On the Mechanism of the Aramination Reaction of Hydroxy Derivatives of Anthraquinone

L. P. Kulev, V. V. Koyash

The aramination reaction of hydroxy derivatives of anthraquinone is of great importance for obtaining certain valuable dyes, primarily acid anthraquinone dyes [1, 2]. Mineral and organic acids (boric, hydrochloric, sulfuric, benzoic, etc.), as well as their salts [3], are used as catalysts that accelerate this reaction. In the opinion of some authors, the role of these substances is reduced to the fact that they give, with hydroxy derivatives of anthraquinone, unstable intermediate products, which then enter more readily into reaction with aromatic amines than do the hydroxy derivatives themselves. For example, in the case of boric acid the reaction proceeds through the stage of formation of an unstable metaboric acid ester [4, 5].

An experimental confirmation of such a course of the reaction is considered to be the fact that, in the condensation of phthalic anhydride with p-chlorophenol in the presence of sulfuric and boric acids, not pure quinizarin is isolated, but its compound in the form of dark-brown crystals [6]. The data of Dimroth and Faust [7] are also cited, according to which acetic-boric anhydride reacts with anthraquinone derivatives to give an acetic-boric ester, which then, on heating in vacuo, is converted into a metaboric ester. However, such an interpretation of the mechanism of the aramination reaction of hydroxy derivatives of anthraquinone is by no means admissible for all substances tested as catalysts of this reaction; for example, it is excluded for hydrohalic acids and some of their salts.

To this it should be added that aramination, as was noted long ago [8, 9, 10, 11, 12], proceeds more rapidly and more readily if the reaction involves not the hydroxyanthraquinones themselves, but their leuco compounds. Under these conditions, even with boric acid, the formation of esters of the structure indicated above is impossible. All this leads to the idea that, in elucidating the role of catalysts in aramination, account must be taken not only of the properties of the hydroxy derivatives of anthraquinone, but also of the properties of the aromatic amines interacting with them—above all, their capacity for complex formation. Indeed, the literature contains numerous examples showing that aromatic amines form complexes with the most varied substances, including those used as accelerators of the aramination reaction. Thus, for example, aniline gives molecular compounds with mineral and organic acids, salts, and many organic compounds [13–17]. If complexes of this kind take part in the aramination reaction of hydroxy derivatives of anthraquinone, it should be accelerated in the presence of substances that form these complexes with aromatic amines, since in complex formation the bond between the hydrogen and nitrogen atoms is loosened, as a result of which the amines become more reactive. To verify these considerations, we prepared, by methods described in the literature [17, 20], various complexes, the properties of which are given in Table 1, and carried out aramination experiments both with the prepared complexes and with mixtures of the individual substances constituting these complexes.

Table 1

In all cases the prepared complexes acted much more actively than mixtures of the components; a particularly noticeable difference was observed at comparatively low temperatures, for example, on heating in alcohol (Table 2).

Table 2

It is evident from the table that the prepared complexes accelerate the anilidation reaction more strongly than the individual catalysts. With the disappearance of the complex the reaction slows down. Thus, for example, aniline and acetic acid at low temperatures give a complex of the composition aniline–acetic acid. With increasing temperature the yield of the anilide derivative decreases noticeably, since the complex of the above composition is destroyed and acetanilide is formed, which already has considerably lower activity than the complex of aniline with acetic acid.

Convincing proof of the connection between the catalytic activity of substances and their complex-forming ability may be provided by experiments with dihydric phenols. Thus, pyrocatechol and hydroquinone, which form complexes with aniline [17], considerably accelerate the aramination reaction, whereas resorcinol, which does not give molecular compounds with the same amine, exerts almost no catalytic action in the anilidation of quinizarin (Table 3).

