No abstract

A new method of measuring the properties of hollow fibre membranes for modelling purposes was developed. Measuring the gas permeability as a function of membrane length and extrapolating to zero length provided enough membrane material information for modelling the flux. These values resulted in the predicted Vacuum Membrane Distillation (VMD) flux within the experimental errors (±5%) for different velocities. The predicted results at different temperatures were mostly within the experimental variation range. However, the error (<10%) was greater at the highest temperature tested, perhaps due to water evaporation in the feed tank. Additionally both the modelling and experimental results show that the flux in VMD was independent of the module packing density under the presented conditions. The results also show that the mathematical model predictions agree well with the experimental results of short experimental duration (1 -2 h).

No abstract

Six modules with different packing densities and lengths were fabricated for conducting vacuum membrane distillation (VMD) experiments. The performances of the modules with different packing densities showed similar results for flux at different temperature and feed velocities. However, shorter modules show both higher flux and global mass transfer coefficient than that of longer modules. It was also found that the increased flux at higher velocity was largely due to the increased average feed temperature rather than reduced temperature polarisation. The flux decay in this study was considered to be caused by gradual compression of the membrane which was confirmed by a compression test. This phenomenon should be noted in the study of VMD, since the pressure difference across the membrane is in general more than 100 kPa and the porosity of the membrane is in general high which results in reduced mechanical strength.