L. E. NEILAND, O. Ya. NEILAND, Academician of the Academy of Sciences of the Latvian SSR G. Ya. VANAG
Unknown
Submitted 1964-01-01 | RussiaRxiv: ru-196401.97663 | Translated from Russian

Full Text

L. E. NEILAND, O. Ya. NEILAND, Academician of the Academy of Sciences of the Latvian SSR G. Ya. VANAG

A NEW METHOD FOR PREPARING TETRONIC ACID

Tetronic acid (I) is the basis of many important natural substances, and both it and some of its derivatives possess physiological activity \((^{1,2})\). I is a highly reactive compound and is used in organic synthesis \((^{3-6})\).

However, the use of I is limited by the fact that, up to now, no convenient method has existed for its synthesis. Usually two methods are used to obtain I. One of them is based on the reduction of \(\alpha\)-bromotetronic acid, obtained by thermal cyclization in vacuo of \(\alpha,\gamma\)-dibromoacetoacetic ester \((^{7,8})\). According to the other method \((^{9,10})\), to obtain I the reaction of sodium or ethoxymagnesium malonic ester with chloroacetic acid chloride is used, leading to the formation of \(\alpha\)-carbethoxytetronic acid, which upon saponification and decarboxylation is converted into I. The methods mentioned are rather difficult, and the yields of I are low. \(\alpha\)-Substituted tetronic acids are usually obtained by thermal cyclization of the corresponding \(\alpha\)-substituted \(\gamma\)-bromoacetoacetic esters \((^{1})\). Attempts to cyclize \(\gamma\)-bromoacetoacetic ester (II) to tetronic acid by heating in vacuo were unsuccessful \((^{8})\), although this would be the simplest method for obtaining I. It was noted, however \((^{11})\), that upon distillation in vacuo of benzyl \(\gamma\)-bromoacetoacetate, a negligible amount of I is formed. It also proved impossible to cyclize II under the action of alkaline agents—aqueous solutions of caustic potash \((^{7})\) and ammonia. The reaction of II with sodium ethoxide or ammonia in a solution of absolute ethanol \((^{12})\) leads to elimination of hydrogen bromide and doubling of the carbon skeleton with ring closure; as a result, 1,4-dicarbethoxycyclohexanedione-2,5 is formed.

Strangely enough, we nevertheless succeeded in cyclizing \(\gamma\)-bromoacetoacetic ester (II) into tetronic acid (I) under the action of aqueous solutions of KOH or NaOH \((^{13})\). II was obtained by bromination of acetoacetic ester or by bromination of diketene followed by treatment of the resulting

Table 1

Conditions for the synthesis of tetronic acid

Starting substances solvent KOH conc., N ratio of moles of starting substance and KOH temp., °C duration, h Yield of III, based on starting substance, %
Diketene, Br\(_2\), abs. methanol CHCl\(_3\) 2 1 : 2 15 2 35
» » ethanol CHCl\(_3\) 2 1 : 2 15 2 41
» » propanol-2 CHCl\(_3\) 2 1 : 2 15 2.5 39
CHCl\(_3\) 2 1 : 2 15 3 0
Acetoacetic ester, Br\(_2\) Ether 2 1 : 2 10–15 3 38.5
Ether 3 1 : 3 10–15 4 42.5
Ether 3 1 : 2 32 3 38
Ether 3 1 : 2 23 3 41.5
\(\gamma\)-Bromoacetoacetic ester 2 1 : 2 20 3 34
0.1 1 : 2 15–16 12 30
5 1 : 2 20–25 2 24
Ether 2 1 : 2 5–15 4 40

bromoanhydride of bromoacetic acid with abs. alcohol (Table 1). Solution II in alkali was obtained by stirring either pure II or a solution of II in diethyl ether, and also in CHCl₃, with solutions of KOH or NaOH. The optimum cyclization conditions were: stirring the ethereal solution of II for three hours with a 2 N–3 N solution of caustic potash at a temperature of 23°. In this way an aqueous solution of tetronic acid was obtained. Isolation of I from aqueous solutions as such still remains an unresolved problem. For this purpose, repeated extraction with diethyl ether is usually used. To determine the reaction yield, we converted I into the sparingly soluble α,α′-isopropylidenetetronic acid (III) by the action of acetone (¹⁴). The yield of I thus determined is 30–42%, calculated on acetoacetic ester or diketene. The aqueous solution of I was also used directly for the synthesis of α-phenylazotetronic acid (¹⁵), phenylhydrazone I (⁷), lactone of 3-oxymethylquinolinecarboxylic acid (⁶), and anhydrobistetronic acid (⁷).

