On a Method for Analyzing Carboxylic Acids by the Electromotive Force of Their Reaction with Metals

L. P. Kulev

We¹ have investigated a series of organic compounds by the electromotive force of their reaction with sodium; moreover, in order to create identical conditions and obtain clearer regularities, the experiments were carried out in benzene solution. The results obtained showed that the electromotive force—or, more precisely, a certain average value depending both on the emf and on the electrical conductivity of the solution under examination—can serve not only as a criterion of the purity of certain substances, but also makes it possible to judge the presence in them of functional groups, their position in the molecule, etc.; that is, it may find application in resolving many questions connected with the structure of organic compounds.

However, the method we developed has a somewhat limited field of investigation, since a number of substances that are sparingly soluble in benzene or in other solvents similar to it cannot be studied; the electromotive force of such substances (for example, many carboxylic acids, their salts, amides, etc.) can be determined in solutions that preclude the possibility of using sodium as an electrode, and therefore other metals must be chosen for them and the method of determination somewhat modified. For this purpose the apparatus we had proposed earlier was reconstructed: the platinum plate was replaced by a platinum crucible, which serves simultaneously as the electrode and as the vessel for the liquid under examination. The crucible is placed in an ebonite socket, at the bottom of which there is a copper rod fitted with a spring. The metal cylinder is tightly closed with an ebonite lid, into the copper bushing of which a rod for piercing sodium is inserted (if this metal is used as the electrode); the lid is provided with a glass cutter-tube with sharpened ground edges. Contact is made through the bottom of the crucible and through the copper bushing of the ebonite lid, which is connected with the metal cylinder. The instruments for measuring the emf are the same as those we used in previous work, i.e., a voltmeter with a scale up to 3 volts and a millivoltmeter with a scale up to 16 millivolts (Keiser-Schmidt, Berlin. Ω 370).

In working with metallic sodium, the procedure is as follows: from a piece of metal 9 mm thick, a cylinder \((d = 7\ \mathrm{mm})\) is cut out with a glass cutter-tube; it is pierced through the opening of the bushing with a brass rod and, in this form, placed into the metal holder of the apparatus. By gently pressing the rod, the sodium is pushed out and immersed in the solution under examination, poured into the platinum crucible. The rod on which the metallic sodium is fixed is of such a length \((6.5\ \mathrm{cm})\) that the cylinder is pushed completely out of the glass holder, while being immersed

into the solution by no more than 3/4 of its height, so that the liquid is in contact only with the fresh surface of the sodium. It should be noted that before the experiment the glass cutter must be slightly moistened with benzene.

For the investigation of carboxylic acids in aqueous solutions, other metals were used instead of sodium (Zn, Fe, Sn, Pb, and Cu—“Kahlbaum” grade) in the form of rods 7.5 cm long and 3.5 mm in diameter.

DESCRIPTION OF THE EXPERIMENTS

0.05 g of the substance was dissolved in 50 cubic cm of freshly distilled distilled water; 5 cm³ of this solution was poured into a platinum crucible, which was placed in the ebonite socket of the cylinder. A metal rod, with its surface freshly cleaned immediately before the experiment, was immersed in the solution under examination (to a depth of 1.2 cm) through the opening of the copper bushing in the ebonite cover of the apparatus; the EMF was recorded at the moment the metal was immersed, then after 30 sec, and after one minute from the start of the experiment. In all experiments with acids, the maximum EMF was observed immediately upon immersion of the metal in the solution; this was followed by a rapid fall to a certain limit, which remained constant for a comparatively long interval of time; therefore, in our experiments attention was paid specifically to the mean EMF reading, i.e., after half a minute from the moment of its appearance. In repeated experiments, exactly coincident EMF values were usually obtained for zinc, iron, and lead; for tin and copper, however, more or less significant fluctuations were observed, which nevertheless did not differ very greatly from the mean value derived from a series of determinations. For example, experiments with normal butyric acid gave the following results (Table 1).

