The Astrophysical Journal, 666: L57–L60, 2007 September 1 䉷 2007. The American Astronomical Society. All rights reserved. Printed in U.S.A.
EXOTHERMIC CHEMICAL REACTIONS CAN DRIVE NONTHERMAL CRYSTALLIZATION OF AMORPHOUS SILICATE GRAINS Chihiro Kaito, Yu Miyazaki, Akihito Kumamoto, and Yuki Kimura1 Laboratory for Nano-Structure Science, Department of Physics, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu-shi, Shiga 525-8577, Japan Received 2007 April 9; accepted 2007 July 13; published 2007 August 9
ABSTRACT To explain how cometary silicates crystallize yet still preserve volatile interstellar ices in their parent comets, we experimentally demonstrate the possibility of chemical-reaction-driven crystallization, which is called nonthermal crystallization, using laboratory-synthesized amorphous Mg-bearing silicate grains. Analog silicate grains ∼50–100 nm in diameter covered with a carbonaceous layer consisting of amorphous carbon, CH 4, and other organics to a thickness of ∼10–30 nm were used as models. The analog silicate grains crystallized via the direct flow of surface reaction energy, which is produced by the graphitization of the carbonaceous layer due to oxidation at room temperature in air, into the silicates. The experimental results imply that amorphous silicates are transformed into crystalline silicates as the grains leave the comet’s surface, rather than as the comet was accreted 4.5 billion years ago. Thus, primordial ices and amorphous silicate grains are predicted to reside in most comets until they approach the Sun. Subject headings: astrochemistry — comets: general — methods: laboratory — radiation mechanisms: nonthermal Mg-bearing silicate grains covered with a carbonaceous layer including CH4 and other organic materials.
1. INTRODUCTION
Several comets, such as comets P/Halley, Bradfield 1987 XXIX, Levy 1990 XX, and Mueller 1993a, show 10 mm emission features characteristic of a mixture between crystalline and amorphous magnesium-rich olivine (Hanner 1999). Since interstellar silicates are in an amorphous state, it has been thought simply that the amorphous silicates gradually crystallized via thermal annealing in the hot inner solar nebula over time and then were transported outward and incorporated into comets (e.g., Shu et al. 1996; Boss 2004). Using this thermal annealing model, chronologies such as the formation ages of comets in the early solar system have been discussed based on laboratory experiments (Nuth et al. 2000). Thus interstellar composition ices, which remained in the outer region, have to be incorporated into comets simultaneously, because ices are easily sublimated during the annealing of silicates at high temperatures (800–1000 K) in the hot inner solar nebula. Recently, a new crystallization route for silicates was proposed, which does not lose volatile interstellar ices from amorphous silicate surfaces before the formation of comets (Yamamoto & Chigai 2005; Yamamoto et al. 2007). Namely, comet nuclei might still contain interstellar ices and silicates. Their model was based on the Greenberg model of cometary dust, composed of a silicate core, an organic mantle, and an outer icy mantle. When the dust is released as a comet approaches the Sun, the icy mantle sublimates quickly and the remaining organic mantle and silicate core are heated to temperatures of several hundred K depending on the heliocentric distance. Although it has been reported that temperatures of several hundred K are insufficient to crystallize amorphous silicate (Hallenbeck et al. 1998), the energy released by the alteration of an organic mantle layer can heat the surface of the silicate core and even flow into the core. This leads to the concentric crystallization of the amorphous silicate core from the surface toward the interior. In this study, we experimentally demonstrate the nonthermal crystallization model using laboratory synthesized amorphous 1
2. EXPERIMENTAL PROCEDURE
In previous studies, we succeeded in forming both amorphous Mg-bearing silicate and crystalline forsterite (Mg2SiO4) grains by the coalescence growth of MgO and SiO2 and/or Mg and SiO grains in a mixed smoke (Kamitsuji et al. 2005b; Kaito et al. 2003). During the thermal annealing process, the crystallization of Mg-bearing silicate grains into forsterite (Mg2SiO4) was directly observed at 800⬚C starting with amorphous Mg-bearing silicate grains in a transmission electron microscope (TEM) (Kamitsuji et al. 2005a). The crystallization initially started at the grain surface and formed a rimlike concentric layer due to prenucleation at 650⬚C–700⬚C. The crystallization process stopped for a long time, which may correspond to the stall state (Hallenbeck et al. 2000; Rietmeijer et al. 1999). The prenucleated crystallites subsequently grew at a significantly higher heating temperature. When an amorphous carbon layer was deposited onto the amorphous Mg-bearing silicate grains, the crystallization temperature locally decreases to 600⬚C due to graphitization of the carbon layer on the grain surface (Kaito et al. 2007). To further decrease the crystallization temperature, amorphous Mg-bearing silicate grains with a carbonaceous layer consisting of amorphous carbon, CH4 and other organics were prepared in our laboratory. The detailed experimental procedure is shown in Figure 1. To observe the deposited structure of the sample before exposure to air, the sample was transferred to the TEM using a specially made transfer holder and observed immediately. After that, the sample was exposed to air and the same position was observed again. 3. RESULTS AND DISCUSSION
Figure 2a shows the amorphous Mg-bearing silicate grains covered with a carbonaceous layer produced in a CH4 gas atmosphere. The sample was transferred to the TEM without exposure in air. The deposited layer exhibited a concave-convex
Corresponding author;
[email protected].
