1. Art and Diseño

46th Lunar and Planetary Science Conference (2015)
2275.pdf
A SILICON ISOTOPIC STUDY OF A FUN-LIKE FORSTERITE-BEARING INCLUSION FROM
ALLENDE. K. Fukuda1, H. Hiyagon1, N. Takahata2, Y. Sano2, and A. Hashimoto3, 1Department of Earth and Planetary Science, Graduate School of Science, The University of Tokyo, Tokyo 113-0033, Japan ([email protected]), 2Atmosphere and Ocean Research Institute, The University of Tokyo, Chiba 277-8564, Japan,
3
Depertment of Cosmoscience, Hokkaido University, Hokkaido 060-0810, Japan.
Introduction: There is a minor group of CalciumAluminum-rich Inclusions (CAIs), so called FUN
(Fractionation and Unknown Nuclear effects) CAIs,
which are characterized by (i) large mass-dependent
fractionation in O, Mg, and Si, (ii) relatively large
mass-independent fractionation in Ca, Ti, Cr, and other
elements, and (iii) little or no excess 26Mg from the
decay of 26Al [e.g., 1, 2, 3, 4]. The origin of FUN CAIs
is still not well understood, but they may have information about an earliest stage of the solar system evolution. Recently, [5] conducted evaporation experiments for Mg- and Si-rich melts simulating precursor
materials of FUN CAIs, and tried to estimate the
chemical compositions of the precursor materials of
some FUN CAIs based on the observed massdependent isotopic fractionation of O, Mg, and Si and
changes in chemical compositions of the melts. To
better understand the chemical compositions of precursor materials of FUN CAIs, we conducted electron
microprobe analyses and in-situ Si isotopic measurements of a FUN-like inclusion from Allende, called
AL1B-F, which shows large mass-dependent fractionation in O and Mg [6].
Sample and Methods: AL1B-F is a forsteritebearing inclusion which shows large mass dependent
fractionation in O and Mg [6]. The central part of this
inclusion is composed of coarse-grained forsteritic
olivine, spinel, and Al-Ti-rich pyroxene. The outer part
is mainly composed of spinel with lath-shaped hibonite
and corundum. Between forsterite-rich core and spinelrich outer part are filled with abundant secondary minerals (e.g., sodalite and nepheline).
A polished section of AL1B-F was mapped in Mg,
Ca, Al, Ti, Si, Na, Cl, K, S, Cr, Fe, and Ni x-rays with
a focused electron beam, 15keV, 200nA beam current,
5s per pixel acquisition time, and resolution of 1 x 1µm
per pixel using the JEOL JXA-8530F field-emission
gun electron probe microanalyzer (FEG-EPMA), the
University of Tokyo. Quantitative analyses were also
conducted with a ~2µm beam, 15keV, and 12nA beam
current .A ZAF matrix correction method was applied.
Silicon isotopic analyses were made using the NanoSIMS 50 at AORI, the University of Tokyo. A primary beam of Cs+ with a diameter of 1µm and an intensity of 10pA was used for the analyses. Primary
beam was rastered by 5 x 5µm during the analysis to
stabilize the secondary ion intensities. Before each
analysis, the sample was pre-sputtered for 3-5 minutes.
A mass resolving power was set to ~3500, which is
sufficient to separate 28SiH- and 29Si- peaks. Secondary
ions, 28Si-, 29Si- and 30Si-, were detected using an electron multiplier (EM) (Tr5) by a peak-jumping mode,
and 18O- was detected simultaneously with 28Si- using
another EM (Tr2). The measured 29Si-/28Si- and 30Si/28Si- were normalized to the literature value of
0.050633 and 0.033474, respectively [7], and those of
the samples were also normalized to the average of the
measured San Carlos olivine data or the Traversella
Al-Ti-rich pyroxene data.
Results and Discussion: Silicon isotopic mesurements were peformed on 9 forsteritic olivines and 3
Al-Ti-rich pyroxenes in AL1B-F. Figure 1 shows the
δ30Si-δ29Si diagram of several olivine and pyroxene
grains. These minerals show a large mass-dependent
fractionation of up to ~22‰/amu. Al-Ti-rich pyroxene
data show rather homogeneous compositions with an
average of +39‰ in δ30Si, while olivine data show a
large range of fractionation with δ30Si from +12 to
+31‰. These results suggest that Al-Ti-rich pyroxene
formed only at the last stage of the evaporation event,
while olivine probably crystallized at various stages of
the evaporation event. Furthermore, all the data for
olivine and Al-Ti-rich pyroxene are located slightly
above the mass-dependent fractionation line. Under the
decribed analytical condition, however, the contribution of 28SiH- to 29Si- is estimated to be at most ~0.1‰.
