Modelling local hygrothermal interaction between airflow and porous materials for building applications

Moisture related damage in buildings is a phenomenon which is familiar to most people. Most of the time it is spontaneously associated with damage due to liquid moisture transport such as plumbing leaks, rising moisture in walls, . . . Yet some materials and objects are so sensitive to moisture that they can already be damaged by water vapour transport through the air. This is especially true for culturally or historically valuable artefacts: even a small amount of damage (like small cracks, . . . ) is unacceptable for these objects. The reason for their high sensitivity for moisture related damage can be found in the used materials: these objects are typically composed of wood or other organic materials which strongly expand in function of the moisture content (and thus indirectly in function of the relative humidity). This means that subsequent fluctuations of the relative humidity in the air can result in an expansion and deformation of the object and the hereby induced tensions can lead to fractures or other damage phenomena (e.g. cracking of paint). To preserve these objects as good as possible it is hence extremely important to keep the relative humidity in the surrounding air as constant as possible. Art objects are typically stored and exhibited in historical (e.g. churches) or monumental buildings (e.g. museums). While it is now standard procedure to place a HVAC installation in large buildings, this was of course not the case for historical buildings. Yet due to the increased demand on thermal comfort, these historical buildings are, when retrofitted, more and more equipped with at least a permanent heating system. The intermittent use of a heating system however results in considerable temperature fluctuations and thus also in important relative humidity fluctuations. Due to the large volume of these buildings the temperature and relative humidity fluctuations will strongly vary in space: during heating the temperature above an air inlet will for example rise much stronger than the temperature in the rest of the building. Yet air flows in a building can also become very complex. It is hence not always possible to intuitively predict where the largest fluctuations will occur and it is even harder to estimate the magnitude of these fluctuations and the associated risk of moisture related damage. There is thus a need for a model that can predict the local fluctuations in air temperature and relative humidity and the associated hygric response of individual objects in order to predict the risk of local moisture related damage. This dissertation is the result of research conducted in search of suitable strategies to model local temperature and relative humidity fluctuations in the air and porous materials for the application in buildings. Chapter 1 starts with an elaborate introduction of moisture related damage induced by relative humidity fluctuations. Next a literature review on existing models and modelling techniques used in hygrothermal simulations for buildings is presented. This overview shows that at present the existing models can not be used for the applications aimed at in this work as they either do not offer the required level of detail or as there are limitations in the used physical models (e.g. only 2D, . . . ). It is also shown in this chapter that a combination of CFD flow simulation and a hygrothermal material model is best suited for the simulation of the local interaction between the air and the porous material. The second chapter focuses on the use of transfer coefficients for the modelling of the hygrothermal interaction between air and porous materials. If it would be possible to use transfer coefficients in the air model then the required computational power would drastically decrease. The study on transfer coefficients starts with an evaluation of the different definitions of the mass transfer coefficient. It is demonstrated that the use of vapour densities in this definition, results in a dependence of the mass transfer coefficient on the temperature difference between fluid and wall. This reduces the applicability of mass transfer coefficients in practice. Yet, when mass fractions are used in the definition, the value of the mass transfer coefficient no longer depends on the temperature difference and the applicability is strongly increased. Next these results are used to study the possibility of using the heat and mass (moisture) analogy to simplify the prediction of local mass transfer coefficients in buildings. It is shown that this is not feasible. The chapter is concluded by checking to which extent the local transfer coefficients remain constant during a transient process in a building environment. It is found that it is not possible to use constant transfer coefficients during long periods of time and that a possible advantage in calculation time associated with the use of transfer coefficients cancels since even in case of a transient moisture response with steady-state flow, the local mass transfer coefficients strongly vary. As it was shown that the use of constant transfer coefficients is not an option for the applications aimed at in this work, the choice was made to directly couple the CFD model and the hygrothermal material model. To reduce the extra computation time imposed by this coupling, the hygrothermal material model is integrated into the CFD solver. This means that no time consuming data exchange between two separate codes is necessary. Chapter 3 presents the combined heat and moisture transport equations which have to be solved by the extended solver. Chapter 4 describes how the transport equations introduced in Chapter 3 can be transformed into a discretized form and which numerical techniques are needed to assure the conservation of mass and energy. Implementation of these discretized transport equations into the CFD solver results in the coupled CFD - material model. The different solver settings are briefly discussed. The well functioning of the coupled model is proved by means of a verification and validation study. In Chapter 5 the newly developed model is applied to a case study: the microclimate vitrine. In such a vitrine valuable and fragile objects (e.g. paintings) are placed for protection against fluctuations in relative humidity. The operating principle of the vitrine is based on the fact that due to the hygric buffering of the object the relative humidity of the small amount of air in the vitrine is stabilized. To make a correct and detailed analysis of the effectiveness of such a vitrine it is of crucial importance to take the local interaction between the air and the object into account. Simulations performed with the new model confirm the stabilizing influence of the vitrine and offer an explanation for the non-intuitive phenomena experienced in practice for this type of vitrines. The new model clearly offers an added value when analyzing the risk of moisture related damage. Chapter 6 concludes this work by summarizing the most important conclusions and by looking forward to possible future research.