Hello,
I am modelling a vascular tissue (intima) cut by a sharp instrument in Abaqus. The scheme of my model in on the picture. The instrument will be translated and simultaneously oscillated with ultrasonic frequency. The tissue is cut just below the metal stent struts.
I can figure out the nonlinear properties of my tissue (Mooney-Rivlin coefficients deduced from experimental data of simple tension) and the loading conditions but the problem is - how to define damage properties for soft tissue in Abaqus? More precisely, I want to find out what “material behaviours” in material definition should I introduce except for the "mechanical-elasticity-hyperelastic".
All the "cutting" models I've found use “Johnson-Cook formulation” or “Ductile damage” but I think they are only for metals. “Traction-separations laws” can’t be implemented because they require the predefined crack path and it is not possible because I can’t predict this path.
This problem is more similar with a “bullet impact in gelatin” – the material is damaged by an instrument but the instrument can be translated in course of the impact https://www.sciencedirect.com/science/article/pii/S175161611630412X
I tried to make the research by myself but I still have several questions:
1. As I have the complete test data maybe it is sufficient for damage modelling to indicate elastic and plastic behavior of the tissue?
2. What “material behaviours” would you recommend me to define in this case?
Thank you in advance,
Nikita.
1) You can still model damage by cohesive elements (CE) even if you do not know damage graowth orientation. Use XFEM (e.g., in ABAQUS. It may help although it may require some expertise with CE based modeling.
2) You may want to check if your materials (tissues) are isotropic or anisotropic. If anisotropic, you may want to apply adequate consititutive laws (also avaialble in major FEA codes at least for deformation).
Hello, Mark,
1) Thank you, it is very helpful information! I will try do use CE.
2) In the simplest and most wide spread modelling of blood vessels they are assumed isotropic so it is my first assumption, at least until the validation of the model. The best constitutive equation will also be chosen (Mooney Rivlin is the simplest one but we are planning to compare them).
Hello Nikita, if you haven't solved your problem yet, just a note on the cohesive elements. You can insert cohesive elements on the element boundaries of your total domain, as mentioned in the link below, so that crack propagation happens in a random fashion instead of following a predefined crack path. Hope this helps.
galileo.pmi.lv/~tarasov/ricoh/index.html
- Badges
- Science topic