Multiple international test standards exist by which the capillary water absorption coefficient (A_w) of building materials can be estimated, as summarized in Table 2. The experimental test setup employed by the majority of ISO and EN standards to determine the capillary water absorption in building materials is simple, requiring minimal equipment, as the experiments involve gravimetrically monitoring the capillary absorption of water, as described by Hall in 1989 [38]. Samples are first preconditioned by drying them out in an oven until constant mass is attained, followed by cooling to the temperature at which water absorption tests should be performed. Once preconditioned, samples are partially immersed in water while being elevated on rods or support pads to allow access of water to the bottom inflow face, as shown in Figure 10. To avoid trapping air under the sample, the samples are immersed at an angle. The water level is kept constant during the test, typically a few mm above the base of the specimen. The samples are then weighed at standard intervals as the waterfront gradually rises to the top of the specimens. Once the waterfront reaches the upper face and the sample achieves the capillary moisture content (w_cap), no further change in weight is observed [37]. The capillary water absorption coefficient can then be determined via the equations corresponding to the chosen standard or model as described in Table 2. Certain methods have somewhat different setups; for example, EN 15801 is more suitable for specimens that are easily eroded when in direct contact with water. In this standard, a bedding layer such as filter paper or cotton is placed at the bottom of the vessel, and water is added to saturate this layer, taking care that the water level does not surpass the height of the bedding. The specimen is then placed on the saturated bedding, acting as the water supply for the capillary absorption test. In this case, the sample does not undergo partial immersion. Other methods, such as ASTM C1585-20 and C1794-19, have the additional step of sealing the side surfaces of the specimen, to restrict water transport solely to the bottom face and prevent evaporation from the lateral faces.

It is worth noting that each method to estimate A_w inherently produces different results due to differences in the equations on which the methods are based. In 2016, Karagiannis et al. [39] compared the capillary water absorption coefficient of three different building materials, brick, stone, and natural hydraulic lime, obtained from three different European standards, referred to as the one-tangent method, the two-tangents method, and the 30 min method. The two-tangents method was found to be the most suitable for determining A_w, especially for materials with non-linear initial absorption. Their work also confirmed the linear relationship between A_w and air temperature, validated through a case study. Earlier that same year, Feng and Janssen [40] aptly pointed out that although a number of studies had focused on investigating the influence of temperature on hygric properties, results tended to differ from one study to the next, leading to opposing conclusions. In their comprehensive experimental study, they determined the capillary absorption coefficient and capillary moisture content of three common porous building materials, autoclaved aerated concrete, calcium silicate board, and ceramic brick, at three different temperatures. While the influence of temperature on capillary moisture content was deemed negligible, there was a notable positive correlation between capillary absorption coefficient and temperature. In addition, their experimental results were in good agreement with theoretical predictions derived from fundamental theories, demonstrating that the mathematical equation applied can be used to correctly predict temperature dependence [41]. This sheds light on the importance of considering the coefficient’s dependence on temperature when characterizing building materials [39]. As such, with any of these methods, it is important to report all experimental conditions, including environmental conditions such as laboratory temperature; this is especially important for isothermal methods.

Isothermal methods focus on studying moisture migration under constant temperature conditions. These methods assume that temperature differentials do not significantly influence the movement of moisture within the building envelope. Isothermal experiments typically involve subjecting building materials or assemblies to steady-state conditions, where the temperature remains constant throughout the test duration. Non-isothermal methods, by contrast, account for temperature differentials that influence moisture migration in building envelope materials, recognizing that temperature variations play a significant role in driving moisture movement. Non-isothermal experiments involve subjecting building materials or assemblies to varying temperature conditions, which mimic real-world scenarios more accurately. Although the focus of this section is on the technical aspects of experimental methods and characterization techniques, with the majority of examples applying to the material scale, it is worth noting that these methods may also be applied at the composite scale. The study of interfacial phenomena as it relates to the transport of moisture between building materials has been ongoing for many years, and continues to be an active area of research [42,43,44,45,46]. As an example of relevant work that employed a standard partial immersion method, Ramirez et al. (2021) studied the brick-mortar interface via a comprehensive experimental program that included capillary absorption tests following EN ISO 15148 [44], and their results demonstrated that imperfect hydraulic contact between materials has an impact on water absorption behavior. The effect of hydraulic resistance was observed as varying degrees of reduction in the water flow through the samples, and was attributed to a discontinuity in the pore structure.