Table 1.

Before preparing the solution, the purity of the distilled water was checked, since upon immersion of metallic electrodes in it an electromotive force arose¹; for example, after repeated distillation of water (in glass apparatus), the following EMF values were obtained: with Zn—4.6; Fe—1.3; Sn—1.3; Pb—2.6; Cu—0.6.

In order to include in the scope of the investigation acids that are sparingly soluble in water, all experiments were carried out in aqueous-acetone solutions, i.e., the weighed portion of the substance was first dissolved in 5 ml of acetone, and the solution was then diluted to 50 ml with water; with metals, such a mixture gave the following EMF values: with Zn—8.4; Fe—3.3; Sn—1.9; Pb—3.2; Cu—1.2.

Freshly prepared acid solutions were tested at a temperature of 15–17°; no evolution of hydrogen bubbles or rise in temperature

when metals were immersed in the liquid (even with strong carboxylic acids) was not observed.

Although a far from complete series of organic acids has been studied, the experimental data obtained already make it possible to establish the presence of certain regularities characteristic of the entire group of compounds (Table 2).

Table 2.

Saturated monobasic acids

EMF in millivolts with metals: Zn, Fe, Sn, Pb, Cu.

Formic acid, as was to be expected, proved to be the most active with all the metals and differs sharply from the other members of the series; beginning with propionic acid, only among the isomeric valeric acids does the difference in EMF become noticeable; the butyric acids can be distinguished from one another only by the electromotive force with tin and copper. Of the unsaturated monobasic acids, solid crotonic acid was studied, for which the following results were obtained: with Zn—100 mV; Fe—50 mV; Sn—14.0 mV; Pb—50 mV; Cu—5.0 mV, i.e., with the introduction of a double bond into the molecule the activity of the acid increases somewhat, and by the character of the EMF CH₃CH=CHCOOH should be placed between propionic and normal butyric acids.

In dibasic acids the difference between the individual members of the series is more sharply noticeable, and their electromotive force is considerably greater (Table 3).

Table 3.

EMF in millivolts with metals: Zn, Fe, Sn, Pb, Cu.

It is highly characteristic that oxalic and malonic acids gave a large EMF even with tin and copper, by which they differ not only from the other dibasic acids, but in general from all the acids tested by us; conversely, ethylmalonic acid with the same metals, especially with tin, gives a negligible electromotive force, which is also observed for pyrotartaric, i.e. methylsuccinic, acid. Unsaturated dibasic acids, being stronger acids, in general behave in the same way as the corresponding saturated ones (Table 4).

Table 4.

All of them exhibit very low activity with tin and fairly considerable activity with copper; the cis-isomers differ comparatively little from the trans-isomers, and the presence of side radicals (cf. citraconic with maleic, and mesaconic with fumaric) has less influence on the magnitude of the EMF than in the series of saturated dibasic acids.

Monobasic hydroxy acids give a similar picture of the change in electromotive force compared with saturated acids of the same basicity; however, at the same time, an essential difference is also observed among them, namely: whereas the fatty acids hardly differ from one another, isomeric α-hydroxy acids with the same number of carbon atoms in the molecule show a large difference in the magnitude of the EMF with all metals (Table 5).

Table 5.

Dicarboxylic hydroxy acids are characterized by low activity with tin, i.e. they display the same peculiarity as some of the acids mentioned above (Table 6).

It is necessary to note that at the first moment of immersing the metal in solutions of these acids with lead and copper, an identical electromotive force arises (with the exception of malic acid) (with Pb—150 mv, with Cu—10–12 mv), which then decreases (after 30 sec) to those indicated in the tab-

beyond the limits, thanks to which it becomes possible to distinguish these acids from one another. In the series of tribasic acids, the same regularities are observed in the change of electromotive force as for dibasic acids, i.e., unsaturated acids and hydroxy acids are more active than saturated ones, while with tin and copper they give a comparatively negligible EMF (Table 7).

