Abstract : Consideration is given to the shock initiation and propagation of detonation in granular or heterogeneous solids. Reactions in initiating shocks were studied by measuring pressure and specific volume in the shocked state (2 to 100 kbar) for the physically similar granular reactive and nonreactive aggregates ammonium perchlorate (a low-power, insensitive explosive) and potassium chloride (an inert comparison material). Longitudinal sound velocities in compacts of the same material, over a range of bulk densities and grain sizes, were also measured. Loci of shocked states for NH4ClO4 and KCl aggregates show a dramatic qualitative difference, leading us to the conclusion that reaction occurs at the shock front. This conclusion opens the possibilities that nonreactive shocks in granular explosives and propellants do not exist, that the amount of initial reaction is related to the shock pressure, and that mechanisms of buildup to steady detonation must take into account reaction from time of shock entry. Finally, a dependence of transition behavior on grain size was demonstrated; the faster transition from shock to detonation occurs for smaller grain size, i.e., larger grain surface per unit volume.
Plane constant-pressure shock waves were used to initiate wedge-shaped pressings of PETN (pentaerythritol tetranitrate). The shocks entered the PETN from brass or Lucite plates. Shock pressures in the plates and depths at which the PETN was initiated were measured with a streak camera. It was found that a 50-kbar shock in the brass was barely sufficient to initiate granular PETN pressed to a density of 1.0 g/cm3. This corresponded to a derived pressure of about 2½ kbar in the PETN pressing. It was further shown that the interstitial gas had no effect on the depth of initiation.
Streak-camera observations have shown that wall interactions are frequently responsible for shock initiation of nitromethane in gap-test geometry. Therefore, pressure calibrations of the gap test for liquids made at the attenuator surface along the charge axis should not be used to interpret the test results. Uncritical acceptance of gap-test values for a series of liquids with different properties may lead to erroneous sensitivity ordering. On the other hand, modified gap-test experiments can be used to obtain fundamental initiation data if appropriate instrumentation shows that wall effects are absent. A valid series of reaction time measurements is presented for nitromethane. Experimentally valid initiation data must be interpreted in terms of divergent reactive flow. G AP tests 1 for determining the sensitivities of liquids to shock initiation of detonation have excellent reproducibility, and, moreover, they are simple and economical. The excellence of the reproducibility suggests that it is worthwhile to interpret gap-test data in terms of shock pressure of initiation. Shock pressures, in principle, can be utilized to compute shock temperatures if additional thermodynamic data are available25; temperatures, in turn, allow one to discuss chemical reaction rates in initiating shocks. Such information, rapidly available from an experiment as simple as the gap test, would be very useful in studies of the mechanism of shock initiation and ultimately in studying the relation between shock sensitivity and chemical structure. However, there is reason to suspect that a fundamental interpretation of gap-test data can be made only if detailed three-dimensional processes occurring in this somewhat complicated geometry are completely understood. Sensitivity data from gap-test experiments are often reported in terms of the number of cards (usually cellulose acetate sheets 0.010 in. thick) making up the attenuator. The number of cards or the total thickness of plastic is reported as the 50% failure thickness. Thickness is converted to pressure in the attenuator by measuring the shock and free-surface velocities along the axis of the donor-attenuator system, or by measuring only the shock velocity if the Hugoniot of the attenuator material is already known. With some difficulty, attenuator pressure may be measured directly with a gage. 6'7 The entering pressure in the liquid can be calculated from the attenuator pressure by solving the interface equations if the shock velocity in the unreacted liquid is known. The experimental measurements, on which the entering shock pressure depends, are made at the center of the attenuator; thus, it is implicitly assumed that whether the liquid detonates is decided along the charge axis. In this note, experiments are reported which show that conditions on the charge axis do not necessarily control initiation; on the contrary, when the initial shock is too weak to cause detonation directly, initiation occurs at the container walls. The assumption that the axial peak pressure entering the liquid initiates a homogeneous thermal explosion leads to confusing inconsistencies that preclude fundamental interpretation, and in some cases, result in erroneous sensitivity ordering of liquid explosives. The liquid explosive used in these experiments was technical grade nitromethane (NM) manufactured by Commercial
Abstract : Iniation and detonation behavior of 13-micron ammonium perchlorate was studied at loading density 1.00 g/cu. cm. Steady detonation velocities were determined experimentally at three diameters and extrapolated to 3.75 = 0.15 mm/ micro sec at infinite diameter. Calculations with the BKW equation of state gave 4.25 mm/micro sec--as good agreement as could be expected for a low-energy chlorine--containing explosive. By introducing 24-kbar flat-topped plane shocks into pellets of various lengths, it was determined that steady full-strength detonation was reached after about 15 mm travel. The growth of pressure in the accelerating wave was followed approximately by means of free-surface measurements on thin Plexiglas at the top surfaces of the pellets, and these measurements indicated the pressure to be 55 = 10 kbar in the full-strength wave. Reducing the air pressure in the pressings to 5 microns left the build-up to detonation unaffected. The position of the shock Hugoniot for the pressings relative to the Hugoniot of the solid crystal is discussed in terms of heat production during collapse and possible reaction processes. (Author)
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)