X-ray computed tomography (CT) is an imaging technology most commonly known for its applications in the field of medicine. Benchtop-scale apparatus operating under the same theory are available for material science laboratories and have been used to investigate water transport and water distribution in building materials. Roels and Carmeliet (2006) used microfocus X-ray radiography to analyze moisture transport in ceramic brick and calcium silicate (CS) brick during capillary uptake experiments. The output of this method is a two-dimensional image that directly shows moisture distribution within the sample. In their method, moisture content was quantitatively determined via logarithmic subtraction of a reference image of a dry sample, from images of the wet samples [54]. Calle et al. (2019) employed a similar procedure involving microfocus X-ray absorption tests calibrated using wet and dry samples, to investigate liquid moisture transport at the interface between ceramic brick and natural hydraulic lime mortar [55]. In 2014, Boone et al. used an enhancing method to obtain water distribution in limestone using CT with good contrast [56]. The enhancement refers to doping the water with 5 wt% caesium chloride (CsCl), to improve the X-ray attenuation of the fluid; this improved attenuation is due to the high atomic number of Cs. Keeping the concentration low, at 5 wt%, ensures that the surface tension of the solution remains very close to pure water, having a negligible impact on the water uptake behavior [56]. Building off from this enhancement method, Yang et al. (2015) applied the technique to investigate water transport in cement-based materials using CT. The introduction of CsCl as a contrast agent by Boone et al. was necessary, as water in cementitious materials exhibits low contrast in CT [57]. In this setup, a sample is partially immersed in a dish of water and is elevated on supports. The researchers demonstrated that it is possible to obtain images of the waterfront moving up the sample through capillary action in situ, using X-Ray CT, and have explored a variety of building materials including cement paste, mortar, and concrete [57,58]. Figure 11 shows an example of the two-dimensional moisture-profile images produced directly by this technique, during a water uptake experiment using mortar. The bright zones indicate areas saturated with water, and the progressive increase in brightness over each exposure period indicates the advancing waterfront through the sample. From the greyscale version of these images, the height of the water uptake was calculated by drawing vertical lines on each image and collecting the grey value along the lines. The point at which the grey value has a sudden drop is taken as the height of the waterfront. Plotting the distance of water uptake against the square root of time, and fitting the relation using the least square method, results in a slope corresponding to the capillary coefficient of the material.

Gamma radiation techniques, including both gamma-ray transmission (attenuation) and gamma-ray scattering, are another sub-class of well-established electromagnetic techniques that have been used to determine the moisture content in building materials non-destructively. In 1967, Hilsdorf listed gamma radiation methods as one of the ways to determine the water content of hardened concrete [59]. More recently, da Rocha et al. (2001) measured moisture profiles of concrete by gamma-ray transmission [60], while others have employed the gamma-ray scattering method to determine water content in concrete and red fired-clay bricks [61,62]. The transmission method requires the source and the detector to be on opposite sides of the measured object, making it more suitable for smaller laboratory-scale specimens; these experiments function based on signal attenuation when a gamma-ray beam passes through the thickness of a material. The scattering method is particularly useful in building science applications, as both the radiation source and the detector can be placed on the same side of the measured object, such as a wall [62]. This is advantageous for large or thick samples, and when the ability to rotate or access both sides of the specimen is limited [61]. An advantage over X-ray methods lies in the penetration depth; as gamma rays have higher energy than X-rays, the penetration depth of this technique is greater, allowing for thicker samples to be tested [61]. The drawback of gamma-ray techniques is that they require a lead-shielded radioactive source, as gamma rays are generated from the radioactive decay of atomic nuclei.

