The statistical approach to the thermal neutron radiography picture analysis is applied to evaluation of the experimental results obtained for drying of rectangular and cylindrical samples of granulated and rigid porous materials. We have shown that the time dependence of the standard deviation of image brightness reects the appearance and motion of the drying front observed during second period of the drying process. The results are discussed within a simple two-region model of the drying sample image.

Externally applied magnetic fields have been used to induce major disruptions in the Tokoloshe tokamak. The investigations have included ‘resonant’ and ‘non-resonant’ external field configurations, using any one of three different stellarator-like windings with ℓ = 1, 2 and 3. In cylindrical geometry, these coils produce magnetic islands with dominant modes m/n = 1/1, 2/1 and 3/1, respectively. In the case where the Mirnov oscillations in the plasma are weak, i.e. where the existing tearing mode islands are much smaller than the islands of the same helicity produced by external coils, a major disruption of the plasma occurs when the coil current exceeds a certain critical value. The time to disruption τd decreases with increasing coil current, consistent with a simple model in which a critical perturbing radial field must be exceeded for each coil to induce a disruption. The characteristic time to reach this critical field is similar for each of the three coils and is approximately the time needed for resistive field penetration to the q = 2 surface. No clear precursor is seen until a rapidly growing locked m = 2 mode becomes significant just before disruption. Simultaneous measurements of B̃θ and B̃r show that this is a tearing-like mode, with the locking orientation dependent on coil current direction. The growth time, τg = 200 μs, is essentially independent of τd, with τg << τd, typically. Comparisons with magnetic field geometry calculations show that the observed disruption limits are consistent with tearing mode destabilization precipitated by, and phase locked with, overlapping or nearly overlapping 2/1 and 3/1 coil islands, where the latter island also overlaps the limiter. This holds for each of the coils and for both ‘resonant’ and ‘non-resonant’ cases. The experimental results are also compared with the case of strong Mirnov oscillations. Here the Mirnov signal can be suppressed when the coil current is sufficiently high. This is clearly due to mode locking (∂/∂t → 0) rather than to mode suppression (B̃θ → 0).

The application of modern digital neutron radiography in the study of the drying process in porous media is demonstrated with two simple examples of cylindrical and rectangular samples made of different materials. The statistical method of image analysis is presented and the standard deviation of grey scale values (referred to brightness) is shown to be a useful measure of receding drying front emergence. The motion of the drying front was analyzed and found neither linear nor proportional to the square root of time and was delineated with nonlinear functions different for each sample shape.

No abstract

Samples of Loxodonta africana ivory were subjected to X-ray diffraction studies which showed that the hexagonal crystals have a strong preferred orientation with their basal planes perpendicular to the longitudinal axis of the tusk. Cylindrical samples were also machined from the tusk with cylinder axes along and perpendicular to the longitudinal axis of the tusk respectively. Three-point bending tests were performed on the cylindrical samples. The results show a clear anisotropy in the bending strength along the two directions. The sample with cylindrical axis longitudinally aligned shows less bending for the same applied force. The results verify a correlation between orientation of hydroxyapatite and, among other things, the mechanical strength. This may be related to the function of the tusk which with normal usage has to be able to sustain large bending forces in a direction normal to the longitudinal axis.