NMR methods for a comprehensive and fast characterisation of mass transport in porous materials

Abstract

Nuclear Magnetic Resonance (NMR) is a powerful tool for non-invasively investigating static and dynamic properties of fluids inside complex and opaque structures. Various properties such as substance distribution, temperature, flow velocities, diffusion properties and much more can be imaged three-dimensionally. The Magnetic Resonance Velocimetry (MRV) allows for the in situ analysis of the local flow velocity of fluids. Such an analysis characterises mass transport properties and helps to validate or improve numerical predictions for porous media. However, the benefit of such validations is lowered by several problems worsening the measurement accuracy. This work addresses systematic errors and the influence of noise, which may reduce the accuracy of MRV measurements. Different techniques are described to minimise or correct displacement and phase errors. In particular, the so-called dual-Velocity ENCoding (VENC) technique is considered. It is described how the ratio between low- and high-VENC value should be chosen in order to obtain the highest possible improvement of the Velocity-to-Noise Ratio (VNR). A new multi-echo MRV sequence is proposed as a compromise between VNR, total measurement time, spatial resolution and displacement errors. For improved VNR, a dual-VENC encoding scheme was used with different velocity encoding steps for the individual echoes. The repetition time TR was optimised to achieve a further VNR improvement. The proposed MRV sequence was implemented on a 7 Tesla preclinical Magnetic Resonance Imaging (MRI) system and used to measure the three-directional flow velocity of water in an Open Cell Foam (OCF) structure and a honeycomb structure with three-dimensional isotropic spatial resolution. The velocity measurements performed for the OCF structure were used for cross-validation with Computional Fluid Dynamics (CFD) simulations. A technique is described to match MRV and CFD velocity maps and their agreement was evaluated by a similarity index. The influence of different artefacts on the similarity index is evaluated in detail. To better understand the origin of different artefacts, the surface and the inside of the OCF structure were analysed separately.