The design of functional materials capable of sequestering CO₂ from the atmosphere or capturing that emitted from human activities is one of the major challenges in materials science today. Porous organic polymers have already proven efficient for the selective adsorption of CO₂. Various combinations of building units and functionalities can create different topologies of porous organic systems. We have investigated the effect of two trigonal connectors (triphenylamine and triphenylpyridine) and three nitrogen–nitrogen linkages (azo, azoxy and azodioxy) on the geometrical and adsorption properties of porous organic polymers. The computational chemistry methods (calculation of binding energies, electrostatic potential maps and grand-canonical Monte Carlo simulations) were used to evaluate the effects of various structural patterns and four different layer stacking modes on the CO₂ adsorption. In the experimental part of the work, we synthesized and characterized (by IR and ¹³C CP/MAS NMR spectroscopy, powder X-ray diffraction, elemental analysis, thermogravimetric analysis and nitrogen adsorption–desorption measurements) a new azo-linked pyridine-based polymer which exhibits high surface area of 606 m² g⁻¹ and the CO₂ uptake of 32 mg g⁻¹, which matches nicely with the results of the GCMC study. The computational results indicated that azoxy and azodioxy linkages strongly promote CO₂ adsorption and this can be tentatively predicted from the calculated electrostatic potential values. Although the correlation between the calculated electrostatic potential values and CO₂ uptake is not always straightforward, it can provide a simple model for evaluating the CO₂ adsorption properties of porous organic polymers prior to synthesis.