Hygrothermal effects and moisture kinetics in a bio-based multi-layered wall: Experimental and numerical studies

Abstract

A bio-based multi-layered reference wall has been developed within the framework of the European ISOBIO project. One of the key points of this project was to be able to perform proper simulations of the hygrothermal transfers occurring inside such walls. Previous published investigations, also performed in the framework of this project, have demonstrated that the classic assumption of instantaneous equilibrium between local relative humidity and water content according to the sorption isotherm is not relevant for bio-based porous materials, where, in practice, a rather slow kinetics of sorption occurs. The theoretical background developed in this previous study is used here to determine the kinetic constants of the bio-based construction materials and to perform 1D hygrothermal simulations. The kinetics constants are determined thanks to a 1D cylindrical tool based on the local kinetics approach, validated against several experiments of sorption. Then, heat and hygric transfers recorded on a demonstrator building (The HIVE, Wroughton, UK) are analyzed and are simulated using two modeling tools: TMC based on the Künzel approach and TMCKIN based on the local kinetic approach. From the simulations, the local kinetics improves the small timescale RH dynamics. The comparison with measurements performed in the demonstrator confirms the relevance of the local kinetics approach.

Modeling of hygrothermal transfers through a bio-based multilayered wall tested in a bi-climatic room

“…Actually, conventional simulations greatly underestimate the time required for the water content to reach the equilibrium. This was shown for bio-based materials: in [10] for cellulosic materials, in [11] and [12] for wood, in [13] for paperboard, by Reuge et al in [1] and [2] for ISOBIO materials and in [14] and [15] for various hemp-lime concretes, but also for more classic compounds such as cements [16,17]. These deviations obtained from conventional models are of particular concern when hygric transfers are subject to permanent fluctuations of relative humidity (RH) such as in building envelope materials, food materials or soils.…”

“…The appropriate models of properties needed for the simulations have been studied in [2]. For the isotherms of adsorption, the Van Genuchten model (VG) [28] has been used.…”

“…The hygric dependence of the vapor diffusion resistance factors µ has been neglected except for the Proclima INTELLO because this is a hygrovariable membrane evolving very significantly as a function of the moisture content (from 550 at 100% RH to 136400 at 0% RH). The data provided by Proclima / F-IBP have been fitted by a logistic power law [1,2] (because of its great values, this parameter must be carefully modeled):…”

“…The main difficulty with such simulations is to correctly describe hygric phenomena in bio-based materials. Reuge et al work on this subject and several studies are already available [1,2].…”

“…Thus, conventional approaches describing hygrothermal transfers in building materials are based on the assumption that for a given local relative humidity ϕ, the corresponding equilibrium local water content w is reached instantaneously (Künzel approach [18]). In our previous studies [1,2,14,15], various bio-based materials were studied and it was demonstrated that such an assumption led to patent inconsistencies such as a water vapor / liquid water equilibrium reached too fast at the local scale and at the sample scale and / or underestimation of the local RH dynamics. Thus, as reported in the aforementioned literature, our previous studies have established that a local kinetics of sorption exists (from water vapor to liquid water and inversely) which can be slow compared to the diffusive phenomena.…”