Salt transport and crystallization in porous building materials L. Pel, H. Huinink, K. Kopinga* Department of Applied Physics, Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands Abstract Salt weathering is a major cause of deterioration of porous building materials. To obtain information about the mechanisms underlying these damage processes we have studied the moisture and ion transport. We measured the time evolution of NaCl saturated samples of fired-clay brick during one-sided drying using Nuclear Magnetic Resonance. The moisture content and amount of dissolved Na ions could be measured quantitatively as a function of position. The NaCl concentration profiles obtained from these data reflect the competition between advection to the surface and redistribution by diffusion. By representing the measured moisture and NaCl profiles in an efflorescence pathway diagram (EPD) also the crystallization can be taken into account. © 2003 Elsevier Science Inc. All rights reserved. Keywords: Porous materials; Moisture transport; Ion transport; Crystallization 1. Introduction Although salt damage has been intensively investigated for several decades [1– 4], the mechanisms and factors that control the formation of salt crystals in porous media and the development of damage by crystal growth are poorly understood. A better knowledge of the movement of water and ions during evaporation and salt crystallization in po- rous materials is required to explain salt damage for differ- ent materials and conditions. For a long time the develop- ment of realistic models for combined moisture and ion transport made hardly any progress, mainly because of the lack of adequate and reliable experimental data. However, using NMR imaging techniques [5] non-destructive mea- surements of the moisture and ion profiles in these materials are possible, offering new possibilities to tackle this prob- lem. Imaging of building materials is difficult, since in general these contain a large amount (up to a few percent) of magnetic impurities, like Fe and Mn. These impurities gen- erate magnetic field gradients up to a few T/m in the pores of the material, causing a very fast dephasing of the nuclear magnetic moments of the pore fluid and the dissolved ions [6]. The rapid signal attenuation due to diffusion within the pores requires the use of NMR pulse sequences involving very short spin-echo times. By using an especially adapted NMR set-up we were able to measure the moisture distri- bution in these materials in a quantitative way with a high spatial resolution [7]. Recently, we adapted this setup to detect also other nuclei, like Na, which offers the possibility to study the salt transport in building materials. 2. Experimental In this study we focus on the salt transport during single- sided drying of a material that has been initially saturated with a NaCl solution. The NMR apparatus used in this study has an iron-cored electromagnet generating a field of 0.8 T. Anderson gradient coils generate a constant gradient of 0.3 T/m in the vertical direction, resulting in a one-dimensional resolution of the order of 1 mm. A Faraday shield has been added to suppress the effect of changes of the conductivity or dielectric properties of the sample [7] due to variations of the moisture and/or ion content. By a computer-controlled switch the LC circuit of the insert can be tuned to 33 MHz for hydrogen or to 8.9 MHz for Na imaging. The sample, which has a cylindrical shape with a diameter of 20 mm and a length of 45 mm, is moved vertically through the magnet with the help of a step motor. It is sealed at all sides, except for the top over which air with a relative humidity of 5% is blown. In this way a one-dimensional drying process is created. A standard Hahn spin-echo sequence with a 90° pulse of 15 s and an echo time TE = 205 s is used for hydrogen. For Na, the 90° pulse width equals 20 s and TE = 370 s. With these settings only the Na nuclei in * Corresponding author. Tel.: +31-40-247-4304; fax: +31-40-243- 2598. E-mail address: k.kopinga@tue.nl (K. Kopinga). Magnetic Resonance Imaging 21 (2003) 317–320 0730-725X/03/$ – see front matter © 2003 Elsevier Science Inc. All rights reserved. doi:10.1016/S0730-725X(03)00161-9