B. V. Tronov, L. N. D’yakonova-Shul’ts and A. N. Kochneva

Reaction Rate of Certain Esters of Mineral Acids with Pyridine.

From the Laboratory of Organic Chemistry of the Siberian Technological Institute.

Measurement of the rate of cleavage of esters by tertiary amines is, in principle, one of the best methods for determining the strength of the bond of organic radicals with oxygen. With other methods one usually has to deal with complex processes proceeding in several stages, and the overall rate of the process depends on each of these stages. For example, the reaction, very convenient from the experimental point of view, of splitting off a hydrocarbon radical by the action of a hydrogen halide acid in fact represents a combination of at least two, and probably a larger number of, reactions.

Hantzsch¹ writes

\[ \begin{gathered} \mathrm{HX} \qquad\qquad \mathrm{X}\quad \mathrm{H}\\ \phantom{\mathrm{HX}}\ \vdots \qquad\qquad\ \ \backslash/\\ \mathrm{R-O-R + HX \longrightarrow R-O-R \longrightarrow R-OR \longrightarrow ROH + RX} \end{gathered} \]

This, of course, greatly hinders the use of experimental results for theoretical conclusions.

With tertiary amines a simple cleavage is quite probable:

\[ \begin{gathered} \mathrm{R}\\ \vdots\\ \mathrm{R-O-R + R'_3N \longrightarrow R-O-NR'_3} \end{gathered} \]

or

\[ \begin{gathered} \mathrm{X-O-R + R'_3N \longrightarrow X-O-N\equiv R'_3}\\ \vdots\\ \mathrm{R} \end{gathered} \]

and it is difficult to imagine an intermediate reaction. It would therefore be very interesting to investigate, with respect to the rate of cleavage by amines, as large a number of esters as possible, both simple and complex, with varied alcohol and acid radicals.

However, a number of obstacles hinder broad application of the method. First of all, amines are too weak a reagent to break the bond of a radical with oxygen in many esters. Among simple esters, only those in which one radical is strongly electronegative react at an appreciable rate (in this respect they approach complex esters).

¹ A. Hantzsch, Ber. 58, 612–41 (1925).

There is a reference in the literature¹) to esters of trinitrophenol. Trinitroanisole gives with pyridine a compound of the composition:

\[ \mathrm{CH_3OC_6H_2(NO_2)_3 + C_5H_5N} \]

Of the several structural formulas possible for this substance, one represents it as the product of addition of the trinitrophenyl methyl ether to pyridine, with rupture of the bond between methyl and oxygen:

\[ \begin{array}{c} \mathrm{C_5H_5N - O - C_6H_2(NO_2)_3}\\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \vert\\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \mathrm{CH_3} \end{array} \]

The electronegative character of the acid radical makes compound esters more active also with respect to tertiary amines. Thus methyl acetate, when heated with trimethylamine, gives acetic acid tetramethylammonium²):

\[ \begin{array}{c} \ \ \ \ \ \ \ \ \ \ \mathrm{O}\\ \ \ \ \ \ \ \ \ \ \ \Vert\\ \mathrm{CH_3 - C - OCH_3 + N(CH_3)_3 \longrightarrow CH_3 - C - ON(CH_3)_4}\\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \Vert\\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \mathrm{O} \end{array} \]

However, for comparing the strength of the bonds of various hydrocarbon radicals with oxygen by determining the rate of cleavage of these radicals under the action of amines, esters of organic acids prove to be of little use. We carried out experiments with the following substances:

\[ \begin{array}{ll} 1)\ \begin{array}{c} \mathrm{O}\\ \Vert\\ \mathrm{C - OC_2H_5}\\ \vert\\ \mathrm{C - OC_2H_5}\\ \Vert\\ \mathrm{O} \end{array} & 3)\ \begin{array}{c} \mathrm{CH_2 - C - N}\\ \vert\!=\!\mathrm{O}\\ \mathrm{C - OC_2H_5} \end{array} \\[3em] 2)\ \mathrm{CH_3 - C - CH_2 - COOC_2H_5} & 4)\ \begin{array}{c} \mathrm{CHOH - COOC_2H_5}\\ \mathrm{CHOH - COOC_2H_5} \end{array} \\[-0.5em] \ \ \ \ \ \ \ \ \ \ \ \ \vert & \\[-0.2em] \ \ \ \ \ \ \ \ \ \ \ \ \mathrm{O} & \\[2em] & 5)\ \begin{array}{c} \mathrm{CH_2 - COOC_2H_5}\\ \vert\\ \mathrm{COH - CH_2 - COOC_2H_5}\\ \vert\\ \mathrm{COOC_2H_5} \end{array} \end{array} \]

These compounds were mixed in equimolecular quantities with pyridine and left at ordinary temperature in sealed test tubes.

