Materials Performance Centre by 2F53T0P

VIEWS: 9 PAGES: 9

									Materials Performance Centre

       Modeling Directions
          Crystal Plasticity Modeling
Prediction of intergranular strains
and mechanical properties
• Relevant to stress corrosion
  cracking in stainless steels
  and Ni-base alloys
• Good expertise in
  Manchester, validated by X-
  ray and neutron diffraction
• Need parallelization to
  allow larger microstructures
  and studies of permutations
  in reasonable timescales
         Grain Aggregate Modeling
Prediction of intergranular
stresses and strains
• Relevant to stress corrosion,
  and intergranular damage
  mechanisms
• Expertise developing in
  Manchester, using 3D
  microstructure data and
  diffraction-based validation
• Need to validate modeling
  approaches and address
  issues due to large model
  size.
               Damage Modeling
Prediction of microstructure
effects on damage development
• Relevant to stress corrosion
  cracking, for example
• Development of current
  work on crystal aggregates,
  derived from tomography
• Work done so far in
  partnership with other
  institutes
• Need to develop further in
  Manchester
            Image-Based Modeling
Dimensional Change of Graphite
• Models constructed from
  tomography data, with crystal
  anisotropy deduced from pore
  orientations
• Model validation against in-
  situ tomography of thermal
  dimensional change
• Aim to predict irradiation
  induced dimensional change
• Currently limited by model size
                Image Based Modeling
Issues                                       Example: fibre composite
•   How large a volume do you need to
    model?
•   What resolution mesh do you
    need?
•   As resolution of XMT systems
    improves, data sets and mesh sizes
    expand
•   Research Needs                           Model
     – Development of visualisation
       methodologies
     – Development of serial mesh
       generation
     – Any size mesh (so far up to 320 GB
       data set)
     – Development of parallelised FE code
     – Development of XFEM

                                             Stress Development
         Grain Boundary Modeling
Prediction of Diffusion and
Segregation
• Relevant to stress corrosion
  and sensitisation kinetics
• Requires molecular
  dynamics methods
• Currently little expertise in
  Manchester in this area, but
  development of capability is
  needed to support other
  work
• Work being done with
  collaborators
                  Flow Assisted Corrosion

• Iron oxidation to give Fe(II) or
  magnetite Fe3O4 at the            Cbulk
  internal metal-oxide interface                              2+
                                                           Fe Cs
                                   Water           2+
• Diffusion of soluble species                  Fe                          H2
• (Fe2+ and H2) across the porous
  Oxide layer
              FACrate K . .([ Fe2 ]eq  [ Fe2 ]0 )                 2+
• Dissolution and reduction of                                      Fe
  magnetite into solution                                        2+
                                                             Fe C0
• Removal of the soluble iron             Oxide
  species (and hydrogen),
  transported into the bulk of
  the flowing solution                               Steel
                                                                      H2
                  Other Areas
• Multiscale modeling
  – Integration of microstructural models with
    fracture propagation models
• Coupled modeling
  – Development of crack tip chemistry
  – Influence of residual stress on crack tip
    deformation

								
To top