Gamma-ray scattering is also referred to as gamma-ray Compton scattering, after the American physicist Arthur Holly Compton who discovered the effect in 1923 while studying X-rays [63]. The phenomenon now refers to the interaction of high-energy photons (either X-ray or gamma-ray) with outer orbital electrons in the absorbing medium [64]. When one of these photons collides with an outer shell electron in the absorbing medium, a few key things occur: the electron is ejected, the photon loses some of its initial energy which increases its wavelength, and the photon is deflected from its original path, where the angle of deflection (θ) is proportional to the energy loss [65]. The probability of Compton scattering is directly proportional to the electron density of the absorber. In medical applications, higher water-density tissues exhibit the highest amount of scatter in the body [65]; similarly, in porous building materials such as concrete, an increase in water content increases the density of scattering centers (electrons), and therefore scattered photon intensity increases as a function of water density [61]. Figure 12 illustrates the basic components of a gamma-ray Compton scattering experiment.

A commonly used system is the NMR-MOUSE (Mobile Universal Surface Explorer) which was first built in 1995 by Blümich et al., following the principles of stray-field imaging (STRAFI) that were developed around the same time [66]. The NMR-MOUSE is a portable system consisting of two main components—a probe, and a spectrometer. As illustrated in Figure 13, the probe is positioned on a computer-controlled high-precision lift powered by a step motor, allowing the operator to step the sensitive slice through the sample to collect an NMR profile with microscopic resolution. Two permanent magnets are embedded in an iron yoke and separated by a gap containing a solenoidal radio-frequency (RF) coil, which excites and detects the NMR signal [66]. The orientation of the components in this setup results in an inhomogeneous static polarizing magnetic field (B₀) and an inhomogeneous RF field (B₁) that are orthogonal to each other above the surface of the probe; this produces a sensitive volume at a fixed distance from the RF coil [67,68]. This technique can be classified as a stray-field NMR technique, as the sample is positioned in the stray fields of the magnets and the RF coil [69]. It is worth noting that the maximum measurement depth is an inherent limitation in this technique; the distance is specific to each instrument’s configuration but is often on the order of tens of millimeters. For example, in the case of the PM25 (Magritek, Germany), the maximum profiling depth is 25 mm. Other SS-NMR devices exist, such as the GARField magnet, whose utility was explored for the analysis of concrete hydration in the built environment [70], and the NMR-MOLE. As with the NMR-MOUSE, these have a sensitive measurement volume outside the device, allowing for non-destructive testing of arbitrarily large planar samples.

In 2004, Casieri et al. used a Bruker Eureka-Mouse 10 to investigate the water content in wood, a porous material often used in construction [71]. As the wood matrix itself contains protons (in cellulose and other macromolecular components), they differentiated between the proton signals from wood and the proton signals from water within the wood structure. Going a step further, they differentiated between cell wall water and free water (lumen water). This is because wood contains macromolecules that can hydrogen-bond water molecules, as well as cell cavities (lumens) that can fill up with water. The three types of protons found in these specimens have different levels of mobility, leading to different spin-spin (T₂) relaxation times, which makes it possible to differentiate between their ¹H SS-NMR signals and determine the moisture fraction in wood. In their study, chestnut, walnut, and sessile oak specimens were conditioned in environments having different relative humidities regulated by saturated saline solutions. They compared results from the ¹H SS-NMR volume fraction method and the standard gravimetric method and attributed the difference to a ”comparing factor” (P), close to one, that is related to the density of macromolecules in the wood. Once this factor was applied, a good correlation between both experimental approaches was obtained. A case was made for the need to adopt a slightly different definition of moisture content when dealing with SS-NMR that is based on volume (porosity index, IP), which can then be easily converted to the conventional gravimetric moisture content (MC), knowing the specific gravity of the dried sample [71]. Others have since applied SS-NMR to investigate moisture content, moisture transport properties, and pore size distribution in a variety of building materials, including gypsum [72], Lecce stone [73], and floor screed [74]. Most recently, Zhaxi et al. used a similar technique as Casieri et al. (2023) to measure the moisture content in brick, sandstone, mortar, and concrete samples [75].

