An experimental apparatus for the investigation of fragmentation of magma by rapid decompression from temperatures up to 950 °C and pressures up to 20 MPa has been brought into operation. With this fragmentation bomb apparatus, cylindrical porous magmatic dome samples (diameter = 17 mm, length = 50 mm) are subjected to rapid decompression resulting from the rupture of a prepared membrane or diaphragm which separates the high-pressure and low-pressure sections of the apparatus. The samples are located in the high-pressure section of this shock tube-type device. The decompression rate is influenced by the gas pressure, the initial temperature and the distance from the membrane to the upper surface of the sample.
In order to investigate the effect of the decompression rate on the nature and products of magma fragmentation the distance from the membrane to the sample surface has been systematically varied so as to generate differing decompression rates at constant initial temperature, pressure and physical state of the sample. Computer simulation of the gas flow accompanying decompression in the high pressure section has been performed using KASIMIR software. The results of these simulations indicate that the decompression rate at the sample surface is sensitive to the sample surface to membrane distance. For the case of an Ar gas medium, initially at 12 MPa and 20 °C, distances of 12, 212 and 402 mm correspond to effective decompression rates at the sample surface during the first 0.15 ms of 80, 44 and 13 MPa/s, respectively.
Two types of modifications have been performed on the high pressure section in order to vary the decompression rate experienced at the sample surface. For low temperature experiments, the samples to be fragmented were fixed within brass tubes of varying lengths which are threaded into brass rods fixed to the base of the high pressure section. Experiments investigating this variable length have demonstrated a significant dependence of fragmentation on the sample position.
For high-temperature experiments we propose to vary the total length of the high-pressure section hydrothermal bomb tube by using a stainless steel extension adapter between the hot bomb and the membrane below the low-pressure section. This will allow the variation of decompression rates to lower values but will not provide the highest decompression rates that are accessible at low temperatures with very short distances. An adaption of the system to allow rapid displacement of the samples from the hot zone to the upper, proximal part of the high pressure section would solve the problem of achieving very high decompression rates at high temperatures.