Drying kinetics is usually observed on a volume-averaged basis, i.e. the mass loss of a wet material as function of time is recorded, without explicitly measuring the internal distribution of the liquid phase within the porous medium being dried. Although in some cases the moisture distribution profiles can be determined a posteriori by taking samples of a material from different depths before drying is complete, or by inserting probes into the material, these methods are generally “destructive” in the sense that they irreversibly modify the microstructure of the system. Moreover, they are applicable to bulk quantities of mechanically robust materials (e.g. a layer of sand) but not to small amounts of fragile particles, such as a sample of a new pharmaceutical ingredient whose drying kinetics needs to be determined. The objective of this work is to demonstrate the use of x-ray computed micro-tomography (XCMT) as a non-invasive technique for the direct observation of evolution of moisture profiles in a sample of a porous material (filter cake) during contact drying. By the measuring of the 3D distribution of x-ray attenuation, which is proportional to the local density, it is possible to identify the solid, liquid and gas phases (see Fig. 1) in cylindrical vacuum contact dryer (heat is applied through the walls and the head-space above the particle layer is connected to a vacuum source).

Simultaneously, the development of the moisture content profiles during contact drying has been calculated. Liquid evaporation and capillary flow are determined by solving material and energy balances combined with the Volume-of-Fluid (VOF) method, which describes free interface (liquid menisci) propagation in a porous medium controlled by the requirements on the contact angle and the mean interface curvature. Output from the simulations is detailed distribution of the liquid phase within the porous medium as a function of time.

The temporally resolved 3D moisture distribution profiles obtained experimentally by XCMT for two different crystal particle assemblies under different drying regimes have then been directly compared with moisture distribution profiles generated by computer simulation in a computer-reconstructed porous medium having the same microstructure as the physical one. These studies can be used for the investigation of the effect of porous microstructure on drying kinetics.

Figure 1: Illustration of tomography visualization technique of packed bed of crystals partially wetted by solvent: the cross-sections of the attenuation distribution obtained by XCMT (left) and 3D reconstruction (right).