Abstract
Findings of an ongoing literature during the four years of the research project, are presented at different stages: six months, 24 months and 48 months. These reviewed potential processes available to reduce oil in water concentration to below 15 parts per million and aspects of crossflow membrane filtration. Six membranes were chosen based on manufacturer/supplier recommendations, with pores sizes ranging from 0.2 μm to 0.005 μm. All membranes were made from ceramics with the exception of one organic membrane. Two test rigs were built: a laboratory and a field test rig. The laboratory rig was used to test all membranes using a synthetic oil in water emulsion, while the field test rig was used to test membranes with industrially produced emulsions on a ship and at a power station. Extensive testing of all six membranes was conducted in the laboratory, where the effects of process parameters were investigated. Based on various selection criteria the best three performing membranes were chosen with which to continue the research: Fairey 0.2 μm, Carbosep 300000 Daltons and Kerasep 15000 Daltons. All six membranes were also tested on a ship. Results showed that despite the different nature of the emulsion the membranes behaved similarly as when tested in the laboratory. The three chosen membranes were tested extensively in the laboratory to determine the best process parameters and to investigate the effect of concentration, which showed that the smallest pore sizes were less affected by increase in concentration. Further extensive testing was carried out both in the laboratory and at a power station to assess the membrane performance under an industrial process set-up and to investigate the effectiveness backpulsing and pressure ramping techniques. Results from these experiments formed the basis for the proposal of an industrial membrane filtration process to produce 1 m3.hr-1 of water containing less than 15 ppm of oil. The capital and yearly maintenance costs of the industrial membrane filtration rig were estimated at approximately £48,000 and £10,000 respectively Additional work included the determination of an accurate and reliable analytical method, which was dependent on oil used to calibrate the instrument and the development of a on-line technique for the visualisation of membrane fouling during crossflow filtration of oil in water emulsions.