Hence, the reason of the offset is not clear, but one
possiblility is a mass-independent isotopic fractionation effect in excess 29Si (or a corresponding deficiency
of 28Si or 30Si), because a small but similar effect was
observed in other FUN CAIs [8, 9].
The relationship between the mass-dependent fractionation degree of Si and that of Mg and O [6] in olivine are similar to those of experimental resuslts by [5].
If we assume that mass-dependent fractionations in
olivine were resulted from a simple one-stage evaporation event and that a precursor composition of AL1B-F
was similar to the FUN 1 composition (53.14 wt%
MgO, 41.27 wt% SiO2, 2.97 wt% Al2O3, and 2.36 wt%
CaO) [5], ~78% of Mg and ~75% of Si must have been
lost (evaporated) from the molten precursor of AL1B-F
based on the experimentally determined isotopic fractionation factors [5]. Because of the presence of abundant secondary minerals, it is not possible to precisely
46th Lunar and Planetary Science Conference (2015)
2275.pdf
determine the bulk chemical composition of AL1B-F.
If we assume that secondary minerals in AL1B-F are
alteration products of primaly melilites with Ak mole%
of 23 or 89, which are compositions of melilites in the
Vigarano forsterite-bearing FUN CAI 1623-5 [10, 11],
the estimated precursor composition for AL1B-F
would be 34.0 wt% MgO, 42.7 wt% SiO2, 12.8 wt%
Al2O3, and 10.6 wt% CaO for Ak mole% of 23, and
37.0 wt% MgO, 48.5 wt% SiO2, 5.6 wt% Al2O3, and
9.1 wt% CaO for Ak mole% of 89, respectively. These
results suggest that the precursor of AL1B-F also have
a Mg- and Si-rich composition like C1, 1623-5, and
CMS-1 FUN CAIs [5, 12]. Further isotopic studies,
(e.g., Ca and Ti isotopes) may be required to to better
understand formation processes and a relationship between AL1B-F and other FUN CAIs.
References: [1] Clayton R. N. and Mayeda T. K.
(1977) GRL, 4, 295–298. [2] Wasserburg G. J. et al.
(1977) GRL, 4, 299–302. [3] Clayton R. N. et al.
(1988) Philos. Trans. R. Soc. London., 325, 483–501.
[4] Krot A. N. et al. (2014) GCA, 145, 206–247.
[5] Mendybaev R. A. et al. (2013) GCA, 123, 368–384.
[6] Hiyagon H. and Hashimoto A. (2008) Meteoritics
& Planet. Sci., 43, #5128. [7] Zinner E. K. (1989) U.S.
Geol. Surv. B., 1890, 145–162. [8] Clayton R. N. et al.
(1984) GCA, 48, 535–548. [9] Molini-Velsco C. et al.
(1983) LPS XIV, 509–510. [10] Davis A. M. et al.
(1991) GCA, 55, 621–637. [11] McKeegan K. D.
(2005) LPS XXXVI, Abstract #2077. [12] Williams C.
D. et al. (2014) LPS XXXXV, Abstract #2146.
[13] Knight K. B. et al. (2009) GCA, 73, 6390–6401.
30
δ29Si (‰)
25
20
15
10
olivine
Al-Ti-rich pyroxene
MF line
5
0
0
10
20
30
40
50
60
δ30Si (‰)
Figure 1: The δ30Si-δ29Si plot for olivine and Al-Tirich pyroxene in AL1B-F. The blue line represents the
Mass-Fractionation (MF) line with the parameters recommended by [13]. Errors are 2σ.
Figure 2: (a) A back-scattered electron image of
AL1B-F. (b) A combined X-ray elemental map in Mg
(red): Ca (green): Al (blue) of AL1B-F. Abbrevations;
Fo = forsterite, Sp = spinel, Px = Al-Ti-rich pyroxene.