Skip to main navigationSkip to main content
The University of Southampton

Research project: In-situ calibration of cohesive zone models for composite damage

Currently Active: 

The success of the next generation of commerical aerospace, marine, automotive and civil engineering applications are predicated on the widespread use of composites in primary structures.  In common with the early history of fracture and fatigue in metals, the majority of the analytical capability for composites consists of empirical fits to large experimental databases to determine "knock-down" factors for static design and to construct stress-life curves or similar aids for "lifing".  The reliance on this predominantly empirical and high cost approach has negative consequences for the introduction of composites into new applications, or the introduction of new materials.  The damage tolerance and/or durability of composite structures is generally determined by a combination of multiple, interacting damage and degredation processes and the abiity to apply accurate computational models to predict performance is therefore highly desirable.

Historically there have been two principal approaches to modelling damage tolerance and durability beyond purely empirical fits:

  1. the bottom up approach of micromechanics, which explicitly models the individual damage modes, and
  2. the top-down approach of continuum damage mechanics (in various guises) which homogenizes the individual damage processes.

Micromechanic approaches have the advantage of a strong physical basis but have the failing that they struggle to cope with the complexity of multiple interacting damage modes and are not readily transferable to structural analysis tools such as finite element analysis (FEA).  Continuum damage approaches are readily integrated with FEA, but have limited predictive capability, particularly in cases where the underlying mechanisms change.  In addition they struggle to capture scaling as they do not incorporate the underlying material and microstructural length-scales.  Ideally the best elements of empirical, micromechanical and continuum damage approaches would be married.  Cohesive zone models (CZM) have been identified as a particularly promising means to achieve this.  CZM are capable of directly describing processes such as off-axis ply cracking, fibre micro-buckling and delamination yet they can be directly implemented in a continuum finite element framework.  In addition, they can cope with both initiation and propagation controlled failure processes, both of which typically operate.  However, in order for a CZM to be truly physically based, and therefore to have a broad predictive capability, it must be independently calibrated from the data t which it is applied.  The CZMs which have been applied to this class of problem have so far been fitted to the same data set they are used to describe.  This is largely due to the difficulty of independently measuring the parameters required in the presence of multiple interacting damage modes, at multiple length-scales and with relatively complicated stress/strain states.  Recent developments in micro-computed X-ray tomography provides a direct and highly informative solution to this problem.  Analogous to the three dimensional imaging acheived in medical applciations, material microstructures and associated damage processes may now be routinely imaged to spatial resolutions of 3-10 µm using laboratory-based micro-focus X-ray equipment, whilst highly coherent, spatially and spectrally homogenous beams available from synchroton sources provide for sub-micron resolution imaging with phase contrast enhanced edge detection.  With the potential for in-situ imaging (i.e. under load), the comprehensive internal mapping of damage and displacement/strain fields at microstructural length scales in three dimensions represents a major advance over established experimental investigation/model validation processes.

The goal of this project is to obtain data for in-situ strains and displacements using high resolution X-ray tomography by which to calibrate, validate and evaluate cohesive zone models capable of describing damage evolution, strength and fatigue life of laminated composites containing notches.

Share this research project Share this on Facebook Share this on Twitter Share this on Weibo
Privacy Settings