Search

Reverse Osmosis (RO) is a widely used technology for water treatment in the beverage industry to produce brewing water, process water or tap water. The weak points of RO-membrane filtration are membrane degradation due to oxidative water constituents or membrane fouling in the difficult-to-clean, spacer-filled feed channel. The fouling can be distinguished into two main categories. The first one is caused by feed waters with a high organic and microbial load. The second type of fouling is triggered by supersaturated salt solutions, which leads to the precipitation of salts on the membrane surface. Both fouling types are destroying spiral wound RO-modules if they are not detected in an early stage. This work presents and validates a new optical online detection method via polymer optical fibres (pOf) for inorganic fouling as CaCO3, CaSO4, BaSO4, SrSO4. The new detection method is tested by three experimental set-ups and compared with common prediction and detection methods such as saturation calculation, measurement of salt rejection, permeability, and the pressure drop in the feed channel. In addition to conventional online analysis methods for detecting deposits in reverse osmosis systems, this study presents an image analysis method that provides reliable evidence of POF sensor operation. The POF sensor is able to detect incipient crystal formation during the RO process. This gives this study the opportunity to discuss current crystallisation theories such as the induction time theory and the role of monohydrated calcium carbonate (MCC) as a precursor in the formation of CaCO3 deposits. [ABSTRACT FROM AUTHOR]

Reverse osmosis (RO) is a widely used technology for providing brewing and process water in the beverage industry. In order to design the membrane process in a resource-saving and cost-effective way, the prediction of salt deposits in the membrane modules is of great importance. Since calcium and carbonate containing waters are very common, the solubility of CaCO3 is often the limiting factor of the raw water yield. The prediction and thus the plant design is largely based on chemical saturation indices as the LSI and S&DSI. However, the calculation of these parameters is done on the basis of simplifications of the water chemistry, which are no longer necessary due to computer-aided calculation methods. Furthermore, there is no valid theory on CaCO3 crystallisation based on fundamental energetic considerations. The classical crystallisation theory, which had been considered valid for decades, has proven to be insufficient. Improved theories are currently under discussion and a validation for CaCO3 crystallisation is not available today. Nevertheless, the polymorphism of CaCO3 seems to be of great value for the description of different saturation and crystallisation zones. Besides the most stable CaCO3 form, calcite, the monohydrate calcium carbonate (MCC) has been characterised as crucial for the description of metastable solutions. Thus, the solubility product of calcite does not provide a suitable basis for predicting CaCO3 precipitation as applied in the LSI and S&DSI. With this work, we introduce a hydrochemical simulation model of the RO process that can be linked to the chemical equilibria of other crystal forms of CaCO3 such as the MCC. The calculations can be executed in the freely available water chemistry software PHREEQC. Besides the prediction of salt deposits, the literature shows a great need for reliable technical monitoring of fouling and especially inorganic fouling (scaling). In this context, we present a new fibre-optic based fouling detection system. Our proposed minimally invasive measurement technique allows spatially resolved measurements in membrane systems. The detection is based on the interaction of light and deposits such as salt on an optical fibre. The fibre optic sensor has the potential to provide the necessary database for validation and further improvement of the simulative prediction. [ABSTRACT FROM AUTHOR]