< Terug naar vorige pagina
Publicatie
Impact of surface oxides on steel on hydrogen absorption and desorption
Boek - Dissertatie
The problem of hydrogen assisted cracking (HAC) of steel has been known for over 140 years. However, it is still not solved nor fully understood. The presence of
atomic hydrogen in a metal can cause important ductility loss, known as “hydrogen embrittlement (HE)”, and can be introduced during the production and processing
of steel or during its use, referred to as internal hydrogen embrittlement (IHE) and hydrogen environment embrittlement (HEE), respectively. Most research on the hydrogen embrittlement of steel deals with the interaction of hydrogen with bulk microstructural features, such as grain boundaries and dislocations, whereas the first contact with hydrogen-containing environments occurs at the metal surface. Steel (when un-polarized) is always covered with a native oxide layer, varying in composition and thickness. Furthermore, in permeation experiments, a passivation layer is often allowed to form before the start of the hydrogen loading. The impact of these oxide layers on the hydrogen transport is, however, not fully understood and often ignored. To have a better understanding of this problem, the impact of surface oxide layers on the hydrogen diffusion needed to be determined. The main objective of this work was to provide experimental evidence concerning the effect of oxide layers on the hydrogen transport through steel. Model thermal oxide layers were used to demonstrate the importance of considering the surface characteristics when investigating hydrogen transport through metallic components. Several oxide types of various thicknesses were tested on various steel alloy substrates: pure magnetite, bilayer (inner magnetite and outer hematite) and magnetite enriched in chromium.
The hydrogen release was investigated by employing an electrochemical technique consisting of consecutive hydrogen charging on one side (on a blank surface) and discharging on the opposite (oxide covered) face. Both the oxide type and the oxide thickness were found to have an influence. Results demonstrated that thermal oxides can greatly limit hydrogen diffusion, with bilayers (hematite/magnetite) having a greater effect compared to pure magnetite layers of comparable thickness. Two main contributions limiting the diffusion through the magnetite layers were determined: a lower diffusion speed in the oxide layer and a barrier effect of the oxide-steel interface. For the pure magnetite layers, increasing the oxide layer thickness had a relatively limited effect, pointing towards the importance of the steel-oxide interface in the hydrogen transfer. This was not the case for the chromium enriched magnetite layer. The presence of chromium introduced defects in the oxide layer, resulting in a lower diffusion speed and causing the effect of the lower diffusion speed in the oxide to become dominant over the effect of the oxide-steel interface. Complementary information on the hydrogen release was obtained by performing potential mapping using Scanning kelvin probe force microscopy. This allowed the simultaneous monitoring of the hydrogen release through an oxidized and a bare section of the steel surface, enabling a more direct comparison. A clearly lower hydrogen release was obtained for the surface with a thermal oxide layer, confirming the results obtained with the electrochemical technique. Modelling showed that a lower diffusion speed in the oxide layer could result in the contrast observed in the potential maps. Additionally, the hydrogen uptake through thermal oxide layers was investigated using melt extraction. As expected from the results obtained for the hydrogen release, the presence of a thermal oxide resulted also in a slower hydrogen uptake, showing that oxide layers could act as an important barrier to the hydrogen uptake, although macroscopic defects in the oxides such as cracks and spallation can greatly reduce this capability.
These results showed the importance of taking the surface condition into account, especially considering that depending on the alloy and environmental conditions, a variety of oxide layers can develop on the surface. As such, it is important to characterize the surface in order to have a full description of the hydrogen transport through metallic components.
atomic hydrogen in a metal can cause important ductility loss, known as “hydrogen embrittlement (HE)”, and can be introduced during the production and processing
of steel or during its use, referred to as internal hydrogen embrittlement (IHE) and hydrogen environment embrittlement (HEE), respectively. Most research on the hydrogen embrittlement of steel deals with the interaction of hydrogen with bulk microstructural features, such as grain boundaries and dislocations, whereas the first contact with hydrogen-containing environments occurs at the metal surface. Steel (when un-polarized) is always covered with a native oxide layer, varying in composition and thickness. Furthermore, in permeation experiments, a passivation layer is often allowed to form before the start of the hydrogen loading. The impact of these oxide layers on the hydrogen transport is, however, not fully understood and often ignored. To have a better understanding of this problem, the impact of surface oxide layers on the hydrogen diffusion needed to be determined. The main objective of this work was to provide experimental evidence concerning the effect of oxide layers on the hydrogen transport through steel. Model thermal oxide layers were used to demonstrate the importance of considering the surface characteristics when investigating hydrogen transport through metallic components. Several oxide types of various thicknesses were tested on various steel alloy substrates: pure magnetite, bilayer (inner magnetite and outer hematite) and magnetite enriched in chromium.
The hydrogen release was investigated by employing an electrochemical technique consisting of consecutive hydrogen charging on one side (on a blank surface) and discharging on the opposite (oxide covered) face. Both the oxide type and the oxide thickness were found to have an influence. Results demonstrated that thermal oxides can greatly limit hydrogen diffusion, with bilayers (hematite/magnetite) having a greater effect compared to pure magnetite layers of comparable thickness. Two main contributions limiting the diffusion through the magnetite layers were determined: a lower diffusion speed in the oxide layer and a barrier effect of the oxide-steel interface. For the pure magnetite layers, increasing the oxide layer thickness had a relatively limited effect, pointing towards the importance of the steel-oxide interface in the hydrogen transfer. This was not the case for the chromium enriched magnetite layer. The presence of chromium introduced defects in the oxide layer, resulting in a lower diffusion speed and causing the effect of the lower diffusion speed in the oxide to become dominant over the effect of the oxide-steel interface. Complementary information on the hydrogen release was obtained by performing potential mapping using Scanning kelvin probe force microscopy. This allowed the simultaneous monitoring of the hydrogen release through an oxidized and a bare section of the steel surface, enabling a more direct comparison. A clearly lower hydrogen release was obtained for the surface with a thermal oxide layer, confirming the results obtained with the electrochemical technique. Modelling showed that a lower diffusion speed in the oxide layer could result in the contrast observed in the potential maps. Additionally, the hydrogen uptake through thermal oxide layers was investigated using melt extraction. As expected from the results obtained for the hydrogen release, the presence of a thermal oxide resulted also in a slower hydrogen uptake, showing that oxide layers could act as an important barrier to the hydrogen uptake, although macroscopic defects in the oxides such as cracks and spallation can greatly reduce this capability.
These results showed the importance of taking the surface condition into account, especially considering that depending on the alloy and environmental conditions, a variety of oxide layers can develop on the surface. As such, it is important to characterize the surface in order to have a full description of the hydrogen transport through metallic components.
Aantal pagina's: 129
Jaar van publicatie:2023
Toegankelijkheid:Embargoed