The mechanism of cyclization of II into I has not yet been elucidated. But the fact that, under the action of alkali, the bromoanhydride of γ-bromoacetoacetic acid is not converted into I suggests that the anion of γ-bromoacetoacetic acid is not capable of cyclization and that the anion of ester (II) is probably what undergoes cyclization:

\[ \mathrm{ \begin{array}{c} \text{[reaction scheme: anion of ester (II) cyclizes with elimination of } Br^- \text{ to a five-membered lactone ester,}\\ \text{then under } OH^- / C_2H_5OH \text{ gives an anionic intermediate, which on } H^+ \text{ affords tetronic acid (I)]} \end{array} } \]

\[ \mathrm{ I + CH_3COCH_3 \xrightarrow[-H_2O]{} \alpha,\alpha' \text{-isopropylidenetetronic acid } (III) } \]

Experimental Part

α,α′-Isopropylidenetetronic acid (III). Method A. To a solution of 4.35 g (0.05 g-mol.) of diketene in 20 ml of CHCl₃, with cooling (0°) and stirring, a solution of 8 g (0.05 g-mol.) of bromine in 10 ml of CHCl₃ is added; thereafter, likewise with cooling and stirring, 4.6 g (0.1 g-mol.) of abs. ethanol is added. After four hours the solution is decomposed with crushed ice. Solution II in CHCl₃ is washed with solutions of NaHCO₃ and sodium chloride, and over the course of half an hour, with stirring at 10–15°, is added to 50 ml of 2 N KOH solution. Stirring is continued for another 1.5 hours. To the aqueous layer are added 4.5 ml (0.05 g-mol.) of conc. HCl; it is treated cold with activated carbon and filtered. To the filtrate are added 5 ml of acetone. After two days crystals of III are separated. Yield 2.45 g (41%, calculated on diketene). M.p. 196–198°, lit. (¹⁴) 200–201°.

Method B. To a solution of 13 g (0.1 g-mol.) of acetoacetic ester in 20 ml of ether, with cooling and stirring, 16 g (0.1 g-mol.) of bro-

for an hour. It is left at room temperature. After 3 hours, 15 g of crushed ice is added. The ether layer is washed with solutions of NaHCO₃ and sodium chloride. With stirring, solution II in ether is added to 67 ml of 3 N KOH solution (0.2 gram-mole) at 20–24°. Stirring is continued for another 2 hours. The aqueous layer is acidified with 9 ml of conc. HCl (0.1 gram-mole) and treated with activated charcoal. To the filtrate is added 10 ml of acetone. After 24 hours the crystals of III are separated, washed on the filter with water and methanol. Yield of III: 4.6–5.1 g (38–42%, calculated on acetoacetic ester). M.p. 197–199°, lit. (¹⁴), 200–201°.

Riga Polytechnic Institute

Received
1 VI 1964

REFERENCES

  1. L. J. Haynes, J. R. Plimmer, Quart. Rev., 14, 291 (1960).
  2. R. Brodersen, A. Kjaer, Acta pharmacol. et toxicol., 2, 109 (1946); Chem. Abstr., 41, 1727 (1947).
  3. F. Micheel, G. Bode, R. Siebert, Ber., 70, 1862 (1937).
  4. G. F. Duffin, J. D. Kendall, Brit. Pat. 828 847, 1960.
  5. E. A. Fehnel, J. E. Stuber, J. Org. Chem., 24, 1219 (1959).
  6. Л. Э. Нейланд, Г. Я. Ванаг, ДАН, 146, 615 (1962).
  7. L. Wolff, C. Schwabe, Lieb. Ann., 291, 226 (1896).
  8. F. Reuter, R. B. Welch, N. S. Wales, J. Proc. Roy. Soc., 72, 120 (1939); Chem. Abstr., 33, 4198 (1939).
  9. E. Benary, Ber., 40, 1079 (1907).
  10. L. J. Haynes, J. R. Plimmer, A. H. Stanners, J. Chem. Soc., 1956, 4661.
  11. A. C. Röttinger, F. Wenzel, Monatsh., 34, 1867 (1914).
  12. W. Wedell, Lieb. Ann., 219, 93 (1883).
  13. Л. Э. Нейланд, О. Я. Нейланд, Г. Я. Ванаг, Author’s Certificate No. 164298.
  14. L. Wolff, W. Schimpff, Lieb. Ann., 315, 151 (1901).
  15. L. Wolff, W. Lüttringhaus, Lieb. Ann., 312, 155 (1900).

Submission history

L. E. NEILAND, O. Ya. NEILAND, Academician of the Academy of Sciences of the Latvian SSR G. Ya. VANAG