Org. Synth. 1929, 9, 50
DOI: 10.15227/orgsyn.009.0050
Submitted by B. F. McKenzie
Checked by H. T. Clarke and R. Phillips.
1. Procedure
To a solution of 500 g. (1.46 moles) of cane sugar (Note 1) in 1 l. of water in a 2-l. flask is added 250 cc. of concentrated hydrochloric acid (sp. gr. 1.16). The flask is heated on a steam bath for twenty-four hours, during which time considerable carbonization takes place. The black solid is filtered off with suction and washed with 300 cc. of water. The filtrate is placed in a large evaporating dish on a steam bath and allowed to evaporate overnight. The black solid residue obtained on the following morning is ground to a powder and placed in a folded filter paper of 34-cm. diameter. This is placed in a 25-cm. funnel fitted with a water-cooled 12-l. flask as described on p. 375. The solid is extracted with 500 cc. of ether for six to eight hours. The ether is distilled and the residue (Note 2) fractionated under reduced pressure. The fraction distilling at 150–160°/15 mm. or 135–140°/10 mm. forms a rather dark liquid which does not completely solidify on cooling.
On redistillation under reduced pressure a fraction boiling over a range of not more than 2° (e.g., 137–139°/10 mm.) is obtained with very little loss; this fraction solidifies almost completely at 30°. The yield is 72–76 g. (21–22 per cent of the theoretical amount).
2. Notes
1. Equally good results may be obtained with starch; the mixture however, must be warmed more slowly as it is apt to foam at the outset.
2. When larger quantities of levulinic acid are to be prepared it has been found by the checkers to be more convenient to fractionally distil the first filtrate under reduced pressure, without evaporating to dryness and extracting with ether. In this case a considerable quantity of tarry residue remains in the distilling flask. The yields are equally good.
3. Discussion
The only practical methods for preparing levulinic acid depend upon the action of mineral acids upon carbohydrates, a reaction discovered by Grote and Tollens,1 who heated cane sugar with dilute sulfuric acid. The procedure described is essentially that of Conrad,2 descriptions of which frequently have appeared3 in the subsequent literature. Improved yields have been reported4 by digesting sucrose under pressure for one hour with dilute hydrochloric acid at 162° in the presence of water vapor. The use of distillation under reduced pressure was suggested by Kent and Tollens.5 Levulinic acid can also be prepared from starch6 and from glucose7 by the action of hydrochloric acid, and from furfuryl alcohol or hydroxymethylfurfural by the action of dilute mineral acids.8

References and Notes
  1. Grote and Tollens, Ann. 175, 181 (1875); 206, 226 (1880).
  2. Conrad, Ber. 11, 2177 (1878).
  3. Fittig and Wolff, Ann. 208, 104 (1881); Neugebauer, Ann. 227, 97 (1885); Seissl, Ann. 249, 272 (1888).
  4. Thomas and Schuette, J. Am. Chem. Soc. 53, 2324 (1931).
  5. Kent and Tollens, Ann. 227, 229, Note 2 (1885).
  6. Rischbieth, Ber. 20, 1773 (1887).
  7. Sah and Ma, J. Am. Chem. Soc. 52, 4881 (1930).
  8. Teunissen, Rec. trav. chim. 50, 1 (1931).

Chemical Abstracts Nomenclature (Collective Index Number);
(Registry Number)

cane sugar


sulfuric acid (7664-93-9)

hydrochloric acid (7647-01-0)

ether (60-29-7)


Furfuryl alcohol (98-00-0)


glucose (492-62-6)