Synthetic membranes are increasingly used for energy-efficient separation of liquid and gaseous mixtures in household applications, environmental technology and the chemical and energy industry. Besides, membranes are used in component-specific sensors in gas and liquid streams, preferably combined with micro-electronic devices. Ceramic membranes have a large potential over their polymer counterparts for applications at high temperature, pressure and in aggressive environments. Ceramic membranes offer the additional advantage that they can be cleaned at high temperatures or by the use of steam (sterilisation). The first enables their extended use in household applications, while the latter is of importance in food processing and medical applications. Future use of ceramic membranes includes their application in chemical reactors or in combination with heterogeneous catalysts. Both options are a chemical engineer’s dream and therefore studied much. Preparation of porous ceramic membranes is expensive, which is a general problem for ceramic materials. This is caused by the fact that ceramic materials cannot be processed in a liquid state as their polymer and metallic counterparts. Much attention is therefore paid to the ‘powder technology’ of ceramic materials in which powder preparation, consolidation, drying and sintering are essential steps. State-of-the-art ceramic membranes are produced by ‘colloidal’ or ‘suspension processing’, which implies that an essential step in the process is the dispersion of particles in a suitable carrier liquid and the subsequent consolidation of the powder into a ‘green cast’. The consolidation step is the subject of part I of this thesis. First a generic overview is given (chapter 1), after which more attention is paid to cast formation by centrifugal casting. First, the case is considered of centrifugal casting of a monodisperse suspension (chapter 2) and, second, of a bidisperse suspension, resulting in a graded cast structure (chapter 3). In chapter 4, the influence of gravity on the development of the cast profile during filtration is discussed, while chapter 5 describes cast formation by making use of an electrostatic force. In chapter 6 a criterion is discussed which predicts the conditions for homogeneous cast formation. In part II several properties of porous ceramic membranes are discussed (e.g., produced by suspension processing as discussed in part I). First, their permeation and tensile strength when applied in a membrane module are considered, which determine their optimum performance (chapter 7). Second, the adhesion strength of a thin membrane layer on top of a substrate is discussed on the basis of experiments in a pressure vessel (chapter 8). The porous ceramic membranes described are used in the osmotic tensiometer, a humidity sensor to be applied in soil environments. This is the subject of part III. This application field implies the measurement of humidities close to 100%. Humidity sensors for this range are and will be an essential element of agriculture and horticulture in a world of increasing prices for clean, potable water. However, no adequate method exists yet when considering criteria as response time, endurance, reproducability and ease of use. A novel technology using a combination of dedicated polymer, a state-of-the-art ceramic membrane and a sensitive pressure transducer may give the desired result. This is the osmotic tensiometer. In chapter 9 several aspects of ‘response time’ for application of the osmotic tensiometer under realistic field conditions are quantified in a transport model. This model is validated with experiments using a novel design of the osmotic tensiometer. Chapter 10 deals with the favorable consequences of the use of cross-linked polymer grains inside the osmotic tensiometer instead of short linear chains. This results in a decrease of response time and a cheap and user-friendly osmotic tensiometer. The features of this second design are essential for a bright future of the osmotic tensiometer. In chapter 11 anomalous temperature changes in conventional and osmotic tensiometers are described. On the one hand, these changes may damage the sensor, while, on the other hand, they may give valuable information on several transport parameters related to tensiometer performance. Model results are compared with experiments using a third, automated tensiometer design.
|Award date||4 Feb 2000|
|Place of Publication||Enschede|
|Publication status||Published - 4 Feb 2000|