In this work, the selectivity of two well-known cryptands, ([2.2.1]cryptand (221) and [2.2.2]cryptand (222), for sodium and potassium cations in both the gas phase and solution are compared to that of two familiar crown ethers, 18-crown-6 (18C6) and dibenzo-18-crown-6 (DB18C6). Intrinsic cation affinity of hosts, cavity size fitting, deformation/strain of hosts as well as solvent effect, for complexation of hosts with cations were calculated at DFT (BP86 and B97-1) level of theory using def2-TZVP triple zeta basis set. The resulting data show that the larger stability of [M ⊂ 221]+ and [M ⊂ 222]+ complexes relative to [M ⊂ 18C6]+ and [M⊂DB18C6]+ complexes is due to the larger intrinsic affinity of 222 and 221 hosts for both Na+ and K+cations. On the other hand, the data show that the size match is in disagreement with larger intrinsic affinity of [Na⊂Host]+ complexes compared to [K⊂Host]+ complexes studied here. The results also showthat the intrinsic affinities of all four hosts for Na+ are greater than those for K+, but the hosts 222, 18C6 and DB18C6 are K+- selective in solution due to the solvent effect. Thus, obviously the main reason for higher selectivity of 222, 18C6 and DB18C6 hosts for K+ compared to Na+ is the solvent effect and not a greater affinity and/or a better size match. The above conclusions were further confirmed by a good correlation between the experimental and calculated formation constants in solution.