L57
L58
KAITO ET AL.
Fig. 1.—Schematic image of the sample, and photos of the tip of the special specimen transfer holder for the TEM. Amorphous Mg-bearing silicate grains mounted on an amorphous-carbon film or on an amorphous-carbon holey film supported by a standard Cu TEM grid was placed on the hole of the transfer holder, as indicated by an arrow. The holder was inserted into the chamber. After evacuation of the chamber to as low as 10⫺6 torr, CH4 gas was introduced up to approximately 10⫺3 torr. Carbon was then evaporated from an electrically heated carbon rod onto the specimen; i.e., a carbonaceous layer was formed on the silicate grains. In addition to CH4 molecules, the carbonaceous layer may also include breakdown products of CH4 molecules, because CH4 gas will decompose due to the high temperature of the evaporation source. Finally, the prepared specimen was transferred under a 200 torr CH4 gas atmosphere, which was introduced to suppress the leak rate of the transfer holder, to the Hitachi H-9000NAR TEM with the closed tip of the transfer holder, as shown in the bottom image.
structure on the spherical amorphous silicate surface. When the carbon layer was deposited on the amorphous silicate grains by vacuum evaporation, the grain surface was covered with a smooth carbon layer (Kaito et al. 2007). In the present case, the CH4 adsorbed on the amorphous silicate grains after introducing the CH4 gas may prevent the migration of the deposited carbon on the grain surface. When the specimen was exposed to air, the deposited surface layer began to react.
Vol. 666
Fig. 2.—(a) Typical TEM image of -prepared carbon-coated Mg-bearing silicate grains, which were kept from contact with air after sample preparation by using a special transfer holder. (b) After exposure to air for 11 minutes, the carbonaceous layer was observed to be smoother. In order to prevent crystallization caused by the electron beam during TEM observation, the intensity of the electron beam was controlled to less than 5–6 A m⫺2, which is the standard observation condition for these specimens. We verified that samples did not crystallize under the electron beam under this electron current.
After 11 minutes, the carbon layer became smoother, as shown in Figure 2b. Since the alteration of the carbonaceous layer does not occur in vacuum, we believe that the oxygen in air acts on the adsorbed hydrogen and/or carbon. A high-resolution TEM image of the surface is shown in Figure 3. The surface carbonaceous layer transformed to a graphitic structure accompanying a turbographene structure. The Fourier transform image of this region clearly showed the graphitization. The chemical reaction energy in the reactions among oxygen, water vapor in air, and hydrogen and/or carbon in the carbonaceous layer may induce graphitization of the carbonaceous layer. The graphitization energy is probably due to the oxidation of the deposited carbonaceous layer. During graphitization, it was ob-
No. 1, 2007
NONTHERMAL CRYSTALLIZATION
L59
Fig. 3.—High-resolution TEM images of a silicate grain after exposure to air for 11 minutes, i.e., corresponding to the specimen shown in Fig. 2b. Growth of forsterite nanocrystallites was observed to accompany the graphitization of the carbonaceous layer.
