; Ott, 1988). Since in situ produced radiogenic 4°Ar in both these meteorites will be very small, due to their very young gas retention ages, this ratio should be characteristic of the Martian atmosphere of recent past. The (4°Ar/129Xe*) ratio for all the ALH 84001 samples is clearly less than the value of 1.4 × 106, even if we consider the total 4°Ar. However, due to its high gas retention age, the in situ produced radiogenic 4°Ar will constitute a major proportion of the total 4°Ar. As we don't know the K contents of the individual samples (density fractions), we cannot cleanly subtract the radiogenic 4°At contribution. As an alternative, we calculate, the 4°Art from the 36Art (Table 4), using the ratio (~°Ar/~6Ar)~ = 1400. The ratio (4°Ar/L29Xe*) for the various samples of ALH 84001 lie in the range 6 - 3 9 × 10 4. Considering the uncertainties involved in evaluating the small amounts of 36Art, the variation in the estimated 4°Ar/~29Xe among the various samples of ALH 84001 is understandable. Hence the 4°ArflJ29Xe* as observed in ALH 84001 is certainly lower
N and noble gases in ALH 84001 by about an order of magnitude as compared to the value found in EET 79001,C. Two possible explanations can be proposed for this observation. An elemental fractionation between argon and xenon during the process of incorporating Martian atmospheric gases into ALH 84001, leading to a lower value of (4°Ar~/~29Xe*). Compared to the value of (36Ar/132Xe)t ~700 as found in EET 79001,C (Becket and Pepin, 1984), a value of only in the range of 50-135 in ALH 84001 samples indicates that the elemental fractionation has depleted Ar in the trapped component of ALH 84001 by factors of 5-14. If the same mechanism is applicable for the observed lower (4°Ar~/~29Xe*) ratios, the depletion of 4°Art should be by the same factor. The (4°Ar,/129Xe*) for ALH 84001 is lower than the value of EET 79001,C by upto a factor of ~23 much higher than expected by elemental fractionation. The second possibility is that the lower ratio of (4°Ar/ ~2'~Xe*) in A LH 84001 indeed represents the Martian atmospheric value at 4 Ga ago. If this is true, it means that the 4°Ar in the Martian atmosphere has not completely evolved at the time of incorporation ( ~ 4 Ga ago) into ALH 84001, while ~29Xehas almost completely evolved (based on similar values of ~2~)Xe/132Xe in ALH 84001 as well as in EET 79001,C). This is a more realistic explanation, and it could be that the elemental fractionation has caused a further lowering in this ratio. Another elemental ratio that shows a large difference as compared to EET 79001,C is (36Ar/14N). This value is ~60 × 10 -6 in EET 79001,C (Wiens et al., 1986) while it varies between (0.5 to 1.5)10 6 for ALH 84001 samples. A fractionation process leading to enrichment of N over Ar can explain the decrease in the 36Ar/HN ratio. Elemental fractionation of the type proposed by Drake et al. (1994) and operating for the noble gases in Nakhla cannot explain the lower 3~Ar/~4N ratio. Drake et al. (1994) have proposed a two stage process to explain the noble gas elemental fractionation in Nakhla: ( 1) solubility of noble gases in water leading to enrichment of heavy noble gases and (2) adsorption of noble gases from aqueous medium by clay minerals (iddingsite), also leading to an enrichment of heavy noble gases (Ozima and Pododsek, 1983). The process (1) above leads to an enrichment of argon over N2 (Ozima and Podosek, 1983), while the process (2) may not lead to much fractionation, leading overall to an enhanced 36Ar/~4N ratio, as against the observed lower ratio in ALH 84001 assuming that a fractionation