[en] The effect of welding on the kinetics of hydrogen attack of 2 1/4 Cr-1 Mo quenched and tempered steel was studied using a highly sensitive capacitance dilatometer, in the temperature range of 490 C to 590 C and hydrogen pressures of 10 to 23 MPa. The strain rate of the weld metal was an order of magnitude greater than that of the base metal at 500 C and 20.3 MPa of hydrogen, but it was the same as that of the base metal at 570 C. The base metal exhibited an activation energy of 256 +/- 5 KJ/mol and a methane pressure dependence of 1.76 +/- 0.4. The weld metal had an activation energy of 313 +/- 36 KJ/mol and a methane pressure dependence of 6.6 +/- 1.5. The bubbles in the base metal formed preferentially at the grain boundaries, but those in the weld metal showed no such preference. The effects of tempering on the hydrogen attack kinetics was studied by measuring strain rates and carbon activities of 2 1/4 Cr-1 Mo steel samples tempered to different extents. Both carbon activity and the hydrogen attack strain rate decreased monotonically with tempering. Tempering for up to 500 hrs at 700 C does not decrease carbon activity below 0.05 and appreciable strain rates were measured at 550 C and 20.3 MPa. Bubble growth mechanism maps were drawn for both carbon and low alloy steels, and the maps give the predicted mechanisms controlling bubble growth at various temperatures and hydrogen pressures. Commercial 2 1/4 Cr-1 Mo Q and T steel was found to suffer hydrogen attack by grain boundary diffusion up to about 25 MPa hydrogen. The weld metal however was found to suffer hydrogen attack by the growth of bubbles by power-law creep of the matrix

[en] It has been realized dor over 30 years that at temperatures of less than roughly two-thirds the absolute melting temperature, diffusion down grain boundaries and dislocations plays a dominant role in determining the rate of interdiffusion in alloys. It now appears that the diffusion induced motion of existing grain boundaries (DIGM) and the nucleation of new grains, and the diffusion induced motion of these new boundaries, may often be the dominant mixing process in many systems. That is, it is the sweeping of the high diffusivity boundaries through the lattice that provides the solute penetration instead of diffusion down stationary boundaries and from there through the lattice, as has been assumed in all attempts to quantitatively describe the kinetics of grain boundary diffusion processes. The purpose of this note reports the rapid transport of iron into pure Ni at 550 ⁰C when the two samples are not in physical contact but both occupy a vycor tube evacuated before sealing. The novelty arises not only from the mode of vapor transport of iron, but from the copious nucleation of new grains on the surface as well as the active role of DIGM in the penetration of Fe into Ni

[en] The effects of Y₂O₃ solute concentration, strain rate, and temperature on solid-solution strengthening in single crystal yttria-stabilized cubic zirconia was investigated. Previous work was extended by studying the flow behavior at strain rates from 1.5 x 10^-5 s^-1 to 8 x 10^-8 s^-1 at 1,200, 1,300 and 1,400 C in the harder <001> orientation. Solute hardening in this system is sensitive to strain rate down to 8 x 10^-8s^-1, but at 1,400 c there was no difference in flow stress between a 9.4 mol% alloy and 21 mol% alloy at a strain rate of 8 x 10^-8 s^-1. The results were explained by the solute drag model

[en] In recent years there has been a recognition of the potential of structural ceramics for use in advanced heat engines and heat exchangers. Compared to metals, ceramics can withstand higher operating temperatures, have lower density, superior wear resistance and chemical stability. Unfortunately, ceramics also have low fracture toughness, resulting in brittle failure. One method of providing increased fracture toughness is by the incorporation of ceramics in composite structures. In this paper the authors report on the elevated temperature mechanical behavior of a fine-grained two-phase material consisting of a YAG matrix with about 20 volume percent of a YAP second phase at temperatures between 1540 to 1680 degrees C

[en] When either uncoated or BN-coated Nicalon fibers are exposed to water saturated with NaCl before being annealed in air at 1,000 C, accelerated degradation of the structure of the fiber occurs. The fiber surface oxidizes to tridymite instead of vitreous silica, and the crystallites of SiC in Nicalon begin to grow. These findings suggest that prior exposure to salt water may cause appreciable debit in the mechanical strength of nicalon fibers with time at 1,000 C in air

[en] Advanced intermetallic materials, such as refractory silicides, exhibit high melting points, high stiffness, low densities, and good strength retention at elevated temperatures. Further, some of these silicides are in equilibrium with terminal refractory solid solution (beta) phases, and therefore, offer the potential for ductile phase toughening. Studies were conducted to elucidate the compressive creep behavior of monolithic Nb₅Si₃ and to generate the constitutive creep law. This, in turn, is required for modeling the creep behavior of the Nb/Nb₅Si₃ two-phase system. Nb₅Si₃ has the ordered tetragonal structure with 32 atoms/cell in both its allotropic forms: αNb₅Si₃ (D8_l Cr₅Si₃-type; a ∼ 0.656 nm; c = 1.187 nm) and βNb₅Si₃ (D8_m W₅Si₃-type; a = 1.000 nm; c = 0.507 nm). αNb₅Si₃ is stable below 1,935 C, while βNb₅Si₃ is stable above 1,645 C. The large lattice parameters as well as the large number of atoms in the unit cell suggest that dislocation creep is unlikely to occur in Nb₅Si₃, because large Burgers vectors and complex dislocation core structures are expected in this material

