Soil Desiccation / Swelling

Question

I am looking at a problem where heat treated ground is undergoing desiccation and volume reduction. 
 Is there a soil behaviour model that can model the deformation (settlement) of the ground with respect to its moisture content? Can this be coupled with a time dependant thermal analysis? From my reading so far, the thermal deformation seems to be linked to the expansion coefficient rather than any water driven off by heat. 
 Do thermal effects alter the moisture content? 
 Which behaviour model do I need to adopt to ensure the moisture content affect deformation (shrinking and swelling)? 
 Many thanks 
 Scot

Stefanos Papavasileiou · Accepted Answer

Dear Scot, 
 Currently, the only mechanism allowing to simulate this is through evaporation, with vapour diffusion. 
 In thermal tabsheet, the user can put value for diffusivity and enables vapour flux to simulate moisture loss by temperature gradient. You can also look at the link to the paper from ECSMGE-2019, section 4: pure.tudelft.nl/.../0499_ecsmge_2019_Bui_Casarella_DiDonna_Brinkgreve_final.pdf Note that the simulation in section 4 uses a user-defined flow model, however, the vapour diffusion is already available for all models. 
 In general, you can start with a simple Linear Elastic or Mohr-Coulomb model to see the behaviour. When heating, enabled vapour diffusion reduces saturation, so suction and mean effective stress (compressive) is increased. 
 So, the expected following phenomena should be: elastic compression (shrinkage), soil thermal expansion, and no increasing plastic shear deformation. If soil thermal expansion is small compared to the elastic shrinkage, then some small ground settlement should be noticed. 
 As always, the exact behaviour depends on the parameters used in the model.

Stefanos Papavasileiou · Answer

Dear Scot, 
 About your query: moisture is transferred by two mechanisms: water flow (unsaturated Darcy's law) and vapour flux created by temperature change (more precisely temperature gradient). So, even if the wet part is thermally inactive (no vapour created), the water pressure gradient would bring water from low suction to high suction zone. 
 
 To the second part, yes, the time-dependent process can be done via the fully coupled flow-deformation analysis, but be aware that such a computation is quite complicated especially if you are interested in combining both heating with water flow and deformation, as advection may be quite critical there, together with unsaturated soil. 
 
 We would be interested in hearing more information about your project, so feel free to create a service request so we can also follow-up in case you have further questions: https://apps.bentley.com/srmanager/ProductSupport

Soil Desiccation / Swelling

Top Replies