mechanism similar to the case of Nakhla is also applicable to ALH 84001. The only possibility that remains is the decrease of N content of the Martian atmosphere from 4 Ga to the present, leading to a higher value of ~6Ar/~4N = 6 × 10 5 as observed in EET 79001,C. The Martian atmospheric evolution proposed by Pepin (1994) also leads to the above evolutionary paths for 36At, N. After the end of the hydrodynamic escape episode (starting at 4.46 Ga and declining at 4.44 Ga), further processing would be only by sputtering. From 4.44 Ga onwards nitrogen abundance evolved due to loss by sputtering as well as by the photochemical processes leading to decrease of nitrogen in the Martian atmosphere and at the same time increasing 6 ~SN to the current value of 620%e (Pepin, 1994). These model evolutionary paths lead to an increase in 36Ar/J4N and 6 ~SN
5427
from 4 Ga ago to the values observed in the present Martian atmosphere. Thus the argon and nitrogen systematics of the Martian atmosphere, as observed in ALH 84001 (corresponding to 4 Ga ago) and in EET 79001,C (corresponding to recent time) are consistent with the evolution model of Pepin (1994). The evolutionary path however can not be constrained with just data from two eras only. Assuming that the highest cosmogenic correlated 6 ~SN of 46%c (also corresponds to the sample with the lowest nitrogen content) has the highest proportion of the Martian atmospheric-N component, we can say that the b tSN of Martian atmosphere at ~ 4 Ga is ->46%0. 4. SUMMARY AND CONCLUSIONS
The nitrogen and noble gas records in ALH 84001 have clearly shown the presence of a Martian atmospheric component in it that is characterised by: ( 1 ) Xenon isotopic ratios J29Xe/132Xe and I36Xe/132Xe, very similar to those found in EET 79001,C and 6~5N -> 46%0. (2) Trapped xenon and nitrogen in ALH 84001 are consistent with a two component mixture of Chassigny type and EET 79001,C type components. (3) In addition xenon shows a small contribution from in situ fission from 23~U and also a ~29Xe excess due to in situ decay of live ~29Iin high temperature fractions. Neutron capture produced krypton from bromine (n, 3') reaction is clearly present, but this is most likely produced in the Martian atmosphere or surface and incorporated into ALH 84001 as a trapped atmospheric component. (4) Noble gas elemental ratios show a fractionation trend enriching the heavy noble gases, similar to what has been found in Nakhla (Drake et al., 1994). (5) Lower 6 ~SN (->46%0) and lower 4°Ar/36Ar (-<1400) for the atmospheric component in ALH 84001, together with the fact that radiogenic 4°Ar and trapped noble gases have similar release pattern, strongly suggest that the atmospheric component was trapped, when radiogenic 4°Ar retention started. Comparing the nitrogen and noble gas records from EET 79001,C (which represents the Martian atmosphere of more recent past) with those of ALH 84001 ( representing Martian atmospheric records of 4 Ga ago), we can clearly trace the following evolution for the Martian atmosphere. ( 1) Similar xenon isotopic composition in both EET 79001,C as well as ALH 84001 suggests that xenon in Martian atmosphere has evolved completely by 4 Ga and remained unchanged. Nitrogen isotopic ratio was less heavy at 4 Ga at 6~SN _> 46%0 and evolved to the present value of 620%0 over time, through the nonthermal escape of nitrogen, leading to heavier nitrogen at present. (2) The radiogenic elemental ratio (4°Ar/ ~29Xe) is much less in ALH 84001, than compared to EET 79001,C. This difference is not only a consequence of elemental fractionation, but indicates that ~2t~Xehas been completely evolved in the Martian atmosphere, while 4°At has not been completely degassed into atmosphere. The stable elemental ratio 36Ar/14N on the other hand is also less in ALH 84001 as compared to EET 79001,C, but due to a different reason. While 36Ar and nitrogen have both evolved at 4 Ga, there has been continuous loss of nitrogen from the Martian atmosphere leading to an increase in the ratio 36At/ 14N as well as 6 ~SN with time. These observations are consis-