But even after standing for 3–4 years, formation of salts could not be established. Thus esters even of comparatively strong organic acids are too little active with respect to amines³) (of course, tertiary ones, since primary and secondary amines will react differently). The same was found for weak inorganic acids.

¹) Michele Giua, C. 1921 III 1271.
²) R. Willstätter, W. Kahn, 35, 2757 (1902).
³) With appreciable speed they react only on heating. (R. Willstätter, W. Kahn, Ber. (1902) 35. 2757.

Isoamyl nitrite did not react noticeably over 3 years. Methyl, ethyl, and isoamyl esters of boric acid did not give the corresponding salts even upon many hours of heating with pyridine in sealed tubes to 200–240°.

All this compelled us to limit the present work to the study of esters only of the strongest, almost exclusively mineral, acids. The following were taken: methyl nitrate, ethyl nitrate, isoamyl nitrate, dimethyl sulfate, diethyl sulfate, and the ethyl ester of benzenesulfonic acid. Pyridine was taken as the cleaving reagent.

It was necessary to study separately the products of the reactions¹, since in the literature on this question there were only very incomplete and partly contradictory data. For sulfates, salts with the methyl ester of nicotinic acid were known². With nitrates, some authors also obtained ammonium salts, e.g., nitric-acid benzylquinolinium from quinoline and benzyl nitrate³; others⁴ found that, in the interaction of esters of nitric acid and amines, decomposition of the ester occurs into nitrous acid and an aldehyde or ketone; thus reacted, for example, nitrates of glycerin and cellulose. The amine here is a catalyst. The contradiction must apparently be explained by the difference in the alcohol radicals. Polyatomic alcohols are in general oxidized appreciably more rapidly.

In the work of L. N. Dyakonova-Shulyats and A. S. Klokova it was established that the methyl, ethyl, isobutyl, and isoamyl esters of nitric acid with tertiary amines at ordinary temperature or on a sand bath give only ammonium salts, and thus the reaction can indeed serve to determine the strength of the bond of these hydrocarbon radicals with oxygen.

Greater difficulties subsequently arose in the actual measurement of the reaction rate.

For the anions \(\mathrm{NO_3'}\), \(\mathrm{C_6H_5SO_3'}\), and \(\mathrm{R--OSO_3'}\)⁵ exact quantitative determination is considerably more difficult than for halides. Therefore, in the present work, which is preliminary in character, we limited ourselves only to an approximate determination of the course of the reaction. A mixture of pyridine with the ester under study, sometimes with addition of a solvent, was poured into a graduated test tube or cylinder. In the case of slowly reacting substances the test tube was sealed.

The course of the reaction was followed by the volume of the precipitated product (all oils are heavier than the original mixture). Observation was aided by the negligible solubility of the salt formed, but some of the pyridine, of course, could be entrained into the lower layer. In some cases the separated oil was weighed. The experiments were carried out at ordinary temperature (16–20°). Sometimes it was necessary to cool the test tube with water. Compounds reacting slowly were investigated without solvent; rapidly reacting ones, chiefly in benzene solution, at

of average activity, experiments were also carried out both in solution and without solvent. The concentration of the solutions was for the most part 1 gram-molecule per liter, sometimes 2.6 gram-molecules per liter; 0.85 gram-molecule and 0.8 gram-molecule per liter. Pyridine and ester were always taken in equimolecular proportions.

In the following tables are given the results of individual experiments and the figures for the average relative activity of the esters, with half the reaction rate for diethyl sulfate taken as unity; it was studied under the most varied conditions, and since in \((C_2H_5)_2SO_4\) both radicals are split off with equal ease, for comparison the figures obtained have to be divided by two. In view of the low accuracy of the determinations, constants were not calculated; instead, the comparison of rates was made according to the time intervals during which the same percentage of substance reacts.

