As it was reported by C&EN , Japanese scientists have reached a new milestone in cyclodextrin research. Hidetoshi Yamada and his colleagues from Kwansei Gakuin University published an article in Science  describing the very first synthesis of small ring cyclodextrins of 3 and 4 glucopyranose units (CD3 and CD4) (Figure 1).
The most common cyclodextrins, as we know, are cyclodextrins of 6, 7 and 8 glucopyranose units (α, β and γ CDs), which can be all produced naturally by enzymatic conversion of starch. By 1965, Dexter French and his colleagues isolated 9, 10, 11 and 12 membered CD rings (δ, ε, ζ, η CDs), however, these are commercially not as useful as the 3 parent cyclodextrins. Since then, large ring CD research sprouted newer and larger cyclodextrins than before and new cyclodextrins up to 35 or more glucopyranose units are possible .
In the other way, the progress was much slower. It is long accepted, that αCD is the smallest naturally occurring cyclodextrin, smaller CDs were theorized to be impossible due to steric overlaps. It was 1994 when the 5-membered cyclodextrin (CD5) was first synthesized by Toshio Nakagawa et al . They were building step by step, a key intermediate was prepared via successive glycosidations which was cyclized and deprotected to give CD5.
The next step happened recently with the Japanese scientists. Since CD3 and CD4 have too small rings to permit the most stable conformations of glucopyranose units, Yamada’s main challenge was to create the glucoyranose ring conformationally counterbalanced between equatorial- and axial-rich forms. They solved the problem by a small bridge between the O-3 and O-6 of glucose (Figure 2). The bridge adjusts the conformation of the scaffold, shifting the conformation in a way to allow the formation of α-1,4 glycosidic bond. This worked so well, that it was easier to form CD3 than CD4. 
It will be interesting to test if they are capable of similar complexation as the regular cyclodextrins, however, they are likely not. A manuscript from 1994 by F.W. Lichtenthaler and his colleagues examined the small ring CDs by molecular modeling and suggested that CD3 will not have a cavity at all and CD4 will only have a partial hydrophobic cavity with 3 Å diameter. The smallest cyclodextrin with a real hole in it would be CD5 (3.9 Å). This work was presented in the 8th International Symposium on Cyclodextrins in Budapest, Hungary (Figure 3) .
 Celia Arnaud (2019) “Smallest cyclodextrins synthesized”. Chemical & Engineering News. April 15, 2019, pp. 10.
 Daiki Ikuta, Yasuaki Hirata, Shinnosuke Wakamori, Hiroaki Shimada, Yusuke Tomabechi, Yuri Kawasaki, Kazutada Ikeuchi, Takara Hagimori, Shintaro Matsumoto and Hidetoshi Yamada (2019) “Conformationally supple glucose monomers enable synthesis of the smallest cyclodextrins”. Science. April 11, 2019.
 Haruhisa Ueda and Tomohiro Endo (2006) “Large-ring cyclodextrins”. In: Helena Dodziuk (ed.) Cyclodextrins and Their Complexes: Chemistry, Analytical Methods, Applications. Chapter 13., Wiley, Weinheim, pp. 370-380.
 Toshio Nakagawa, Koji Ueno, Mariko Kashiwa, and Junko Watanabe (1994) “The Stereoselective Synthesis of Cyclomaltopentaose. A Novel Cyclodextrin Homologue with D.P. Five.”. Tetrahedron Letters, 35(12) p9. 1921-1924.
 Stefan Immel, Jürgen Brickmann, and Frieder W. Lichtenthaler (1995) “Small-Ring Cyclodextrins: Their Geometries and Hydrophobic Topographies”. Liebigs Annalen (Molecular Modeling of Saccharides), 6 pp. 929-942.
 Frieder W. Lichtenthaler and Stefan Immel (1996) “The Lipophilicity Patterns of Cyclodextrins and of Non-glucose Cyclooligosaccharides”. In: József Szejtli, Lajos Szente (eds.) Proceedings of the Eighth International Symposium on Cyclodextrins, Kluwer, Doordrecht, pp. 3-16.