Embed contact line pinning

[Theo Witkamp]

Hi All, 

I am currently working with two-phase in combination with embed. I use contact-embed.h of Mathilde Tavares and am getting some unexpected results. 

I append two videos where the setup is the same but the only difference is the bottom surface being either embed or the computational domain wall. 

I notice that in the case of embed, the thin tail is pinned and the film is getting pulled towards it, which is of course incorrect. The center of mass should practically stay static and the tail should merge into the bigger part of the film. 

I use a contact angle of 97.5 and the cs[] value at the embedded boundary is 0.96 which should not cause numerical pinning. 

Does the pinning occur because initially the contact angle is very low? 

I finally attach two other movies (gauss_.mp4) where one case pins in a non-physical way. What I can conclude is that indeed once the interface creates a two small contact angle with the embed, the contact line gets pinned. 

Best,

Theo