Table 6.

EMF in millivolts with metals: Zn, Fe, Sn, Pb, Cu

Table 7.

EMF in millivolts with metals: Zn, Fe, Sn, Pb, Cu

Aromatic carboxylic acids show similarity to the corresponding representatives in the aliphatic series, but the presence of the benzene ring in the molecule at the same time imparts a special character to the EMF of these acids; for example, C₆H₅COOH, as a representative of monobasic acids, resembles CH₃COOH, but differs from it in being more active toward Pb than toward Fe; phthalic acid, on the other hand, is very close in magnitude of electromotive force with all metals to ethylmalonic acid and differs very strongly from the first two members of the series of dibasic fatty acids (Table 8).

Table 8.

Aromatic hydroxy acids behave in a distinctive way: according to the EMF with Zn, Fe, Pb, and Cu they are arranged normally, i.e. the strongest, salicylic acid, comes first, then meta-hydroxybenzoic acid and, finally, para-hydroxybenzoic acid; with tin, however, the order changes and they must be arranged by activity as follows: \(m>p>o\). If one takes into account that an insignificant electromotive force with tin is characteristic of most dibasic acids, the behavior of monohydroxybenzoic acids with this metal may be represented as follows: the electronegative hydroxyl group situated in the ortho position to the carboxyl as it were replaces a second carboxyl group, as a result of which salicylic acid behaves with metals analogously to dibasic acids, namely phthalic acid. In meta- and para-hydroxybenzoic acids, owing to the greater separation of OH and COOH from each other, this influence is nullified, and the acids acquire the characteristic features of monocarboxylic acids, i.e. with tin their electromotive force is greater than with copper; moreover, para-hydroxybenzoic acid is inferior in activity to meta-hydroxybenzoic acid, since its hydroxyl and carboxyl groups are the most isolated from one another.¹ The radical \(\mathrm{CH_3}\) in ortho-toluic acid does not exert such an influence on the carboxyl as does the electronegative hydroxyl group; therefore \(\mathrm{C_6H_4(CH_3)COOH}\) is close to benzoic acid, while cumic acid, which has a side chain in the para position, is inferior in activity to both.

Mandelic acid differs substantially from the hydroxy acids of the fatty series, for which a considerable EMF with tin is characteristic. In its behavior with metals it approaches ortho-hydroxybenzoic acid, which apparently can be explained by the immediate proximity of the hydroxyl and carboxyl groups attached to the same carbon atom connected with the benzene ring.

Mellitic acid is the most active of all the aromatic acids we have investigated and occupies a special position in the magnitude of its electromotive force.

Alongside the carboxylic acids, some acid amides were tested (Table 9).

Table 9.

EMF in millivolts with metals

All of them are less active than the corresponding acids, but in the main they retain a similar order of change in electromotive force.

If the EMF values we obtained for carboxylic acids are compared with their electrolytic dissociation constants, a regular relationship is not always observed between them, as is evident, for example, from the following comparison (Table 10).

¹ A similar picture may also be observed when these acids are compared by their dissociation constants:

1. Benzoic acid . . . 0.0060
2. o-Hydroxybenzoic . . . 0.102
3. m-Hydroxybenzoic . . . 0.0087
4. p-Hydroxybenzoic . . . 0.00286, i.e. with increasing distance of the hydroxyl group from the carboxyl, the constant decreases, and for p-hydroxybenzoic acid it becomes smaller than for benzoic acid.

Table 10.

Whereas the dissociation constant characterizes the degree of dissociation of acids, and consequently makes it possible to judge their “strength,” the magnitude of the EMF in many cases changes less sharply than would be expected when comparing acids with one another (cf., for example, maleic and fumaric); and sometimes even a lower electromotive force corresponds to stronger acids (the saturated monobasic acids beginning with propionic). Such an absence of a direct dependence between the EMF and the dissociation constant is natural if one takes into account that the former depends on a whole series of quite special causes and is determined under conditions different from those of the electrical conductivity by which the latter is established. It should be noted that the solvent exerts a very substantial influence on the electromotive force of carboxylic acids.

