An experimental and computational investigation of the corrosion behaviour and mechanical performance of a bioabsorbable medical grade magnesium alloy
View/ Open
Date
2023-05-29Author
van Gaalen, Kerstin
Metadata
Show full item recordUsage
This item's downloads: 89 (view details)
Abstract
Magnesium-based medical implants have the potential to overcome several complications
that arise with permanent metallic implants. Magnesium-based implants are biodegradable
such that they can undergo absorption once their load-bearing function has been completed,
thereby avoiding a secondary removal surgery. However, several challenges must be overcome
as non-uniform and highly localised pitting corrosion mechanisms can lead to early failure
of implants. Despite this, the majority of experimental studies to date tend to consider bulk
measures of corrosion by measuring mass loss, while the role of localised surface corrosion
mechanisms have been largely ignored, or evaluated using qualitative approaches. As a con sequence, there is limited quantitative understanding on how pit formation (e.g. severity
and spatial distribution) affects the overall mechanical performance of magnesium-based al loys. The objective of this thesis is to investigate the relationship between spatial corrosion
performance with the corresponding mechanical integrity of a magnesium WE43 alloy for
orthopaedic implants by means of an experimental and numerical approach.
In this thesis, a three-dimensional automated detection framework, ’PitScan’ was developed
to systematically evaluate the extent and morphology of surface-based pitting corrosion of
cylindrical magnesium specimens undergoing corrosion through in-vitro immersion. PitScan
used a Python-based approach to analyse micro-computed tomography images (µCT) of
cylindrical specimens undergoing corrosion. It was used to establish relationships between
local pitting parameters and mechanical performance, which was determined through uniax ial mechanical testing. Additionally, a finite element surface-based degradation model was
implemented to further explore the correlation between the severity of localised corrosion
mechanisms and mechanical integrity. These experimental and computational approaches
were then used to investigate the influence of a plasma electrolytic oxidation (PEO) surface
treatment on the corrosion performance of a Magnesium alloy. Finally, the degradation model
was utilised to demonstrate its potential within the design development chain of degradable
implants.
The Pitscan algorithm was able to systematically identify pitting morphology on the corrod ing surface, enabling full spatial characterisation of pitting parameters, including pit density,
pit size, pit depth as well as pitting factor according to ASTM G46-94. It was found that
bulk measures of mass loss during corrosion were not suitable predictors of the mechanical
integrity of corroding magnesium specimens. Instead, PitScan showed that features linked
to the reduction of the cross-sectional area seem to be the best predictors for the remaining
strength. This thesis also provides the first quantitative evidence that a surface-based non uniform corrosion model could capture both the geometrical and mechanical features of a
magnesium alloy undergoing corrosion. By considering a wide range of corrosion scenarios,
it was demonstrated that parameters described in ASTM G46-94 showed weaker correlations
to the mechanical integrity of corroding specimens, compared to parameters determined by
Pitscan. Similar to experimental observations, the minimal cross-sectional area parameter
was the strongest predictor of the remaining mechanical strength (R2 = 0.98), with this re lationship being independent of the severity or spatial features of localised surface corrosion.
It was also shown that uniform degradation models are not suitable to predict the mechan ical performance of samples undergoing corrosion. This thesis also demonstrated that PEO surface treatment on the magnesium alloy continued to protect the samples from corrosion
throughout the entire corrosion process, and not just in the early stages of corrosion. Fur thermore, the phenomenological surface features were used to calibrate the surface-based
corrosion model to fully predict the mechanical performance of both unmodified and PEO
surface modified magnesium WE43. Finally, the surface-based corrosion model was used to
demonstrate how simulated design adaptations could optimize material usage, while main taining similar mechanical integrity during corrosion. Overall, this provides significant tech nical advances and enhances the scientific understanding of corrosion in magnesium alloys,
and could inform future work in this area.