Pyrene-based metal-organic frameworks (MOFs) have tremendous potential for various applications. With infinite structural possibilities, the MOF community often relies on simulations to identify the most promising candidates for given applications. Among thousands of reported structures, many exhibit limited reproducibility - in either synthesis, performance, or both - owing to the sensitivity of synthetic conditions. Geometric distortions that may arise in the functional groups of pyrene-based ligands during synthesis and/or activation cannot easily be predicted. This sometimes leads to discrepancies between in silico and experimental results. Here, we investigate a series of pyrene-based MOFs for carbon capture. These structures share the same ligand (1,3,6,8-tetrakis(p-benzoic acid)pyrene (TBAPy)) but have different metals (M-TBAPy, M = Al, Ga, In, and Sc). The ligands stack parallel in their orthorhombic crystal structure, creating a promising binding site for CO2. As predicted, the metal is shown to affect the pyrene stacking distance and, therefore, the CO2 uptake. Here, we investigate the metal's intrinsic effects on the MOFs' crystal structure. Crystallographic analysis shows the emergence of additional phases, which thus impacts the overall adsorption characteristics of the MOFs. Considering these additional phases improves the prediction of adsorption isotherms, enhancing our understanding of pyrene-based MOFs for carbon capture.