In 2006, Bortolotti et al. used an mq-ProFiler (Bruker Biospin, Italy), a small (5 × 5 × 10 cm³) handheld ¹H SS-NMR, to study the capillary water absorption kinetics of porous building materials. By pairing this experimental technique with magnetic resonance imaging (MRI) and applying the Washburn model of water capillary rise, they quantified the sorptivity of two stone samples: untreated Lecce stone, and Lecce stone treated with a hydrophobic acrylic resin commonly used to protect cultural heritage and building materials [73]. The MRI technique allowed for the visualization of water at the sample scale, and monitoring of the waterfront to calculate sorptivity, while SS-NMR gave information at the pore scale. Their work demonstrates that SS-NMR can be successfully applied for in situ analysis of pore structure and capillary water uptake of porous materials. Colinart and Glouannec studied the drying of gypsum as a reference building material, using the NMR-MOUSE PM25 [72]. Calibration of the NMR signal was accomplished by wetting three specimens to achieve 10%, 20%, and 40% gravimetric moisture content, allowing the moisture to redistribute at standard conditions, and then collecting ¹H SS-NMR profiles. The drying experiment was performed by unwrapping only the upper face of a wetted sample, to ensure evaporation only occurred through that face, and collecting repetitive NMR profiles for 30 days. Applying kinetic and T₂ analyses to the collected profiles, they demonstrated that drying proceeds faster at higher moisture contents and that evaporation occurs from the larger pores before the smaller ones. These studies all demonstrate the utility of ¹H SS-NMR for the analysis of moisture properties of building materials.

Marynowicz and Kucharczyk [76] developed a method to simultaneously determine both the capillary absorption coefficient and the water diffusion coefficient of porous building materials using infrared thermography, i.e., a thermal imaging camera. This method has the advantage of using more accessible equipment compared to the aforementioned electromagnetic methods, making it available for broader use across different laboratories. Their setup involves placing the lower end of a specimen in water, as in the standard gravimetric methods. However, it is modified by suspending the specimen from a high precision balance, while a thermal imaging camera on a tripod monitors the side of the specimen, as shown in Figure 14. The camera records images at a frequency of 0.1 Hz and mass changes are continuously recorded by connecting the balance to the same computer as the camera. Although the sample and the water are at the same temperature when initiating the test, differences in their emissivity result in different apparent temperatures on the thermal images. Images collected throughout the test show the rising front of capillary absorption as a change in the apparent temperature distribution on the sample’s surface, as shown in Figure 14. The temperature distribution along a line in the vertical direction with respect to time is selected for analysis, which first involves correcting for the uniform temperature drop observed over the whole sample, attributed to evaporation. The saturation was then calculated at each position with respect to time, employing the assumption that temperature and saturation are correlated. From this calculated saturation, the apparent water mass in the sample was determined, resulting in a water absorption curve with respect to √t, from which A_w is extracted as the slope of the initial linear portion. Since evaporation causes the sample surface to cool down, thermal images should not be taken immediately after drying. This limitation was highlighted by Janz and Johansson (2000); further, for their particular application studying moisture content at the interface between mortar and brick, the spatial resolution of the camera was an additional a limiting factor [77].

In this technique, the relative temperature is used instead of the absolute temperature. Therefore, a camera with moderate accuracy is sufficient, making this a relatively low-cost method to assess capillary water intrusion in porous building materials. Marynowicz and Kucharczyk compared their thermographic measurements to gravimetric measurements collected simultaneously and noted that close agreement was obtained at the sample center, where the difference was 3%. An overall 7.5% difference was observed between the gravimetric A_w and the thermographic A_w, mean averaged across six lines spanning half the sample width [76].