Some effects of addition of inert suspended solids (fine glass particles) to both reacting (water and acetic anhydride) and nonreacting systems (pure water or water/glycerol mixtures) are described. Depressurization profiles from experiments with solids present were compared with analogous profiles without solids. Experimentation was on scales of one and ten liters. Computer control meant that highly reproducible experimental sequences were followed. Mean nozzle to particle diameter ratios of between 6 and 500 were used. The particles were narrow size-range cuts of nearly spherical glass beads with a density of 2500 kg m−3. Hollow glass beads with a density of 600 kg m−3 were also used. Glass particle concentrations were up to 16% v/v and relief set pressures were between 3 and 5 bara.
Many variables affect the pressure vs. time profile during blowdown, and thus experimental (factorial) design techniques were used for efficient planning of experiments and interpretation of results. In general the glass had little statistically significant effect. There was limited evidence that the hollow glass particles could slightly increase depressurization rates. Tests with runaway reactions highlighted some inherent difficulties when comparing systems with and without glass particles present. In all cases, the vented fluids were less concentrated in glass than those in the reactor. The consistency of the amounts of liquid carryover during blowdown was poor in smaller-scale tests; this may have important ramifications for small-scale tests that are used to establish the flow regime during venting, as advocated in the DIERS methodology. © 2006 Health and Safety Executive. Crown Copyright. Process Saf Prog, 2006