Vaporization Rate of Forsterite in Hydrogen Gas

Forsterite is one of the most prominent minerals in the solar system and its vaporization behavior should be studied in detail for the understanding of the Mg-Si and isotopic fractionation. Hashimoto [1] and Wang et al. [2] measured the evaporation rate of forsterite in vacuum (continuous evacuation), which is expected to be much smaller than that in the presence of hydrogen gas from the study of SiO2 vaporization in vacuum and hydrogen gas [3,4]. Vaporization rate of forsterite was determined in a wide range of hydrogen gas pressure and the results are reported here with an application to vaporization processes in the solar nebula. In this report, the term "vaporization" is used for transformation of solid to vapor in general, and "evaporation" is used strictly for free evaporation typically realized in the Langmuir experiment. About 30 mg of synthesized single crystals of forsterite was used for each experiment. Experimental temperature was 1700 degrees C and the hydrogen gas pressure ranged from 10^-7 to 3 x 10^-5 (?) bar. Because high pressures (>10^-5) cannot be achieved with turbomolecular pump, an rotary pump was used in two runs with a pressure correction. Vaporization rate per a unit area and a unit time was obtained by assuming that vaporization took place at a constant rate from every surface. The results are shown in Fig. 1. At pressures below 10^-6 bar, the vaporization rate does not depend on hydrogen pressure and is the same as that for vacuum, but at higher pressures it appears to be proportional to 1/2 power of P(sub)H2. On the basis of this dependence, Nagahara and Ozawa [5] considered that the reaction of forsterite with an adsorbed hydrogen atom on the crystal surface may control the vaporization at higher hydrogen pressures. The overall P(sub)H2 dependence, however, should be explained by summation of vaporization flux (in gram cm^-2S.ec^-1) by evaporation (J(sub)evap) and by reaction with hydrogen (^J(sub)react). At pressures below 10^-6 bar, J(sub)evap, which does not have P(sub)H2 dependence, dominates. At pressures above 10^-6 bar, contribution of J(sub)react, which depend on P(sub)H2, becomes predominant. The results of Fig. 1 are best fitted by the equation J(sub)total= J(sub)evap + J(sub)react = 8 x1 0^-7 + 0.27 P(sub)H2, indicating that J(sub)react linearly depends on P(sub)H2 with reaction constant of 0.27. According to our preliminary experiment s at higher hydrogen pressure, the relation can be extended at least up to 10^-4 bar. The linear dependence suggests that a plausible rate-conrolling reaction is Mg2SiO(sub)4 + H2 --> 2Mg + SiO + H2O + O2 which takes place on the surface of forsterite. The equation indicates that the reaction is breakdown of a forsterite molecule by an hydrogen molecule, where oxygen ions occupy the very surface, which easily reacts with hydrogen to form gas species. The results are applied to estimate the limit of residence time of forsterite dusts in the solar nebula. The time of total evaporation of a sphere of forsterite with radius of 0.1 micrometers to 1 cm was calculated (Fig. 2). Within the plausible range of hydrogen pressure in the solar nebula, forsterite vaporizes very quickly. Even a grain with 1 mm radius completely vaporizes in a few days at 1700 degrees C. If the activation energy for the reaction is similar to that of evaporation of forsterite determined by Hashimoto [1], the grain completely vaporizes in a month at 1600 degrees C. Because the experiments were made in molecular flow of hydrogen gas that is continuously evacuated, the estimated vaporization time gives probably the lower limit. References: [1] Hashimoto, A. (1990) Nature, 347, 53-55. [2] Wang et al. (1993) LPS XXIV, 1479-80. [3] Nagahara H. et al. (1993) Meteoritics, 28, 406-407. [4] Nagahara H. (1994) LPS XXV. [5] Nagahara and Ozawa (1994) NIPR Symp. (abstract). Figure 1, which appears in the hard copy, shows the vaporization flux of forsterite plotted against hydrogen (~total) pressure. Solid circles for experiments in hydrogen atmosphere with a turbomolecular pump and open squares for vacuum experiments are fitted by a curve shown. Open circles are for runs at higher hydrogen pressure evacuated by a rotary pump. Figure 2, which appears in the hard copy, shows the time required to completely vaporize a spherical grain of forsterite in hydrogen gas at 1700 degrees C.