Explosive Limits Model Predictions

The most extensive test which has been made of this conduction model for thermal explosion is to be found in the work of Vanp e on the explosion of CH2O + O2 mixtures. He used a calibrated thread of 10 per cent Rh-Pt alloy of 20 m diameter (jacketed by a 50-m quartz sleeve) suspended at the center of a cylindrical vessel to measure directly his reaction temperature during the induction periods preceding explosion. By Uvsing He and Ar as additives and vessels of different diameters he was able to verify the dependence of the critical explosion limits on vessel size and on thermal conductivity of the gas mixture. In addition, he was able to check the maximum predicted temperature at the center of the vessel just prior to explosion and also the value of 8c = 2 [Eq. (XIV.3.12)], the critical explosion parameter for cylindrical vessels. Finally, with a high-speed camera, he was able to show directly that the explosions in this system do start at the center, the hottest region, " and propagate to the walls. 

Figure 11. Comparison of predicted explosion limits with experimental data experimental explosion limits [from Lewis and von Elbe (1961)] — model prediction, [from Wu et aJ. (1990)].

The isothermal model, constituted by eqs. (16) and (18) alone, does not predict the occurrence of the third explosion limit. In particular, when applying the generalized criterion to this model, it is found that the predicted values of the second explosion limit increase continuously as the initial pressure increases. For values larger than about 100 Torr the curves representing the normalized objective sensitivities become flatter and the maximum tends to vanish. Using the non-isothermal model, given by eqs. (15) to (17), the explosion limits predicted by the... 

High pressure explosive loading was carried out on both z- and y-cut crystals at pressures between about 25 and 60 GPa ([83S01, 77S01]). The z-cut crystals responded in the plus-x orientation with current pulse wave shapes as predicted by the three-zone model. Nevertheless, limited experiments in the minus-z orientation of lithium niobate do not show the positive currents expected from the three-zone model. 

Figure 1 Observed photon luminosity vs. time during a year after the supernova explosion predicted from standard cooling theory of neutron stars. Shown are three representative nuclear models PS (dashed), FP (solid), and BPS (dot-dashed). The detection limit from Ginga is shown as a horizontal line.

The second aspect is to provide the researcher with a practical and simple methodology for predicting the dynamics loads on mechanical structures. From numerous experiments and within the limits of scalability of the explosions, the data are correlated and fitted by least-squares polynomials. In both cases, the incident blast wave (i) and the pressure loading (r) on plane surfaces, the pressure signals are modeled. The damping coefficients ki and... 

The requirement that the reaction not be explosive is a very subtle one. Essentially it requires that the chain-carrier growth be self-limiting in some way. This constraint is related to the dependence of the NO removal on H2 addition [4-7], water concentration [4-7], and O2 level (this work) in that, when the reaction is explosive, these other effects also are not predicted accurately by the model. When NIffl simply dissociates rapidly, i.e. 

Further stochastic simulation studies now in progress are concerned with fluctuation and nucleation in evolving chemical systems (e.g., limit cycle oscillations, combustion and explosions) and at the transition to spatial dissipative structure (cf.. Figure 10). In the latter case, for example, stochastic simulations verify the existence of critical long-range spatial correlations predicted in a stochastic theoretical study of the model (cf.. Ref. 17). 

---