HySchool Webinar #2: Raymond Mushabe (UiB) & Alessandro Campari (NTNU)

Raymond Mushabe
PhD-student at UiB, Centre for Sustainable Subsurface Resources (CSSR) 

In-situ visualization of microbial hydrogen consumption using high-resolution PET-MRI

The efficiency of short- and long-term underground hydrogen storage UHS in subsurface porous media is one of the limiting technical challenges facing the renewable energy industry. The stored H2 is one of the most important electron donors for many subsurface microbial processes, e.g., microbialinduced sulphate reduction and methanogenesis, which can convert H2 to H2S and CH4, causing permanent gas loss and H2 contamination. Therefore, understanding the microbial H2 metabolisms is essential for estimating the storage and withdrawal efficiency in UHS and improving the selection criteria for future storage sites. A halophilic sulfate-reducing stain was used as the model bacterium to quantitatively assess the consumption of H2 in in 6 cm x 1.5 cm sand and glass bead packs. The bacterium can utilize H2 as electron donor and sulfate as electron acceptor producing H2S, for growth. Besides, the high accumulation of bacteria can form biofilms and cause pore-clogging. In this study, state-of-the art visualization techniques were utilized to study hydrogen consumption and bacteria growth in 6 cm x 1.5 cm sand and glass bead packs. A multi-modal magnetic resonance imaging (MRI)- positron emission tomography (PET) scanner was used to study both static and dynamic phenomena, respectively. Sand and glass bead packs were saturated with bacteria solution (a sulphate-reducer oleidesulfovibrio alaskensis), both without and in the presence of hydrogen. The whole experiment was conducted under anaerobic conditions for the bacteria to survive and grow. In-situ visualization provided insight into the dynamics of bacterial growth and hydrogen consumption rates: MRI provided information on the spatial fluid saturation at micrometer scale. PET provided fluid displacement dynamics during injection. of brine, nutrients and bacteria at high temporal resolutions. We, hence, observed bacterial growth and fluid flow redistribution at resolutions not previously used to study these phenomena at the core scale.

Alessandro Campari
PhD-student at NTNU

A Machine Learning Approach to Predict the Susceptibility of Materials to Hydrogen Embrittlement

Hydrogen is widely considered a promising energy carrier capable of mitigating the human impact on the environment while making the countries energetically independent in the long term. Nevertheless, safety aspects represent the major bottleneck for the widespread utilization of hydrogen technologies. Industrial equipment operating in a pure hydrogen environment is prone to a variety of material degradations. Hydrogen embrittlement (HE) is the best-known hydrogen-induced damage and manifests itself as a reduction in tensile ductility, fracture toughness, and fatigue performance of the affected materials. It may cause component failures at stress levels significantly below the nominal tensile strength of the material, often resulting in undesired releases of hazardous substances in the environment. The occurrence of HE relies on the synergy of several factors, such as the hydrogen concentration, the operating conditions (i.e., temperature and pressure), the level of internal and applied stress, and the microstructure and chemical composition of the material. However, the mutual influence of these factors is still difficult to evaluate, and this results in serious difficulties in planning inspection and maintenance activities of hydrogen technologies. In this study, the experimental data of tensile ductility tests carried out on several materials exposed to hydrogen under different operating conditions were analyzed through an advanced machine learning approach. This study aims to provide critical insights into the susceptibility to hydrogen embrittlement for various materials at different operating conditions. In particular, the embrittlement index was estimated to predict the likelihood of component failures. The model demonstrated accurate and reliable predicting capabilities. The outcome of this study can increase the understanding of hydrogen-induced material damages and facilitate the decision-making process in planning inspection and maintenance of hydrogen technologies.

Subscribe to upcoming HySchool Webinars here

Join on your computer, mobile app or room device
Click here to join the meeting
Meeting ID: 383 478 208 261
Passcode: H6nLjn
Download Teams | Join on the web
Learn More | Meeting options