Villiam Bortolotti

Models of water imbibition in untreated and treated porous media validated by quantitative magnetic resonance imaging

Fluid imbibition affects almost every activity that directly or indirectly involves porous media, including oil reservoir rocks, soils, building materials, and countless others, including biological materials. In this paper, magnetic resonance imaging (MRI) has been applied to study water imbibition in a porous medium, in which capillary properties are artificially changed. As a model system, samples of Lecce stone, a material of cultural heritage interest, were analyzed before and after treatment with a protective polymer (Silirain-50 or Paraloid PB72). By using MRI, we can visualize the presence of water inside each sample and measure the height z(t) reached by the wetting front as a function of time during experiments of capillary absorption before and after treatment. The sorptivity S, defined as the initial slope of z versus t1/2, has been determined before treatment and through both treated and untreated faces after treatment. Very good fits to the data were obtained with theoretical and empirical m...

Changes of the local pore space structure quantified in heterogeneous porous media by 1H magnetic resonance relaxation tomography

Magnetic resonance imaging and relaxation analysis are combined in a spatially resolved technique (relaxation tomography), which is able to quantify the parameters connected to the local structure in the internal regions of a porous material saturated by water, giving information on the pore space structure beyond the nominal instrumental resolution. Voxel-by-voxel longitudinal (T1) and transverse (T2) relaxation curves are acquired in order to obtain T1, T2 and S(0) maps, where S(0) is the extrapolation to zero time of the total equilibrium magnetization corrected for T2 decay. The proposed method permits evaluation of the porosity (ratio of pore space to total volume), at different length scales, from the sample to the voxel, not all achievable by traditional methods. More striking is its ability to describe how porosity is shared among different classes of surface-to-volume ratios of diffusion cells (the regions that the individual water molecules, starting at their particular positions, can experience...

Quantitative determination of porosity: A local assessment by NMR imaging techniques

The local determination of porosity is an extremely valuable target because of different applications in petrophysics. In fact, obtaining a reliable profile of porosity or saturations could considerably improve the evaluation of transport properties of porous media, especially if applied to multiphase flow test (i.e., for relative permeability characterisation). However, the best procedure to adopt for these kind of studies is currently under debate, involving different experimental choices both for acquisition and hardware solutions. The choice of a reliable procedure could be particularly important especially if heterogeneous samples (i.e., vugular or fractured carbonates) are approached. We have investigated selected carbonate cores, characterised by fractures and large vugs, using different instrumental solutions in order to assess usefulness of application of NMR imaging methods as a porosity measurement tool for heterogeneous samples. Local porosity data have been obtained and discussed relative to conventional measurements and petrographical aspects of rock samples.

Water vapor absorption in porous media polluted by calcium nitrate studied by time domain Nuclear Magnetic Resonance

Nuclear magnetic resonance relaxation analysis of liquid water (1)H nuclei in real porous media, selected for their similar composition (carbonate rocks) and different pore space architecture, polluted with calcium nitrate, is presented to study the kinetics of water condensation and salt deliquescence inside the pore space. These phenomena are responsible for deterioration of porous materials when exposed to environmental injury by pollution in a humid atmosphere. The theory is well described for simple pore geometries, but it is not yet well understood in real porous media with wide distributions of pore sizes and connections. The experiment is performed by following in time the formation of liquid water inside the pore space by T(1) and T(2) relaxation time distributions. The distributions allow one to see the effects of both the salt concentration and the pore space structure on the amount of water vapor condensed and its kinetics. It is shown that, for a given lithotype, even with different amounts of pollutant, the rate-average relaxation time T(1ra) tends to increase monotonically with NMR signal, proportional to the amount of liquid water. T(1ra) is often inversely associated with surface-to-volume ratio. This suggests a trend toward the filling of larger pores as amounts of liquid water increase, but it does not indicate a strict sequential filling of pores in order of size and starting with the smallest; in fact, relaxation time distributions show clearly that this is not the case. Increased amounts of salt lead to both markedly increased rates and markedly increased amounts of water absorption. NMR measurements of amounts of water, together with relaxation time distributions, give the possibility of information on the effect of pollution in porous materials exposed to humid atmospheres but sheltered from liquid water, even before the absorption of large amounts of moisture and subsequent damage. These phenomena are of importance also in other fields, such as the exploitation of geothermal energy.

Magnetic resonance lifetimes as a bridge between transport and structural properties of natural porous media

Abstract Results are reported of an NMR spin-lattice relaxation study on a suite of sandstone cores fully saturated with brine. Relaxation curves of 1 H nuclei of water filling the pores, obtained by Inversion Recovery pulse sequence, were analyzed to extract relaxation lifetime values. These lifetimes have been recently proposed as suitable to characterize the relaxation behavior of fully saturated sandstones as well as able to predict permeability in conjuction with porosity. An improvement in the comprehension of interplay among geometry and transport properties of the pore space can be obtained from the comparison between these lifetimes and conventionally measured properties. The results obtained are in agreement with previous results on sandstones and reproduce very well results on model systems. They led to a functional dependence among permeability ( k ), porosity (φ) and average lifetimes ( T ls ) that allowed a good estimation of k from φ and T ls . The dependence of the error in the permeability estimation in terms of the variations of φ and T ls exponents in the relationship that predicts k was also discussed. The reason of the correlation between k and T ls lies in their common dependence on the surface-to-volume ratio S / V p . In this way, 1 H spin-lattice relaxation lifetimes behave as a bridge between structural and transport properties and may be viewed as a dynamically weighted version of S / V p . The discussion of these results can help in clarifying the informative content of 1 H spin-lattice relaxation curves on a broad spectrum of heterogeneous systems with high surface-to-volume ratio.

PERFIDI filters to suppress and/or quantify relaxation time components in multi-component systems: An example for fat–water systems

Parametrically Enabled Relaxation FIlters with Double and multiple Inversion (PERFIDI) is an experimental NMR/MRI technique devised to analyze samples/voxels characterized by multi-exponential longitudinal relaxation. It is based on a linear combination of NMR sequences with suitable preambles composed of inversion pulses. Given any standard NMR/MRI sequence, it permits one to modify it in a way which will attenuate, in a predictable manner and before data acquisition, signals arising from components with different r rates (r=1/T1). Consequently, it is possible to define relatively simple protocols to suppress and/or to quantify signals of different components. This article describes a simple way to construct low-pass, high-pass and band-pass PERFIDI filters. Experimental data are presented in which the method has been used to separate fat and water proton signals. We also present a novel protocol for very fast determination of the ratio between the fat signal and the total signal which avoids any time-consuming magnetization recovery multi-array data acquisition. The method has been validated also for MRI, producing well T1-contrasted images.