|
隕石や地球試料で求められた元素(主に酸素、窒素、希ガス)の同位体比に基づいて惑星(地球や隕石母元体)の起源と進化を研究した。種々の隕石につき求められた酸素同位体比(170/160, 130/170)データを統計的に処理し(bootstrap法)、太陽系内の酸素同位体比分布状態を推定した。惑星・隕石母元体の形成過程に制約を課すことが期待できる。
Mass-independent O isotopic fractionations in CAIs have long been attributed to the mixing of nucleosynthetic components, but no other major elements have showed such nucleosynthetic contribution arguing against this hypothesis. Hence, Clayton (1) proposed that self-shielding of CO in the proto-solar nebula resulted in the preferential production of 17-18O isotopes, which were then accommodated in solid objects via H2O. Kuramoto & Yurimoto (2), and Young and Lyons (3) also employed the same process to explain the mass-independent O isotope fractionation in CAIs. As a consequence of this process, these authors concluded that the solar O is close to the extreme O isotope observed in CAIs (D17O ~ -50 permil), but differs from those of planetary bodies such as Earth or the Mars (D17O ~ 0). However, since the average values for D17O for bulk chondrites and achondrites were close to zero, and their variance became smaller with increasing size of a planetary object, Ozima and Podosek (4) argued that all the planetary objects were derived by random sampling of a whole planetesimal population which represents D17O value of the protosolar nebula or the Sun. In this preliminary statistical argument, we assumed a specific relation between variance (standard deviation) and the size of planetary object that may not in general be valid. To circumvent this difficulty, here we used a more general statistical approach. The result supports the previous conclusion that terrestrial objects (meteorite parent bodies, Mars, Earth, Moon etc) should have the same O isotope as the Sun. We also discuss its implications on the evolution of the early solar system.
For D17O data of 16 different classes of chondrites compiled by K. Lodders (written communication), we applied a multistep-multiscale bootstrap method (5) to derive a relation between a standard deviation (s) and the size (N) of planetary objects (a chondrite parent body). Here, we assumed that each chondrite parent body has randomly sampled N-ch bodies of planetesimals. Similarly we calculated a standard deviation for D17O data for 12 different classes of achondrites (Lodders, ibid). From the s-N relation with the observed standard deviation for achondrites, we estimated the number of planetesimals (N-ach) which accreted to have formed achondrite parent body. This gave (N-ach)/(N-ch) of 9 ~ 30. If we assume a spherical shape of a radius R for a meteorite parent body, the ratio of their radius (R = (N)^1/3)) is 2 ~ 3. The latter ratio appear to be reasonable for the relative size between chondrite and achondrite parent body. Also considering that chondrite and achondrite bulk samples have mean values of D17O which are close to 0, we suggest that the planetary objects (chondrite and achondrite parent bodies, the Mars, and the Earth) formed from progressive random accretion of planetesimals, and hence should have the same D17O ratio as the solar nebula which represents the average D17O of a whole planetesimal population.
|