Thus, from Table 11 it is evident that for monobasic acids a decrease in the electromotive force in the presence of acetone is characteristic; oxalic acid, on the contrary, shows considerably greater activity in aqueous-acetone solution than in pure water. This interesting phenomenon, of course, must still be studied in detail on a whole series of dibasic acids and with the use of various concentrations of aqueous-acetone solutions;

...but the fact that dissociation in the stronger acids of this series proceeds in two stages, i.e., according to the scheme:

\[ \mathrm{H_2R \rightleftarrows H^+ + HR^-} \]

\[ \mathrm{HR^- \rightleftarrows H^+ + R^{--}} \]

makes it likely to expect, for oxalic acid, a peculiar behavior with metals and other regularities associated with the appearance of electromotive force than, for example, for monobasic carboxylic acids¹. It is hardly the case here that acetone has an activating effect² on

Table 11.

the cleavage of the monovalent negative ion of a dibasic acid (according to the second scheme), but it is possible that $\mathrm{CH_3COCH_3}$, combining with molecules or ions of both monobasic and dibasic acids, forms complex complexes with varying degrees of activity toward metals³. It is probably with a similar influence of acetone that the previously noted peculiar behavior of certain acids with tin is connected; this was observed neither in pure water nor in aqueous-alcoholic solution for the same acids at the same concentration (Table 12).

Table 12.

Thus, the values of electromotive force that we have obtained for carboxylic acids, depending on a whole series of physical and chemical causes which often are absent or are of a fundamentally different character in determining the constants of electrolytic dissociation, cannot serve as a criterion for comparing these acids with one another in strength; at the same time, however, just like the dissociation constants, they are undoubtedly a constitutive property of the compounds, i.e., they express the peculiarities of their chemical structure. In some cases the change in emf proceeds parallel to the change in the dissociation constants, as is observed, for example, for acids of the aromatic series (cf. benzoic acid and the isomeric hydroxy acids); in others it is of an entirely different character (emf with tin), and this circumstance, it must be supposed, may make it possible—even with greater certainty than by other methods—to distinguish structural features and to establish the identity of a given acid.⁴)

CONCLUSIONS

1. When metals are immersed in aqueous solutions of carboxylic acids, an electromotive force arises that is characteristic of each representative of this class of compounds; tin proved to be especially sensitive to certain changes in molecular structure, namely, the emf with this metal decreases very strongly upon the introduction of:

1) side radicals, hydroxyl groups, and a double bond into the molecules of dibasic and tribasic acids;

2) hydroxyl groups into the molecule of an aromatic acid, especially in the position adjacent to the carboxyl group, and also of a second carboxyl.

For monobasic acids of the fatty series, on the contrary, the introduction of hydroxyl increases the emf (with the same metal) approximately 2–3-fold (see the hydroxy acids).

1. Each of the acids we investigated gives, with metals, values of emf so characteristic and, at the same time, reproducible with sufficient accuracy in repeated experiments that it becomes possible easily and quickly to establish the identity of a given acid.¹) If one takes into account that an exceedingly small amount of substance is required for the analysis (experiment has shown that for strong carboxylic acids a quite distinct emf pattern is obtained even when their content in solution is 0.001%), then the advantage of this method over many others is obvious. We carried out a series of trial experiments on the analysis of acids, and, despite slight deviations from the standard, previously established values of electromotive force, each time one or another representative of the group of compounds studied was unmistakably determined from the character of its change.

Example: for analysis there was given an aqueous-acetone solution of an unknown acid, for which the following emf values were obtained with metals:

with Zn 240–245; Fe—95–100; Sn—1.5–1.7, Pb—50; Cu—5–6.

