[CP2K-user] Conflict of results

[Monu Joy]

Hi there
I have cell-optimized a MOF structure (CYCU-3) using cp2k and it converged nicely but the resultant structure looks a little weird, which means the ligands are kind of twisted and I thought that could be the minimum structure. Just to double-check, I used another software CASTEP but that resulted in a reasonable geometry (with respect to the experimental structure). This is not for a comparison between cp2k and CASTEP but I would like to learn where I was wrong in the cp2k input description/model. The input geometry and the cut-off value (925) for both cp2k and CASTEP are the same but CASTEP used PBESol where cp2k used PBE. In both cases, the Hexagonal symmetry was fixed.
I know both packages are completely different in various means, especially pseudopotentials and plane waves, etc. 
Any help would be highly appreciated. 
P.S. All the input/output structures and files attached.     

-Monu Joy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210108/f11d5eda/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CYCU-3.zip
Type: application/x-zip-compressed
Size: 2873144 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210108/f11d5eda/attachment.bin>