Dataset for poster presentation: Magnetic Resonance imaging of water transport and drying-wetting cycles in cement pastes

This dataset supports the results included in the poster presentation ‘Magentic resonance imaging of water transport and drying-wetting cycles in cement pastes: Upscaling towards applications – water transport in C-S-H agglomerates studied by MRI measurements’, Magdalena Janota, PJ McDonald, DA Faux, A Gajewicz-Jaromin, at the 39th Cement and Concrete Science Conference, 9-10 September 2019, University of Bath, UK.

The durability of cementitious materials is closely related to water transport within the material. The dynamic microstructure of calcium silica hydrate (C-S-H) - the main hydration product of cement - is researched widely and may impact transport. The macro-scale redistribution of water during drying / wetting in different types of pores is investigated by Magnetic Resonance Imaging (MRI).

Traditional techniques such as gravimetry or permeability testing to measure sorption offer at best limited information on the distribution of water within a sample. On the other hand, magnetic resonance imaging (MRI) is a very powerful method not only to visualise where water is within a sample, but also, through spin relaxation time contrast to provide information on the locally filled pore size distribution. In this study, we use 1D and 3D SPRITE MRI methods sensitive to short relaxation time to separate and resolve with circa 0.5 mm resolution water in gel (nanometre) and capillary (micron) sized pores during repeated wetting and drying cycles of 60 mm size samples.

The study aims to explore the extent to which time-dependent microstructural changes occurring in response to water content changes cause anomalous sorption behaviour. A series of water distribution maps are shown along with preliminary analysis using a new transport model under development.

We observe that capillary and gel pores (T2 = 400 μs) are initially emptied. Smaller interlayer spaces first increase (150 μs) because larger pores collapse and because a residual surface layer of water is left behind in gel pores. In this sample, the drying is surprisingly uniform with position, especially given the signal loss decreases close to t0.5.

The conference proceedings are published by University of Bath in association with Cementitious Materials Group, Institute of Materials Minerals and Mining, London, UK.

https://researchportal.bath.ac.uk/en/publications/39th-cement-and-concrete-science-conference-2019

https://purehost.bath.ac.uk/ws/portalfiles/portal/197842011/39th_Cement_and_Concrete_Science_Conference_2019.pdf