5428
S . V . S . Murty and R. K. Mohapatra
tent with the h y d r o d y n a m i c escape model of Pepin ( 1 9 9 4 ) for the evolution of Martian atmosphere. Acknowledgments--We thank the Meteorite Working Group at JSC,
Houston for providing the ALH 84001 sample. Critical comments from three anonymous reviewers have greatly helped in improving the presentation. REFERENCES
Ash R. D., Knott S. F., and Turner G. ( 1995 ) Evidence for the timing of the early bombardment of Mars. Meteoritics 30, 483. Ash R. D., Knott S. F., and Turner G. (1996) A 4-Gyr shock age for a Martian meteorite and implications for the cratering history of Mars. Nature 380, 57-59. Becker R. H. and Pepin R. O. (1984) The case for a Martian origin of the shergottites: Nitrogen and noble gases in EET 79001. Earth Planet. Sci. Lett. 69, 225-242. Becker R. H. and Pepin R. O. (1986) Nitrogen and light noble gases in Shergotty. Geochim. Cosmochim. Acta 50, 993-1000. Becker R. H. and Pepin R. O. (1993) Nitrogen and Noble gases in a glass sample from the LEW88516 Shergottite. Meteoritics 28, 637-640. Bogard D. D. and Johnson P. (1983) Martian gases in an antarctic meteorite? Science 221, 651-654. Bogard D.D., Nyquist L.E., and Johnson P. (1984) Noble gas contents of shergottites and implications for the Martian origin of SNC meteorites. Geochim. Cosmochim. Acta 48, 1723-1739. Bogard D. D. ( 1995 ) Exposure-age-initiating events for Martian meteorites: Three or four? Lunar Planet. Sci. 26, 143-144. Drake M., Swindle T. D., Owen T., and Musselwhite D. S. (1994) Fractionated Martian atmosphere in the nakhlites? Meteoritics 29, 854-859. Dreibus G. and W~inke H. (1985) Mars, a volatile rich planet. Meteoritics 20, 367-381. Dreibus G., Burghele A., Jochum K. P., Spettle B., Wlotzka F., and W~inke H. (1994) Chemical and mineral composition of ALH 84001. A Martian orthopyroxenite. Meteoritics 29, 461. Eugster O. (1994) Orthopyroxenite ALH 84001: Ejection from Mars(?) 15 Ma. Meteoritics 29, 464. Eugster O. and Michel T.H. (1995) Common asteroid break-up events of eucrites, diogenites, and howardites and cosmic ray production rates for noble gases in achondrites. Geochim. Cosmochim. Acta 59, 177-199. Gilmour J. D., Whitby J. A., Ash R. D., and Turner G. (1995) Xenon isotopes in irradiated and unirradiated samples of ALH 84001. Meteoritics 30, 510-511. Gilmour J. D., Whitby J. A., and Turner G. (1996) Carrier phase of xenon in ALH 84001. Meteor. Planet. Sci. 31, A51. Goswami J. N,, Sinha N., Murty S. V. S., Mohapatra R. K., and Clement C. J. (1997) Nuclear tracks and light noble gases in Allan Hills 84001: Preatmospheric size, fall characteristics, cosmic-ray exposure duration, and formation age. Meteor. Planet. Sci. 32, 91-96. Grady M. M., Wright I. P., Franchi I. A., and PiUinger C. T. (1993) Nitrates in SNCs: Implications for the nitrogen cycle on Mars. Lunar Planet. Sci. 24, 553-554. Jagoutz E., Sorowka A., Vogel J. D., and Wanke H. (1994) ALH 84001: Alien or progenitor of the SNC family? Meteoritics 29, 478-479. Jones J. H. (1985) The youngest meteorites, I, A 180 m.y. igneous age for the shergottites--The constraint of petrography. Lunar Planet. Sci. 16, 406-407.
