Tag Archives: CDT

Digital twins and seeking consensus

A couple of weeks ago I wrote about our work on a proof-of-concept for a digital twin of a fission nuclear reactor and its extension to fusion energy [‘Digitally-enabled regulatory environment for fusion power plants‘ on March 20th, 2019].  In parallel with this work and together with a colleague in the Dalton Nuclear Institute, I am supervising a PhD student who is studying the potential role of virtual reality and social network analysis in delivering nuclear infrastructure projects.  In a new PhD project, we are aiming to extend this research to consider the potential provided by an integrated nuclear digital environment [1] in planning the disposal of nuclear waste.  We plan to look at how provision of clear, evidence-based information and in the broader adoption of digital twins to enhance public confidence through better engagement and understanding.  This is timely because the UK’s Radioactive Waste Management (RWM) have launched their new consent-based process for siting a Geological Disposal Facility (GDF). The adoption of a digital environment to facilitate a consent-based process represents a new and unprecedented approach to the GDF or any other nuclear project in the UK. So this will be an challenging and exciting research project requiring an innovative and multi-disciplinary approach involving both engineering and social sciences.

The PhD project is fully-funded for UK and EU citizens as part of a Centre for Doctoral Training and will involve a year of specialist training followed by three years of research.  For more information following this link.

Reference:

[1] Patterson EA, Taylor RJ & Bankhead M, A framework for an integrated nuclear digital environment, Progress in Nuclear Energy, 87:97-103, 2016.

Image: Artist’s impression of geological disposal facility from https://www.gov.uk/government/news/geological-disposal-understanding-our-work

 

Crack tip plasticity in reactor steels

Amplitude of temperature in steel due to a cyclic load with a crack growing from left to right along the horizontal centre line with the stress concentration at its tip exhibiting the peak values. The wedge shapes in the left corners are part of the system.

At this time of year the flow into my inbox is augmented daily by prospective PhD students sending me long emails describing how their skills, qualifications and interests perfectly match the needs of my research group, or sometimes someone else’s group if they have not been careful in setting up their mass mailing.  At the moment, I have four PhD projects for which I am looking for outstanding students; so, because it will help prospective students and might interest my other readers but also because I am short of ideas for the blog, I plan to describe one project per week for the next month.

The first project is about the effect of hydrogen on crack tip plasticity in reactor steels.  Fatigue cracks grow in steels by coalescing imperfections in the microstructure of the material until small voids are formed in areas of high stress.  When these voids connect together a crack is formed.  Repeated loading and unloading of the material provides the energy to move the imperfections, known as dislocations, and geometric features in structures are stress concentrators which focus this energy causing cracks to be formed in their vicinity.  The movement of dislocations causes permanent, or plastic deformation of the material.  The sharp geometry of a crack tip becomes a stress concentrator creating a plastic zone in which dislocations pile up and voids form allowing the crack to extend [see post on ‘Alan Arnold Griffith‘ on April 26th, 2017].  It is possible to detect the thermal energy released during plastic deformation using a technique known as thermoelastic stress analysis [see ‘Counting photons to measure stress‘ on November 18th 2015] as well as to measure the stress field associated with the propagating crack [1].  One of my current PhD students has been using this technique to investigate the effect of irradiation damage on the growth of cracks in stainless steel used in nuclear reactors.  We use an ion accelerator at the Dalton Cumbrian Facility to introduce radiation damage into specimens the size of a postage stamp and afterwards apply cyclic loads and watch the fatigue crack grow using our sensitive infra-red cameras.  We have found that the irradiation reduced the rate of crack growth and we will be publishing a paper on it shortly [and a PhD thesis].  In the new project, our industrial sponsors want us to explore the effect of hydrogen on crack growth in irradiated steel, because the presence of hydrogen is known to accelerate fatigue crack growth [2] which is believe to happen as a result of hydrogen atoms disrupting the formation of dislocations at the microscale and localising plasticity at crack tip on the mesoscale.  However, these ideas have not been demonstrated in experiments, so we plan to do this using thermoelastic stress analysis and to investigate the combined influence of hydrogen and irradiation by developing a process for pre-charging the steel specimens with hydrogen using an electrolytic cell and irradiating them using the ion accelerator.  Both hydrogen and radiation are present in a nuclear reactor and hence the results will be relevant to predicting the safe working life of nuclear reactors.

The PhD project is fully-funded for UK and EU citizens as part of a Centre for Doctoral Training and will involve a year of specialist training followed by three years of research.  For more information following this link.

References:

  1. Yang, Y., Crimp, M., Tomlinson, R.A., Patterson, E.A., 2012, Quantitative measurement of plastic strain field at a fatigue crack tip, Proc. R. Soc. A., 468(2144):2399-2415.
  2. Matsunaga, H., Takakuwa, O., Yamabe, J., & Matsuoka, S., 2017, Hydrogen-enhanced fatigue crack growth in steels and its frequency dependence. Phil. Trans. R. Soc. A, 375(2098), 20160412