It is worth it to go first go over these structural differences so that their applications are clearer. Cucurbiturils are polymeric forms of glycoluril, whereas cyclodextrins are polymeric forms of glucose. The final cyclic forms of these molecules are vastly different. I draw your attention to the figure below: (1) notice the obvious shape difference between the two hosts. The cucurbituril is more or less a perfect ring (Faulkner and Kenwright actually describe it more appropriately as a barrel) $^{(4)}$, whereas the cyclodextrin can be thought of as a tapered cylinder; (2) notice that in cucurbiturils, there are carbonyl groups that are exocyclic to the main cavity of the host molecule, while cyclodextrins feature hydroxyl groups exocyclic to the cavity $^{(1)}$.
Though this may seem unassuming at first, this is a key difference in terms of the guests they can complex, and how they complex them. In aqueous solution, both cyclodextrins and cucurbituril cavities are filled with water, as confirmed by diffraction studies $^{(2)(3)}$. For neutral guests, the driving force behind the inclusion for both molecules is precisely the displacement of this water from the internal cavity and the inclusion of the guest. However, the story becomes entirely different when one talks about polar guests. Owing to the carbonyl functionality, the ion-dipole interactions between positive charges on the guest and carbonyl oxygens lining the cucurbituril cavity openings becomes a major force, whereas the peripheral hydroxyl groups in the cyclodextrins generally do not engage in strong interactions with the included guest (there are exceptions, generally these occur with amphipathic molecules). I have included a photo of a cucurbituril complex below with testosterone: notice the carbonyl group interactions.
Species |
log($K_d$) for association with $\ce{Ca^{2+}}$ (solvent: $\ce{H2O}$) |
cucurbit[6]uril $^{(5)}$ |
2.80 |
β-cyclodextrin $^{(7)}$ |
0.633 |
18-crown-6 $^{(5)}$ |
< 0.5 |
This means that broadly, cucurbiturils are far, far better at binding polar molecules than cyclodextrins are. And, due to their larger size, you can fit a whole host of molecules in their cavities. They are more versatile than cyclodextrins in this regard: the largest cucurbituril (with 10 monomers) has a cavity size of 12.7Å; the largest cyclodextrin has a cavity size of 9.5Å. You can even fit a whole cucurbituril inside a cucurbituril—this has actually happened, and is called a molecular gyroscope (figure below) $^{(8)}$, and is probably up there with my favorite supramolecular constructions ever.