BRXN / TOTB not recognized correctly, and extra B atom inserted in COF-1 output

[Ali-SE98]

I’m trying to generate the COF-1 structure using the built-in building blocks, but I’ve run into two related problems:

1- Building-block name mismatch

In both the pyCOFBuilder documentation and the supplementary information of the original paper, the boroxine ring is represented as BRXN and COF-1 is said to be represented as:

T3_BRXN_BOH2-L2_BENZ_BOH2_H_H-HCB_A-AA

However, in the codebase and examples, the building block is called TOTB (1,3,5,2,4,6-trioxatriborinane). When I try to use T3_BRXN_BOH2 the parser does not recognize it—even though BRXN is documented as a valid T3 core.

2- Incorrect atom assignment and extra B insertion

When I force the use of TOTB as the tritopic SBU, the output CIF puts nitrogen atoms instead of oxygen.

Additionally, the builder inserts an extra boron atom between the benzene ring and the boroxine ring—this atom does not exist in the actual COF-1 structure.

I have observed similar issue of extra atom between the two rings for borazine and triazine SBUs, as well as boronic acid condensations, so this may be a common parsing or template problem.

[Ali-SE98]

import pycofbuilder as pcb

cof = pcb.Framework('T3_TOTB_BOH2-L2_BENZ_BOH2_H_H-HCB_A-AA')
cof.save(fmt='cif', supercell = [1, 1, 1], save_dir = '.')

[lipelopesoliveira]

Hi @Ali-SE98 ,

Thank you for pointing out these inconsistencies. I will take a look at it this weekend.