Preparation of Psicose, Sorbose and Tagatose

Reliable information relating to a ketose of formula I came first from the research of Ohle and Just.8 They prepared the first crystalline derivatives of D-psicose, namely l,2-isopropylidene-3,4-anhydro-D-psicose (V), and some of its esters by means of a Walden inversion in the detosyla-tion of 3-tosyl-l,2-isopropylidene- a -D-fructose (IV).12 Attempts to 

113 (1948)]. Some writers have used the name allulose instead of psicose, but there appears to be no good reason for changing the historical trivial name that was originally suggested by Ohle and Just. The name ribohexulose has been proposed by Wolfrom, Thompson and Evans as a systematic name for this ketose. 

The synthesis of D-psicose as a colorless sirup ([ ]% + 3.1° in water) by Steiger and Reichstein13 may be regarded as the first authentic preparation of this ketohexose. The Kiliani-Fischer cyanohydrin synthesis furnished D-allonic lactone (VII) from D-ribose. This lactone, on reduction with sodium amalgam, gave D-allose (VIII) which was transformed into D-psicose (I) by refluxing with pyridine. Pyridine had been introduced into the Lobry de Bruyn-Van Ekenstein reaction by Fischer, Danilov and their coworkers.13  

More recently, Wolfrom, Thompson and Evans14 obtained an amorphous ketohexose, with a specific rotation of +4.7°, by treatment of 

D-ribonyl chloride tetraacetate (IX) with diazomethane. The 1-diazo-l-desoxy-/cefo-D-psicose tetraacetate (X) which was formed was transformed into /ceio-D-psicose pentaacetate (XI) by refluxing with cupric acetate in acetic acid. Deacetylation with barium hydroxide gave D-psicose (I). 

---