Salt crystallisation at the surface of a heterogeneous porous medium

Evaporation of saline solutions from a porous medium often leads to the precipitation of salt at the surface of the porous medium. It is commonly observed that the crystallized salt does not form everywhere at the porous medium surface but only at some specific locations. To explain this phenomenon, we consider efflorescence formation at the surface of a saturated heterogeneous porous column (finer porous medium in coarse porous medium background) in the wicking situation. We study the impact of permeability and porosity contrasts on the efflorescence formation location from a simple visualisation experiment and Darcy’s scale numerical simulations. We show that the porosity is the most sensitive parameter for our experiment and that efflorescence forms at the surface of the medium of lower porosity. A simple efflorescence growth model is then used to explain why the efflorescence continues to grow at the surface of the lower porosity medium without spreading over the adjacent surface of the greater porosity medium. Introduction. – Understanding of evaporation from porous media in the presence of dissolved salts is important in relation with several applications, such as soil physics [1] or the underground sequestration of CO2 [2]. Another important field is the protection of buildings and of our cultural heritage [3] because of the severe damages caused by the salt crystallisation [4–6], that often results from the evaporation process. Depending on various factors such as the type of salt or the evaporation conditions [6], salt can precipitate within the porous medium where it forms subflorescence or at the surface of the porous medium where it forms efflorescence. The latter can be an important issue for the conservation of old paintings and frescoes [3]. Although the problem has motivated many studies, e.g., [3–6] and references therein, owing to the significance of related practical issues, the understanding of the efflorescence formation and growth at the surface of a porous material is surprisingly not very advanced. As discussed in this paper, this is because efflorescence results from a series of complex coupled phenomena between the internal transport of salt within the underlying porous materials, the external evaporation and the non-equilibrium growth process associated with crystallisation. However, developing theories and modelling of efflorescence formation and growth (as a first (a) E-mail: [email protected] step toward the still more involved problem of surface degradation) does not appear out of reach. The present paper is viewed as a step in this direction. Even small samples of real porous media are rarely homogeneous at Darcy’s scale and thus characterised by spatial variations in permeability and porosity. An archetypical example is a brick wall with properties markedly different in the bricks and the inter-brick mortar. The question thus arises as to whether the efflorescence will form at the surface of the bricks, of the mortar or both. To gain insight into the effect of structural heterogeneities on the location of the efflorescence formation and growth, we developed the simple experiment described in the next section. Experiment. – The experimental set-up is sketched in fig. 1. Evaporation takes place at the top surface of the porous medium and the dissolved salt is transported across the sample to the evaporative surface from the reservoir in contact with the porous material at its bottom. As sketched in fig. 2, the experiment is performed for a homogeneous column (a pack of 1mm glass beads) as well as for a heterogeneous one: a column containing a radially symmetric fine textured inclusion in the middle of a coarser porous medium. The coarser porous medium is the same as in the homogeneous column whereas the inclusion is a consolidated porous medium made of sintered glass Fig. 1: (Colour on-line) Sketch of the experiment. A column of porous medium of height L = 35 mm is set in a hollow cylinder (38 mm in diameter and 50 mm in height). The bottom portion of the sample is submerged in a near saturated aqueous solution of sodium chloride (C =C0 = 25 g NaCl/100 g solution; the saturation concentration is Csat = 26.4 g NaCl/100 g solution). The aqueous solution level in the brine reservoir is such that the medium remains saturated during the experiment duration owing to capillary rise. The system is set in a cylindrical enclosure (not shown) of controlled relative humidity (RH ≈ 7%) and temperature (T ≈ 22.4 ◦C) (see [7] for more details). The distance δ between the porous medium surface and the cylinder entrance rim is 15mm. Fig. 2: (Colour on-line) The experiment is performed for a homogeneous porous column (left) as well as for a heteroge-