11:15 〜 11:30
[PCG20-09] 気相凝縮実験による漸近巨星分枝星周非晶質ケイ酸塩ダストの化学組成の制約
キーワード：凝縮、実験、星周ダスト、ケイ酸塩
Mid-infrared spectroscopic observations of oxygen-rich asymptotic giant branch (AGB) stars have shown prominent dust features at 10 and 18 µm originating from Si–O stretching and O–Si–O bending vibrations of amorphous silicate, respectively. Synthetic optical functions derived from observed spectra of circumstellar/interstellar dust and laboratory dust analogs known as “astronomical silicate”[1,2] have been widely used in the interpretation of observational data, while the nature of astronomical silicates, such as structure and chemical composition, remains poorly constrained. Laboratory amorphous silicate dust analogs with various Mg/Si ratios and Mg/Fe ratios cannot reproduce the relative strength of 10 and 18 μm [3,4]. Most of the synthetic experiments were performed in simple Si-O-Mg(-Fe) systems and the effects of Al, Ca, and Na are scarcely discussed. Fe-free silicate glass of the solar abundance (Mg, Al, Si, Ca, and Na) is proposed as a realistic dust analog [5] but still cannot replace the astronomical silicate. In addition, it is unclear if the spectra are different when Fe metallic particles are within or outside amorphous silicate grains. In this study, we did condensation experiments in the Si-Mg-Fe-Na-Al-Ca-Ni-O and in the Si-Mg-Na-Al-Ca-O system to examine the effects of Fe metallic particle and Al content.
Experiments were performed using the induction thermal plasma (ITP) system (JEOL TP-40020NPS, [6]). Starting materials with compositions of Si:Mg:Fe:Na:Al:Ca:Ni = 1:1.03:0.85:0.06:0.08:0.06:0.05 (CI-02), Si:Mg:Na:Al:Ca = 1:1.03:0.06:0.08:0.06 (CI-03), 1:1.03:0.06:0.75:0.06 (CI-04) were evaporated. CI-02 and CI-03 were prepared with the CI chondritic composition [7]. The products were analyzed by XRD, FT-IR, and EPMA and observed by TEM.
FT-IR spectra of all run products reproduce slightly convex upward slopes at 20–25 µm of circumstellar dust emissivity, unlike previous dust analogs such as Mg-Fe-Si-O glasses [4] nor astronomical silicate [1,2]. This could be attributed to the presence of Al, Ca, and Na. There is no significant difference between the spectra of CI-02 (with Fe) and CI-03 (without Fe) regardless of the presence of Fe metallic particles. Dust emissivity of a silicate-rich AGB star (Z Cyg) was derived from observations at multiple variable phases as a function of temperatures of the inner dust shell [8]. The Al-rich sample (CI-04) showed the most similar spectral shapes to the observed dust emission of Z Cyg in terms of the 11 µm shoulder and relative strength of 10 and 18 µm peaks. However, the peak at 14 µm due to Al–O vibration [9] is stronger than the observation.
Our main finding is that the circumstellar silicate dust must not be pure Mg-Fe-silicates but contain Al and possibly other cations such as Ca and Na. Based on the present results, the circumstellar silicate dust must be enriched in Al relative to the CI chondritic composition (Al/Si=0.08) although not as much as CI-04 (Al/Si=0.75). However, amorphous presolar silicate grains generally have low Al content [10]. This contradiction could be explained by two hypotheses: the circumstellar dust is (1) homogeneous silicate dust with Al/Si somewhere between 0.08 and 0.75. (2) a mixture of silicate dust with CI chondritic chemical composition and a small amount of Al-rich silicate dust.
Our future work will be condensation experiments with starting materials in various Al/Si ratios to examine the content of Al in amorphous silicates more systematically.
[1] Draine B. T. and Lee H. M. (1984) ApJ, 285, 89. [2] Ossenkopf V. et al. (1992) A&A, 261, 567. [3]Jäger C. et al. (2003) A&A,408,193. [4] Dorschner J. et al.(1995) A&A, 300, 503. [5]Speck, A. K. et al (2015). ApJ, 809, 65. [6] Kim T. H. et al. (2021) A&A, 656, A42. [7] Lodders K. et al. (2009) in Landolt-Börnstein Landolt–Bornstein (Berlin: Springer) [8] Onaka T. et al. (2002) A&A, 388, 573. [9] Tarte P. (1967) AcSpA 23, 2127. [10] Nguyen A. N. et al. (2016) ApJ, 818, 51.
Experiments were performed using the induction thermal plasma (ITP) system (JEOL TP-40020NPS, [6]). Starting materials with compositions of Si:Mg:Fe:Na:Al:Ca:Ni = 1:1.03:0.85:0.06:0.08:0.06:0.05 (CI-02), Si:Mg:Na:Al:Ca = 1:1.03:0.06:0.08:0.06 (CI-03), 1:1.03:0.06:0.75:0.06 (CI-04) were evaporated. CI-02 and CI-03 were prepared with the CI chondritic composition [7]. The products were analyzed by XRD, FT-IR, and EPMA and observed by TEM.
FT-IR spectra of all run products reproduce slightly convex upward slopes at 20–25 µm of circumstellar dust emissivity, unlike previous dust analogs such as Mg-Fe-Si-O glasses [4] nor astronomical silicate [1,2]. This could be attributed to the presence of Al, Ca, and Na. There is no significant difference between the spectra of CI-02 (with Fe) and CI-03 (without Fe) regardless of the presence of Fe metallic particles. Dust emissivity of a silicate-rich AGB star (Z Cyg) was derived from observations at multiple variable phases as a function of temperatures of the inner dust shell [8]. The Al-rich sample (CI-04) showed the most similar spectral shapes to the observed dust emission of Z Cyg in terms of the 11 µm shoulder and relative strength of 10 and 18 µm peaks. However, the peak at 14 µm due to Al–O vibration [9] is stronger than the observation.
Our main finding is that the circumstellar silicate dust must not be pure Mg-Fe-silicates but contain Al and possibly other cations such as Ca and Na. Based on the present results, the circumstellar silicate dust must be enriched in Al relative to the CI chondritic composition (Al/Si=0.08) although not as much as CI-04 (Al/Si=0.75). However, amorphous presolar silicate grains generally have low Al content [10]. This contradiction could be explained by two hypotheses: the circumstellar dust is (1) homogeneous silicate dust with Al/Si somewhere between 0.08 and 0.75. (2) a mixture of silicate dust with CI chondritic chemical composition and a small amount of Al-rich silicate dust.
Our future work will be condensation experiments with starting materials in various Al/Si ratios to examine the content of Al in amorphous silicates more systematically.
[1] Draine B. T. and Lee H. M. (1984) ApJ, 285, 89. [2] Ossenkopf V. et al. (1992) A&A, 261, 567. [3]Jäger C. et al. (2003) A&A,408,193. [4] Dorschner J. et al.(1995) A&A, 300, 503. [5]Speck, A. K. et al (2015). ApJ, 809, 65. [6] Kim T. H. et al. (2021) A&A, 656, A42. [7] Lodders K. et al. (2009) in Landolt-Börnstein Landolt–Bornstein (Berlin: Springer) [8] Onaka T. et al. (2002) A&A, 388, 573. [9] Tarte P. (1967) AcSpA 23, 2127. [10] Nguyen A. N. et al. (2016) ApJ, 818, 51.