served that nanocrystallites of Mg2SiO4 of ∼10 nm in diameter grew on the silicate surface. Since the Mg2SiO4 crystallites grew concurrently with graphitization, we believe that the energy released by graphitization was conducted into the amorphous silicate surface and used to crystallize silicate grains. In fact, the amount of oxygen in a cometary coma would not actually be enough to oxidize the carbonaceous layer and produce the energy needed to crystallize amorphous silicate grains. One other possible driving mechanism of graphitization is solar radiation. Yamamoto and colleagues have discussed the crystallization of amorphous silicate due to the energy released by chemical reactions in the carbonaceous layer via solar radiation. After crystallization of the silicate surface, the crystallization energy was also conducted to the interior, but the energy was insufficient to drive further crystallization, unlike the case of crystallization by direct heating. In the case of our heating experiment (Kamitsuji et al. 2005a; Kaito et al. 2007), some crystallites produced on the surface layer acted as nuclei for further crystallization, similar to grain growth. Since the mean free path of the phonon was on the order of the crystallite size, the concentric crystallization rings appeared as dark contrasts, as indicated by arrows A and B in Figure 4. The sample shown in Figure 4 was the result of exposing an amorphous carbon film to air for 4200 minutes. A concentric black-contrast outer layer, as indicated by arrow A, which corresponds to the crystallization of silicate, would be formed initially using the reaction energy of graphitization in the carbonaceous layer. After that, the second layer with black contrast, indicated by arrow B, would appear as the result of the successive crystallization of silicate using the crystallization energy of the first layer. The crystallization energy from area A was also conducted to the central part of the particle. The excess energy due to the crystallization caused the formation of concentric crystals in B. Due to the volume alteration from amorphous silicate with a lower density to crystalline silicate with a higher density, it may also crystallize concentrically. The sizes of the present amorphous silicate grains and their carbonaceous surface layer were 50–100 nm in diameter and 10–30 nm in thickness, respectively. Therefore, the sample corresponds to the model size of Yamamoto et al. (2007), which is based on the interstellar dust model of Greenberg (1982). We estimated the crystallinity of our nonthermally crystallized silicate grains. In the case of the grains shown in Figure 4, the
Fig. 4.—TEM image of a silicate grain deposited on an amorphous carbon film after exposure to air for 11 minutes. The carbonaceous layer is slightly thicker than that of the silica particles located on a hole of the holey carbon film. As a result of uniform crystallization on the surface of the silicate grains, concentric dark contrasts are visible as indicated by arrows A and B. Please refer to the text for details.
silicate core and the carbonaceous layer are 50 nm in diameter and 10 nm in thickness. Since the thickness of the crystallization region is approximately 8.3 nm, the degree of crystallinity of our silicate grains is roughly 36%, which excellently corresponds to the value estimated by Yamamoto’s theory (45%) (Yamamoto et al. 2007). The degree of crystallinity is sufficient to explain the intensity of crystalline silicate features in the 8–13 mm spectra observed in comets. Therefore, our experimental results show that 300 K is a high enough temperature to obtain crystalline silicates even though thermal annealing requires temperatures in excess of 800 K. We also measured the mid-infrared spectra of these specimens. Unfortunately, significant changes in the mid-infrared silicate spectral features have not yet been observed. The crystallization of Mg-bearing silicate grains depends on their composition, i.e., Mg-rich silicate grains crystallize more easily (Miyazaki et al. 2007). Since our silicate grains have a range of compositions in magnesium and silicon, the number of crystallized grains is insufficient to show the crystalline silicate feature. However, since nonthermal crystallization should depend on both the composition of magnesium and silicon in the grain as well as on the gas atmosphere, we predict that further experiments could show a greater degree of crystallization of silicate grains via the nonthermal heating model. Here we propose that nonthermal crystallization should occur in the coma of comets and will yield crystalline silicates in addition to those produced in the early solar nebula.
We thank T. Yamamoto and H. Kimura for useful suggestions. This work is supported by MEXT and JSPS under a Grant-in-Aid for Scientific Research.
L60
KAITO ET AL.
Vol. 666
REFERENCES Boss, A. P. 2004, ApJ, 616, 1265 Greenberg, J. M. 1982, in Comets, ed. L. L. Wilkening (Tucson: Univ. Arizona Press), 131 Hallenbeck, S. L., Nuth, J. A., III, & Daukantas, P. L. 1998, Icarus, 131, 198 Hallenbeck, S. L., Nuth, J. A., III, & Nelson, R. N. 2000, ApJ, 535, 247 Hanner, M. S. 1999, Space Sci. Rev., 90, 99 Kaito, C., et al. 2003, Meteoritics Planet. Sci., 38, 49 ———. 2007, Advances in Geosciences, in press Kamitsuji, K., Sato, T., Suzuki, H., & Kaito, C. 2005a, A&A, 436, 165
Kamitsuji, K., Suzuki, H., Kimura, Y., Sato, T., Saito, Y., & Kaito, C. 2005b, A&A, 429, 205 Miyazaki, Y., Kumamoto, A., Kimura, Y., Saito, M., Kurumada, M., Suzuki, H., Koike, C., & Kaito, C. 2007, Meteoritics Planet. Sci., submitted Nuth, J. A., III, Hill, H. G. M., & Kletetschka, G. 2000, Nature, 406, 275 Rietmeijer, F. J. M., Nuth, J. A., III, & Karner, J. M. 1999, ApJ, 527, 395 Shu, F. H., Shang, H., & Lee, T. 1996, Science, 271, 1545 Yamamoto, T., & Chigai, T. 2005, Highlights Astron., 13, 522 Yamamoto, T., Chigai, T., & Kimura, H. 2007, ApJ, submitted