Kaiser W. A. (1977) The excitation functions of Ba(p, x)mXe (m = 124-136) in the energy range 38-600 MeV, the use of cosmogenic xenon for estimating burial depths and real exposure ages. Phil. Trans. Roy. Soc. London A285, 757-779. Mathew K. J. and Murty S. V. S. (1993) Cosmic ray produced nitrogen in extra terrestrial matter. Proc. Ind. Acad. Sci. (Earth Planet. Sci.) 102, 415-437. Mathew K. J., Kim J. S., and Marti K. (1997) Xenon components in Martian meteorites: Evidence for atmospheric evolution? Lunar Planet. Sci. 28, 885-886. Marti K., Eberhardt P., and Geiss J. (1966) Spallation, fission, and neutron capture anomalies in meteoritic krypton and xenon. Z. Naturforsch. 21a, 398-413. Marti K., Kim J. S., Thakur A. N., McCoy T. J., and Kiel K. (1995) Signatures of the Martian atmosphere in glass of the Zagami meteorite. Science 267, 1981-1984. McKay D. S., et al. (1996) Search for past life on Mars: Possible relic biogenic activity in Martian meteorite ALH 84001. Science 273, 924-930. Mittlefeldht D. W. (1994a) ALH 84001, a cumulate orthopyroxenite member of the Martian meteorite clan. Meteoritics 29, 214-221. Mittlefeldht D. W. (1994b) Erratum for Mittlefeldht (1994a). Meteoritics 29, 900. Miura Y. N., Nagao K., Sugiura N., Sagawa H., and Matsubara K. (1995) Orthopyroxenite ALH 84001 and Shergottile ALH77005: Additional evidence for a Martian origin from noble gases, Geochim. Cosmochim. Acta 59, 2105-2113. Murty S. V. S. and Goswami J. N. (1992) Nitrogen, noble gases, and nuclear tracks in lunar meteorites MAC88104/105. Proc. Lunar Planet. Sci. Conf. 22, 225-237. Murty S. V. S. (1997) Noble gases and nitrogen in Muong Nong tektites. Meteor. Planet. Sci. 32, 687-691. Ott U. (1988) Noble gases in SNC meteorites: Shergotty, Nakhla, Chassingny. Geochim. Cosmochim. Acta 52, 1937-1948. Ott U. and Begemann F. (1985) Are all the Martian meteorites from Mars? Nature 317, 509-512. Ott U. and L6hr H. P. (1992) Noble gases in the new shergottite LEW88516. Meteoritics 27, 271. Owen T., Biemann K., Rushneck D. R., Biller J. E., Howarth D. W., and Lafleur A. L. (1977) The composition of the atmosphere at the surface of Mars. J. Geophys. Res. 82, 4635-4639. Ozima M. and Podosek F. A. ( 1983 ) Noble Gas Geochemistr3,. Cambridge Univ. Press. Pepin R. O. (1994) Evolution of the Martian atmosphere. Icarus 111, 289-304. Romanek C. S., et al. (1994) Record of fluid rock interactions on Mars from the meteorite ALH 84001. Nature 372, 655-657. Spergel M. S., Reedy R. C., Lazareth O. W., Levy P. W., and Slatest L. A. (1986) Cosmogenic neutron Capture-produced nuclides in stony meteorites. J. Geophys. Res. 91D, 483-494. Swindle T. D., Caffee M. W., and Hohenberg C. M. (1986) Xenon and other noble gases in shergottites. Geochim. Cosmochim. Acta 50, 1001-1015. Swindle T. D., Grier J. A., and Burkland M. K. (1995) Noble gases in orthopyroxenite ALH 84001: A different kind of Martian meteorite with an atmospheric signature. Geochim. Cosmochim. Acta 59, 793-801. Wiens R. C., Becker R. H., and Pepin R. O. (1986) The case for a Martian origin of shergottites II. Trapped and indigenous gas components in EET 79001 glass. Earth Planet. Sci. Lett. 77, 149158. Wright 1. P., Grady M. M., and Pillinger C. T. (1992) Chassigny and the nakhlites: Carbon bearing components and their relationship to Martian environmental conditions. Geochim. Cosmochim. Acta 56, 817-826.
