*3.1. Samples*

All solvents investigated in this study (2-pentanone, n-hexane, cyclohexane, n-octane, cyclooctane, and 1,3,5-triisopropylbenzene (TiPB)) were purchased from Sigma-Aldrich (Germany). Different binary mixtures of the mentioned solvents (cyclooctane/n-octane, cyclooctane/2-pentanone, n-octane/2-pentanone, n-hexane/cyclohexane, and TiPB/2 pentanone) were then prepared by mixing the two solvent components at different mass proportion with a help of a syringe. Due to possible evaporation issues during the preparation of the mixture, the estimated mass proportion was controlled by 1H high-field liquid-state spectroscopy measurements. The binary mixtures were sealed in a glass container in order to keep the concentration constant during the NMR measurement.

The MOF UiO-66(Zr) was purchased from Strem Chemicals (USA). According to the manufacturer, this MOF has a particle size of 0.2–0.5 μm, 1000–1600 m2 g−<sup>1</sup> BET, and a pore volume of 0.3–0.5 cm3 g−<sup>1</sup> [59]. Following purification at 220 ◦C overnight under vacuum conditions, the MOF powder was stored under argon atmosphere. The activate MOF retains its crystalline structure as demonstarted by the experimental diffraction patterns (Figure S4, in the Supplementary Materials). All further handling involving MOF samples was also performed under an argon atmosphere to avoid any possible water adsorption from the atmosphere.

To investigate the applicability of proton NMR relaxometry as a fast analytical tool to test the separation power of a certain MOF material, the above mentioned mixtures of the two solvents with a volume proportion of about 1:12 (unless else stated) was used. A total of 65 μL solvent was added for 10 mg of dry MOF. This solvent amount was chosen considering the total amount of solvent that the MOF can uptake. Shortly before starting the measurements, the needed amount of MOF powder was loosely placed in a glass bottle under an argon atmosphere and at room temperature. The mixture of solvents was then gently poured on the top of the MOF using a syringe. The bottle was then immediately closed with a cap sealed with an elastic band and then covered with Teflon coating to prevent the evaporation of hydrocarbons. The prepared system was then left for about 2 h at room temperature to reach an equilibrium state. This equilibration time was confirmed by monitoring the changes in the relaxation times and the 1H spectra of the solvent mixture after different contact times with the metal–organic framework (Figure S5, in the Supplementary Materials). Then a filter was used to remove the MOF and the left solvent mixture was measured by NMR. Each sample was weighed before and after NMR

measurements to be sure that no solvent evaporated within the frame of the measurement takes place.
