Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
STRONTIUM AND RUBIDIUM IN STONE METEORITES Paul W. Cast Department of Geology and Mineralogy University of Minnesota The contents of Rb, Sr, in some instances K, and the Sr87/Sr86 ratio have been measured for a number of calcium-rich achondrites and hypersthene-bronzite chondrites. Most of the measurements reported here were done on the same mass spectrometer, under closely similar conditions, in order to minimize any errors arising from varying mass discrimination effects. Much of this data has been reported previously and requires little comment. TABLE 1 Rb and Sr in Ca-rich Achondrites Achondrite Rb Sr Sr87/Sr86 Pasamonte 0.21 79. 3 0. 7012; 0.7005; 0.6997; 0.702 Sioux City 0. 25 65.7 0. 7011; 0.7018 Nuevo Laredo 0. 37 72. 1 0. 7027 Moore County 0. 16 78.6 Note that the enrichment of Sr relative to Rb in these achondrites is so great that there would be a negligible increment added to Sr87 even in 5 A. E. The presumed primordial ratios of Sr87/Sr86 are in good agree- ment with one another; a value of 0. 7015 is suggested as the best estimate of the primordial ratio. (Note that the results for Pasamonte are in serious disagreement with that reported by Schumacher (1956) on the same sample, 0. 686 ± 0. 006.) Analyses have been made on four chondrites, three of which give results that appear to be quite reliable. The last four entries in Table 2, 85
TABLE 2 K. Rb, and Sr in ChondritesO) K Rb Sr 87/86 Age(2) Forest City 820 2.75 10.3 0.755 4.8 A. E. Modoc 3.45 13.T_ 0.757 5.0 A. E. 862 3.03(3) 10.8 872 3.04 -- 0.7555 4.6 A. E. (4) Richardton(5) 9. 8 0.756 818 2.95 11.4 844 3.17(3) 9.9 2.73 n.d. 821 3. 02(3) 10.0 803 2.67 n.d. <10 <0.4 n.d. 884 3.10 n.d. 838 2.93 n.d. 2.74 9.3 (1) Absolute values in parts per million. (2) The ages given are model ages in the sense that they date the hypo- thetical differentiation of calcium-rich achondrites from chondritic material. They are really dates for the isolation of Sr with an 87/86 ratio of 0. 7015 from material of the mean composition of the chon- drites, calculated using a Rb87 half-life of 50 A. E., rather than the recent value of 47 A. E. (Flynn and Glendenin 1960). (3) These numbers represent duplicate determinations on two aliquots of the same solution of the sample. (4) The 4. 6 A. E. age is probably more reliable. (5) Note that the variance in Rb and Sr contents is very large, both for replicate analyses on the same samples and between samples. This variance appears to be a property of the meteorite itself, and is not associated with experimental error. 86
on Richardton, represent analyses of four 0. 7 g samples, using different chemical procedures. It is of great interest that digestion with perchloric acid alone, for 15 hours, did not succeed in bringing into solution the mete- oritic K and Rb (10 ppm K by this procedure compared with~830 by others). The small amount of residue from this digestion appeared to be a single mineral, by optical criteria. Powder photographs of the residues afforded several weak lines corresponding to plagioclase feldspar, as might be expected from the inferred high K and Rb content. Perhaps the sampling problem is due to the presence of large grains of plagioclase, inhomogeneously distributed, in which most of the K and Rb is situated. Reynolds: We have done some density separations on Richardton samples, and find that the K is not associated with material of density similar to feldspars. Gast: The identification is somewhat uncertain, since the index of re- fraction of the residue in question was too high for plagioclase. (Sub- sequent to the conference, a density separation was made on the insoluble residue. The density of the mineral under consideration was found to be less than 2. 7. Thus plagioclase cannot be ruled out on the basis of density.) Anders: Could the material be maskelynite, with a small amount of in- cluded plagioclase to give the weak x-ray lines? Gast: Perhaps, although the grains displayed a definite birefringence, which would not be expected of a glassy material. Anders: It could be strained maskelynite; we have observed a number of of strained silicate glasses in meteorites. If one takes a grand average of the data on Richardton, the model age obtained is 4. 5 A. E. The problem of sample variability, noted in the Richardton results and present to some extent for Modoc as well, becomes acute when studying Beardsley. (Table 3.) The first four lines refer to analyses on Beards ley I, a sample obtained from Nininger which was collected two years after infall. Beardsley II (data below the dashed line) was obtained from Anders, ultimately from the Perry collection of the University of Michigan, and was collected the day after infall. The ages 5. 8 and 5. 5 A. E. have been previously reported. Beardsley II is very different. It has the highest Rb content known for any meteorite, and the highest K content measured in a chondrite. While the total Sr is not changed from that of Beardsley I, the radiogenic 87
1227 1247 TABLE 3 K, Rb and Sr in the Beardsley Chondrite K (ppm) Rb (ppm) Sr (ppm) 87/86 Age (A. E.) Sample I 917 4.90 11. 2 0.812 5. 8 895 4.83 10.4 0.812 5.5 905 4.80 10.8 - _. _ 4. 67 14.5 14.6 Sample II 10.6 n. d. 0.960 4. 25 Sr87 clearly follows the Rb. The age derived (which is known to high accuracy because of the very high Sr87/Sr86 ratio) is much lower than the ages derived from the data on Beardsley I. In considering these ages, it is very important to keep in mind that there can be only one primordial Sr87/Sr86 ratio, by the definitions used in constructing the model. If this is not 0. 7015, as assumed, but rather some higher value in order to bring the data on the two Beardsley samples into agreement (for example, 0. 73), then it is necessary to explain the Sr87/Sr86 ratio in the achondrites in some special way. Any such explan- ation must lead to the conclusion that Beardsley is significantly different from all the other meteorites studied. A more promising explanation seems to be that of leaching. [A dis- cussion of this topic followed, with Gast, Anders, Hurley, Goles, Turkevich. Broecker. and Arnold participating, among others. It was pointed out that the leaching would have to be such that it would remove Rb and radiogenic Sr but not primordial Sr. The results on perchloric acid digestion were cited as an objection to a large effect by leaching. Gast indicated that he intended to test this hypothesis experimentally, by attempting to leach out Rb and Sr87 artificially.] Addendum: Several experiments attempting to leach the alkalies out of the second Beardsley sample were performed subsequent to the conference. In the first attempt (A), two grams of the powdered 88
meteorite were washed in warm water (~70o C) for 24 hours. The alkalies in the water were then determined in the usual way. The procedure was repeated in boiling water for 4 hours on the same powder (B). The results of these experiments are shown in Table 4. Eighteen and 55 per cent of TABLE 4 Per Cent Extractable Alkalies in Beards ley II K Rb Cs A 3.5% 12.6% 35% B 1.8% 5.7% 20% the Rb and Cs respectively were removed in these experiments. The results support the hypothesis that rubidium and radiogenic strontium were removed from the first Beardsley sample while it was in the ground. The separation of Rb and Cs from K during this experiment further sug- gests that all the alkali metals are not in the same phase in this meteorite. REFERENCES Flynn, K. P., and Glendenin. L. E. (1959) Phys. Rev. 116, 744. Schumacher, E. (1956) Helv. Chim. Acta. 39. 538-547. 89
