The extraction stage of edible oil in the oil industry is commonly performed by using toxic solvents (e.g. hexane) and processes with high energy consumption (e.g. distillation, evaporation) to recover the solvent, which represents around 70-75 wt% in the oil-solvent mixture. In this paper, a membrane-based extraction method using nanofiltration (NF) membranes is presented. Commercial nanofiltration membranes made of different polymers (Desal-DK-polyamide NF from GE-osmonics Ò , NF30 polyethersulfone NF from Nadir Ò , STARMEM TM 122 polyimide from MET Ò and SOLSEP NF030306 silicone base polymer SOLESP Ò ) were selected and tested to recover the solvent from soybean oil/solvent (10-20-30% w/w oil) mixtures at various separation pressures and constant temperature in a dead-end filtration set up. The selection of the solvent was made in order to compare solvents obtainable from renewable resources, such as ethanol, iso-propanol and acetone, with solvents traditionally used in the industry (i.e. cyclohexane and n-hexane). The structural stability of the membranes towards the different solvents used in this work was verified visually, by the variation of the membrane area and by means of permeate flux assessments. Desal-DK and NF30 showed poor filtration performance and even visible defects after exposure to acetone but a good performance was obtained for the nanofiltration membranes STARMEM TM 122 and SOLSEP NF030306 with ethanol, iso-propanol and acetone. For example, considering a mixture with 30% edible oil in acetone, STARMEM TM 122 shows a flux and oil rejection of 16.8 L m -2 h and 70%, respectively. For the same conditions, SOLSEP NF030306 exhibited a flux of 4.8 L m -2 h with 78% rejection, which shows the potential application of nanofiltration membranes in the oil industry.

No abstract

Dear Editor,We wish to reply to the comments made on our paper ''Performance of nanofiltration membranes for solvent purification in the oil industry'', by S. Darvishmanesh, T. Robberecht, J. Degrëve, and B. Van der Bruggen and published in J Am Oil Chem Soc 88, 1255-1261 (2011) [1] by Albert Dijkstra [2]. We are very pleased that the paper invoked discussion, which should be the objective of all scientific research. Nevertheless, it is clear that the discussion in the letter to the editor by Albert Dijkstra is rather subjective and based on a difference in opinion arising from his experience in the oil industry in comparison with the paper's objectives.He raised the issue of energy consumption. We claimed that conventional solvent recovery by distillation consumes about 50% of the total energy in edible oil extraction. In our paper, we stated that nanofiltration requires much less energy than distillation. Dijkstra believes that we overestimated the energy savings because most of this energy is supplied by the vapors leaving the desolventizer/toaster and thus there is little net energy saving. This may be true but is not relevant to our study, in which a comparison of technologies is made without taking into account the overall energy balance. His comment assumes that part of the heat that is supplied is waste heat and should therefore not be taken into account, or in other words, to paraphrase George Orwell (in his book Animal Farm) with some adaptation, all energy is equal but some energy is less equal than the rest.The analysis ignores the efforts currently being made on usage of waste heat, which may turn waste heat into a valuable resource. Moreover, technologies become available that produce significantly less waste heat so that this heat may not be available for distillation in a later stage. In view of this trend, it seems inappropriate to make a difference between different sources of energy. This is also a matter of opinion, but it cannot be denied that utilizing low grade thermal energy is more and more necessary [3].However, the proposal of membrane filtration as an alternative for distillation is indeed not optimal. A combination of membrane filtration and distillation as an alternative for the combination of a desolventizer/toaster and classical distillation might be even more advantageous and should be investigated, in view of suggested advantages of hybrid processes [4].Regarding the use of isopropanol, it is not at all straightforward to consider a heating and cooling cycle more energy efficient than a pressurization and depressurization cycle. Much depends on the way these technologies are implemented. A heating and cooling cycle can be operated most efficiently by integrating heat pump technology. Yet it is not at all clear this is technically feasible or economically viable. Moreover, the application of heat pump technology in industry is mostly an emerging field [5]. On the contrary, a pressure recovery device can be used to recover the applied pressure during nanofiltration. These devices ...