In the summary tables there are also given figures expressing the activity of haloid compounds with the same radicals¹.

EXPERIMENTS WITHOUT SOLVENT:

EXPERIMENTS IN BENZENE SOLUTION

EXPERIMENTS IN m-XYLENE:

EXPERIMENTS IN CARBON TETRACHLORIDE:

EXPERIMENTS IN ETHYLENE CHLORIDE

EXPERIMENTS IN ETHYLIDENE CHLORIDE

EXPERIMENTS IN ISOAMYL CHLORIDE

Experiments in Chlorobenzene

Experiments in Ethyl Ether

Experiments in Acetophenone

EXPERIMENTS IN ETHYL BENZOATE:

EXPERIMENTS IN ISOAMYL NITRATE:

EXPERIMENTS IN NITROBENZENE:

Relative rate of reaction of various esters and halogen compounds with pyridine

Relative rate of reaction of sulfates with pyridine in various solvents (conc. 1)

Conclusion

We have studied too few esters to make any theoretical conclusions.

The experiments presented do not make it possible to decide definitely what type of bond cleavage—ionic or neutral—takes place in the present case. Compounds of three radicals were investigated: methyl, ethyl, and isoamyl. If the rate of reaction for ethyl esters is taken as unity, and nitro esters and sulfuric esters, as well as alkyl halides, are taken with pyridine, then the following numbers are obtained:

Rate of reaction $\mathrm{RX} + \mathrm{C_5H_5N}$
with $X =$

Measurement of the rate of action of alcoholates on halogen compounds¹) and the rate of decomposition of simple phenyl ethers²) and acetic acid esters³) by hydrogen bromide gives:

The reaction of halogen compounds with amines is, as was indicated earlier⁴), typical of ionic bond rupture; interaction with alcoholates proceeds predominantly according to the neutral type; still more sharply is this type of decomposition expressed, as the numbers in the table show, in the decomposition of ethers by hydrogen halides.

Measurement of the rate of interaction of nitrates and sulfates with pyridine, judging by the ratios between methyl and ethyl, belongs to the ionic type, but with isoamyl esters fluctuations are observed.

The most interesting result of the work is the comparison of the mobility of various acid residues:

The activity of the NO₃ group has been taken as unity.

As is seen from the numbers given here, the series of activities is as follows:

\[ \mathrm{Cl} < \mathrm{NO_3} < \mathrm{Br} < \mathrm{I} < \mathrm{C_6H_5SO_3} < \mathrm{ROSO_2} \]

In rate of reaction with pyridine, nitrates somewhat exceed chlorides, but are strongly inferior to bromides.

The group C₆H₅SO₃ is close in its mobility to I.

Especially outstanding are the sulfates, in which rupture of the radical bond (the first one) with oxygen occurs tens of times more readily than even in iodide compounds. This is not in any connection with the dissociation constants of the corresponding acids. This extraordinary chemical activity must evidently also explain the unexpectedly strong toxicity of dimethyl sulfate. From the theoretical side, the activity of sulfates for the time being remains incomprehensible.

¹) M. Conrad, C. Brückner, Ztschr. phys. Chem. 4, 631—637 (1889). E. V. Tronov and L. G. Boldyreva. Work not yet published.
²) E. V. Tronov and L. V. Ladygina. The work is being printed in Ber. d. Deutsch. Chem. Ges.
³) E. V. Tronov and N. K. Subbotina. The work is being printed there.
⁴) E. V. Tronov, Zh. R. Kh. O. 58, 1278 (1926).

1. The greater part of the compounds were compared at a concentration of $1$ g-mol; $\mathrm{CH_3ONO_2}$ and $\mathrm{C_6H_5SO_2OC_2H_5}$ were taken at other concentrations and compared with a solution of ethyl sulfate of the same concentrations. ↩↩↩↩

2. R. Wolffenstein, Ch. Al. 17, 1247. ↩

3. Ya. I. Mikhailenko and N. A. Kudryasheva, “Reports on scientific and technical work in the republic,” 6, 60. ↩

4. A. Angeli, C. 1921, 4, 762. ↩

5. Sulfates, even the most active dimethyl sulfate, readily give up only one hydrocarbon residue, so that under ordinary conditions the reaction does not proceed as far as formation of the sulfuric-acid ammonium salt. ↩