https://www.sciencedirect.com/search?qs=penetration%20of%20cohesive%20elements%20upon%20compression%20in%20Abaqus%20explicit&show=25&sortBy=relevance

https://www.sciencedirect.com/search?qs=penetration%20of%20cohesive%20elements&show=25&sortBy=relevance

Why do you use 4 layers of cohesive layers? You should use only one. Then define two element-based surfaces for the other two materials and define a contact pair interaction between them.

Dear Luca Di Stasio ,

Thank you for your reply. I have tried a single element layer with a node to surface contact for the other two materials, but it didn't work too. by element based contact pair you mean general contact or node to surface algorithm? moreover, I have used the edit mesh to transform the single layer element to a single node layer. But I do not know how to connect the new geometry to the previous one. I mean in the assembly I see a cohesive layer with a finite thickness but in the mesh module, there is just a single node layer. And once I run the simulation, it can't recognize the tie constraints and the cohesive layer is in the air!

You need to define the interfaces of the two materials with two lines of nodes that are distinct but juxtaposed, i.e. in the same place. Then, create a layer of cohesive elements between these two lines of nodes. Define the two surfaces as element-based surfaces, and create a contact pair interaction.

If possible, use matching meshes at the interface, it makes things easier to control.

Dear Luca Di Stasio ,

Thank you for your reply, as you said, I moved the 3 layers at the same placed to be juxtaposed. Then defined an element based contact pair between the two layers and tied the single cohesive layer(did it with edit mesh option) to the the top and bottom layers with a node to surface Tie constraint. However, I got "abaqus system error code 3" error. I changed the mesh size of the parts as well as matching meshes but the problem didn't resolve. Finally, I suspected to the Tie constraints and suppressed all of them, and the explicit solver started to solve! do you have any suggestions for this issue?

Thank you

First, do not use 3 layers of cohesive elements, but only 1. Furthermore, you don't need to tie the cohesive elements to each surface: if the mesh is continuous, continuity of the solution is ensured. By continuity of the mesh, I mean the cohesive elements are defined between the nodes making up the surfaces of the two materials. Cohesive elements are a numerical tool to insert a cohesive behavior between two surfaces. You need to define a contact pair interaction to control the behavior in compression, to avoid interpenetration, and in free tangential sliding. Then, cohesive elements control the behavior in traction, i.e. surfaces moving away from each other, where they insert a cohesive behavior, which means they offer some (by some I mean: to be defined) resistance to the motion, i.e. they tend to keep the surfaces attached to each other. Furthermore, they offer the same kind of restraint to tangential sliding, by a mechanics that is different from that of friction (Coulomb or viscous).

Thus, tie constraints have no place in the definition of cohesive interaction, as they define the rigid link between surfaces. The tie constraint ensure continuity of stresses and displacements between the two surfaces, i.e. the seal the surfaces as if the mesh were continuous. For this reason, they are often used to join regions with very different meshes.