• Korean Membrane Journal
• /
• v.8 no.1
• /
• pp.1-12
• /
• 2006

In semiconductor industries, dissolved oxygen is one of the most undesirable contaminants of ultrapure water. A method for dissolved oxygen removal (DOR) consists in the use of polymeric hollow fibres, loaded with a catalyst and fed with a reducing agent such as hydrogen. In this work, PVDF hollow fibres loaded with Pd were characterized by means of perporometry, scanning electron microscopy (SEM), energy dispersive X-ray (EDX). The hollow fibre analyzed shows a five-layer structure with remarkable morphological differences. An estimation of pore diameters and their distribution was performed giving a mean pore diameter of 100 nm. The permeance and selectivity of the fibres were measured using $H_2,\;N_2,\;O_2$ as single gases, at different operating conditions. An $H_2$ permeance of $37 mmol/m^2s$ was measured and $H_2/O_2$ and $H_2/N_2$ selectivities of ca. 3 were obtained. $H_2$ permeance was 1/3 when a water stream flows in the shell side. Catalytic fibrebehaviour was simulated using a mathematical model for a loop membrane reactor, considering only $O_2$ and $H_2$ diffusive transport inside the membrane and their catalytic reaction. Dimensionless parameters such as the Thiele modulus are employed to describe the system behaviour. The model agrees well with the experimental reaction data.

• Membrane and Water Treatment
• /
• v.2 no.4
• /
• pp.251-266
• /
• 2011

This work deals with the use of High Resolution Scanning Electron Microscopy (HRSEM) to verify ultrafiltration membrane selectivity at the end of the production line as well as membrane ageing. The first part of this work is focused on new membranes. It is shown that it is better to use sputtering metallization than vacuum deposition, as this latter technique entails thermal damage to the skin layer. Moreover, the impact of the metallization layer on the determination of the membrane pore size is studied and it is observed that no impact of the metallization step can be clearly defined for a metallization layer ranging from 3 to 12 nm. For example, an average pore size of 16.9 nm and a recovery rate of 6.5 % are observed for a 150 kDa cellulose acetate membrane. These results are in agreement with those given by the manufacturer: pore size ranging from 10 to 15 nm and recovery rate ranging from 5 to 10 %. The second part of this work focuses on the study of membrane ageing. A PVDF hollow fibre membrane is studied. It is shown that a 65 % decrease in the permeate flux can be linked to a decrease in the number of pores at the surface of the membrane and a decrease in the recovery rate. In conclusion, a mapping of the pores is performed for several new hollow fibre membranes used to produce drinking water, made of different materials, with different geometries and molecular weight cut-off. These results provide reference data that will help better understand the phenomena of membrane fouling and membrane ageing.