An experimental investigation of the internal methane pressure in hydrogen attack
- PDF / 801,435 Bytes
- 9 Pages / 603.28 x 788 pts Page_size
- 0 Downloads / 154 Views
I.
INTRODUCTION
R A P I D deterioration of the mechanical properties of steels may occur upon exposure to high pressure (>5 MPa) high temperature (200 to 600 ~ hydrogen, after an "incubation period" of negligible change. The deterioration is due to the formation of methane from the reaction of diffusing hydrogen and carbon from the cementite or matrix. The methane accumulates in grain boundary and other bubbles, which then grow under the influence of an increasing intemal pressure. Bubble growth and coalescence lead to degradation of all mechanical properties and eventual failure. Successful predictions of the incubation time ti and other important parameters, e.g., attack rate and total void volume, are a major target of theoretical models. In addition to providing a fundamental rationale for the empirical Nelson diagrams l in current use for design of vessels to contain hot, high pressure hydrogen, such models may be of value in predicting the behavior of new alloys in unconventional hydrogen environments. The modeling studies presently available2-5 have encountered difficulties because of the lack of hard values of some of the important input parameters needed in the mathematical formulations. Among these parameters is P, the methane pressure in the bubbles and voids, which is considered the main driving force for their growth. Shewmon's 2 model emphasized a diffusion controlled growth mechanism for bubbles in the critical incubation period. In that model P was assumed constant, with equilibrium values, Peq, which later were shoWn to be too high. Sagues et al 3 and Shih and Johnson 4 developed dual-mechanism creep and diffusion models, which predicted diffusion controlled bubble growth only in the very first stages, and creep as the controlling mechanism in later stages. Early diffusion controlled bubble growth gained additional support in the MENACHEM NATAN, Scientist, is with Martin Marietta Laboratories, 1450 South Rolling Road, Baltimore, MD 21227. HERBERT H. JOHNSON, Professor, is with Materials Science and Engineering Department, Comell University, Ithaca, NY 14853. Manuscript submitted November 16, 1981.
METALLURGICALTRANSACTIONS A
experimental work of McKimpson and Shewmon, 5 who, however, agreed that a creep mechanism takes over and controis bubble growth beyond the incubation period. Recent work by Oldfield et al 6 and Odette and Vagarali7 provide quite accurate estimates for Peq, which have already been used in calculations by McKimpson and Shewmon. 5 In this paper we present a first experimental study of the internal methane pressure in HA bubbles. Of course, the pressure cannot be determined inside a single bubble; rather, we considered a macroscopic or average pressure, and related it through simplifying assumptions to the single bubble case. The average methane pressure is determined from measurements of total bubble volume and methane content via an appropriate equation of state. It will be rigorously equal to the single bubble pressure if all bubbles nucleate at the same time, i.e., no tempe
Data Loading...