From the table we find emf values almost exactly coinciding with those given:

with Zn—240; Fe—100; Sn—1.7; Pb—50; Cu—5.5,

which were obtained for aconitic acid, corresponding to the correct solution. Sometimes a less distinct picture is obtained, which was observed, for example, for certain saturated monobasic acids,

⁴) Acids poorly soluble in water and in aqueous-acetone solutions can probably be investigated in the form of sodium salts, as is confirmed by some recent experiments of ours not included in the present work.

having EMF values close to one another; in this case they can be distinguished from one another by their behavior in different solvents (water, aqueous-acetone, and aqueous-pyridine solutions), in which they already give an appreciable difference (see the EMF of isomeric valeric acids in water and in aqueous-acetone solution). It should also be noted that so far we have tested only five metals chosen at random, and it must be supposed that, with a larger number of them and with other combinations, still clearer regularities may be obtained both for entire homologous series and for individual representatives of carboxylic acids.

The analysis is considerably simplified if, in determining the EMF, one uses not separate portions of the solution, but one and the same portion (5 ml), observing a definite order (according to the electromotive series) when immersing the metals in it; under these conditions, of course, other EMF values are obtained, but likewise quite definite ones and characteristic for each acid.

Table 13

1. By the electromotive force with metals one can determine the degree of purity of certain substances; in this way, for example, minute impurities of acetic acid are easily detected in acetone and other liquids miscible with water and, in the absence of acid, giving an insignificant EMF.¹

In conclusion it should be pointed out that the method we are developing undoubtedly has a number of shortcomings both in the apparatus and in the very procedure for determining the EMF, and still requires further improvement. Nevertheless, the experimental data obtained by us indicate that determination of the electromotive force under new conditions, with the use of a series of metals, considerably broadens the field of investigation of those substances that could hardly be addressed when working with metallic sodium, and promises to yield no less interesting results in solving a number of questions connected, for example, with the structure of complex compounds and, especially, with the analysis of certain organic substances, first and foremost carboxylic acids.

Along with this, in further work special attention should be devoted to the study of a whole series of questions relating to the electrical conductivity of solutions, polarization at electrodes and its change upon addition of one or another organic substance,² the behavior of metallic alloys in solutions of these substances, etc., which will make it possible to illuminate more deeply and more fully many of the phenomena that we observed in preliminary experiments with aqueous and aqueous-acetone solutions of carboxylic acids.

References

1. B. V. Tronov and L. P. Kulev—Determination of the activity of hydrogen in organic compounds from the EMF of the reaction with sodium. Zh.O.Kh., vol. IV, issue II, 1933.

2. B. V. Tronov and L. P. Kulev—Determination of the electromotive force of reactions with metals as a method for investigating organic compounds. Izvestiya TII, 1936.

3. M. Leblanc—A Manual of Electrochemistry. 1931, p. 94 ff.

4. G. Meyer—Analysis and Determination of the Structure of Organic Compounds, pp. 359–403. Also contains an extensive survey of the literature on the analysis of carboxylic acids.

5. Walden,—Electrochemie nichtwässriger Lösungen. Leipzig. 1924.

6. Chem. Zentr. II, 211, 1916.

1. The determination of the degree of purity of distilled water should also be included here, as was reported above (see “Description of the experiments”). ↩↩↩↩

2. Izgaryshev, for example, observed a strong increase in polarization upon addition of methyl and ethyl alcohols. In this respect it would be interesting to test acetone as well, especially in the presence of carboxylic acids. ↩↩

3. The formation of complex molecules affecting the magnitude of the EMF was noted by us for certain systems in work with metallic sodium (B. V. Tronov and L. P. Kulev, “Complexes of nitro compounds of the benzene series with acids,” Zh.O.Kh., vol. V, issue 3). ↩