[en] In this study we present a new constitutive model for Ni3Al and Ni3(Al, X) alloys that was developed to represent many of the unusual plastic flow behavior found in L12 intermetallics while maintaining consistency with the experimentally-observed evolution of dislocation substructure. In particular, we sought to develop a model that would not only predict the anomalous increase of the yield strength with increasing temperature, but would also capture other important flow characteristics such as extremely high work-hardening rates that change anomalously with temperature, and a flow stress that is partially to fully reversible with temperature. For this model, we have treated work-hardening as arising from two different sources. Thermally-reversible work hardening is accounted for using the description of screw dislocation motion proposed by Caillard, which involves exhaustion of mobile dislocations by cross-slip locking of the dislocation core and athermal unlocking. Thermally-irreversible work hardening is accounted for using an approach consistent with the theoretical framework proposed by Ezz and Hirsch, which involves both the multiplication of Frank-Reed sources and the interaction of edge-dislocation segments with cross-slip locking events and the dislocation forest. Both work-hardening contributions were incorporated into the rate formulation for thermally-activated plastic flow proposed by Kocks, Argon and Ashby. We will show simulation results for the flow response of Ni3(Al, X) crystals over a wide range of temperatures in the anomalous flow regime, and we will compare these findings with experimental data

[en] The effect of 33.5 vol% SiC whisker loading on high-temperature deformation of 1 wt% MgO-38.5 wt% zirconia-mullite composites was studied between 1,300 and 1,400 C. At strain rates of 10^-6 to 5 x 10^-4/s the creep resistance of zirconia-mullite composites without SiC reinforcement was inferior to monolithic mullite of similar grain size. Analysis of the results suggested that the decreased creep resistance of mullite-zirconia composites compared to pure mullite could be at least partially explained by mechanical effects of the weaker zirconia phase, increased effective diffusivity of mullite by zirconia addition, and to the differences in mullite grain morphology. With SiC whisker reinforcement, the deformation rate at high stress was nearly the same as that of the unreinforced material, but at low stress the creep rates of the SiC-reinforced material were significantly lowered. The stress dependence of the creep rate of unreinforced material suggested that diffusional creep was the operative mechanism, while the reinforced material behaved as if a threshold stress for creep existed. The threshold stress could be rationalized based on a whisker network model. This was supported by data on other whisker-containing materials; however, the threshold stress had a temperature dependence that was orders of magnitude higher than the elastic constants, leaving the physical model incomplete. The effects of residual stresses and amorphous phases at whisker/matrix interfaces are invoked to help complete the physical model for creep threshold stress

[en] Physics-based models for predicting the mechanical behavior of Ni-based superalloys as a function of microstructure features require the use of microstructure data for calibration and verification. Accurate representation of the heterogeneity of microstructure features requires accurate selection of the representative microstructure data size (i.e. image size). Thus, this work is carried out to address the influence of microstructure data size on the accuracy of a discrete dislocation dynamic model in predicting the critical resolved share stress (CRSS) of IN100. Microstructure features from backscattered electron images were extracted using image processing techniques. Single point statistics (e.g. area fraction, precipitate size, and distance between γ' particles) and higher order statistics using two-point correlations were calculated from segmented 2-D images. Modified Bhattacharyya Coefficient analysis techniques were employed to calculate three-dimensional particle size distributions. Results indicate a significant influence of the microstructure data size on the calculated CRSS.

[en] Large scale, atomistic simulations of the core structure and mobility of ½[111] screw, edge and mixed dislocations in ternary multicomponent alloys (e.g. High Entropy alloys), NbTiZr, Nb_1.5TiZr_0.5 and Nb_0.5TiZr_1.5, are presented. The core structure of ½[111] screw dislocations continuously varies from compact to 3-fold with decreasing Nb content. The screw dislocation core structures in NbTiZr and Nb_1.5TiZr_0.5 are calculated using Embedded Atom Potentials (Johnson-Zhou) and compared with first-principles calculations of the screw dislocation in a quasi-random structure. In both simulations the dislocation core spreads on different (110) glide planes as the composition varies along the dislocation line in stoichiometric NbTiZr. The Nb-rich composition Nb_1.5TiZr_0.5 shows a compact core with very little core structure variation along the dislocation line in both First Principles and atomistic simulations. The screw dislocation deposits interstitial and vacancy dipole debris as it moves under stress. Average solute-dislocation core interaction energies in NbTiZr, Nb_1.5TiZr_0.5 and Nb_0.5TiZr_1.5 are derived from the average interatomic potential derived for each of the three systems. The interaction energies are used to determine the critical stress for the motion of ½[111] screw dislocations in the three systems as a function of temperature using the Suzuki model of kink migration controlled mobility developed for concentrated BCC random alloys. This analysis shows that the relatively high barrier for kink migration caused by fluctuations in solute concentration along the screw dislocation line and the dipole dragging stress associated with the screw dislocation motion results in a shallow fall-off of critical stress with temperature in these alloys as compared to simple BCC metals. Finally, the screw dislocation to edge and mixed dislocation critical stress ratio in NbTiZr are shown to be low, ∼2 at 5 K, in contrast to simple BCC metals, where it could be as high as 100–1000.