Table 3

A substantial influence on the activity of the catalyst should be exerted also by the stability of its complex with the amine. The stability of the complexes was determined as follows. Into a flask with a reflux condenser were introduced 1 g of the substance being tested and 25 ml of benzene. The mixture was heated on a boiling water bath for 10, 20, 30, etc. minutes; the residue insoluble in benzene was filtered off, washed with 25 ml of benzene, and dried in a vacuum desiccator to constant weight. The loss in weight after definite intervals of time characterizes the stability of the complex toward heating. Experiments carried out by us to determine the stability of certain complexes upon heating in benzene showed that the most stable of them exhibited the least activity in the aramination of hydroxy derivatives of anthraquinone, while the least stable, conversely, exhibited the greatest activity. For example, the complex of boric acid with aniline almost completely decomposes upon 6-hour heating in benzene, whereas the complex of the same amine with metaphosphoric acid, which is far inferior in catalytic action to the former, decomposes over the same interval of time by only 33–34%.

Table 4

Decrease in weight of the complex, %, as a function of heating time in minutes.

Despite the fact that the experiments on determining the stability of complexes are limited only to those catalysts which do not dissolve in benzene, they nevertheless testify quite convincingly to a close relationship between the catalytic activity of additives and the stability of the molecular compounds they form with aromatic amines.

The stability of the complex depends not only on the nature of the catalyst, but also on the nature of the amine. A substantial difference is observed, for example, in the stability of zinc chloride complexes with different amines.

Table 5

A very considerable acceleration of the reaction occurs in the aramination of leuco derivatives of hydroxyanthraquinones even under conditions when they

was carried out in the presence of weakly active catalysts.

With strong catalysts, for example with boric acid, the reaction with leuco compounds was accompanied by resinification of the product, whereas under milder conditions (in an alcoholic medium) with the same catalyst dianilide derivatives were obtained in good yield; these were almost absent when other accelerating additives were used. On heating leucoquinizarin in alcohol with the prepared salt (aniline + boric acid), we obtained a compound not described in the literature, apparently having the following structure:

The substance separates from alcohol in the form of beautiful lustrous black crystals.

Found, % nitrogen: 2.57. Calculated, % nitrogen: 2.60. Molecular weight (cryoscopic method): calculated—445.532; found—436.176.

The formation of this substance is a consequence of the loosening of the bonds of both hydrogen atoms of the amino group. Thus, boric acid exerts the strongest catalytic action in the aramination of oxy derivatives of anthraquinone, which may be explained by the following reasons. The coordination number of boron in the complexes it forms is, as a rule, equal to 4. The atoms and atomic groups surrounding the boron atom are located at three corners of a tetrahedron, the fourth corner of which remains free. Here a positive electric field of considerable intensity is created. Thus, if the radius of the boron ion is taken as 0.2 angstrom, the “ionic-field strength” will be equal to 75, i.e., much greater than, for example, for aluminum. Hence the tendency of boron compounds to form complexes with various substances, including organic ones, is understandable.

A. A. Grinberg points out [18] that the pronounced tendency of boron to form ammines as a result of interaction with ammonia, as well as its ability to give intracomplex compounds with hydroxyl-containing organic substances, is explained by the field strength. On this basis one may envisage the following mechanism of the catalytic action of boric acid in the aramination of oxy derivatives of anthraquinone. Boric acid, adding complexly to the aromatic amine, loosens the bonds of the hydrogen atoms in the amino group and makes them mobile.

At the same time it reacts with the hydroxyl groups of oxyanthraquinone, forming a boric-acid ester of the intracomplex type (at higher temperatures a metaboric-acid ester of the same type may be formed).

Both complexes are not stable under the conditions of the aramination reaction and readily decompose, liberating boric acid, which again enters into interaction with the amine and the oxy derivative of anthraquinone.

\[ \text{quinizarin--}2\,\mathrm{B(OH)_2} + 2\,\mathrm{ArNH_2} \;-\;\mathrm{HN\cdots B(OH)_3} \longrightarrow \text{diaminoanthraquinone} + 4\,\mathrm{B(OH)_3} \]

With other catalysts, for example hydrogen halide acids, the accelerating action is due only to their influence on the aromatic amine, in which the hydrogen of the amino group becomes mobile as a result of complex formation; the ethers of the above-mentioned “intracomplex” type cannot give such substances, which explains their comparatively low activity.

It should be borne in mind, however, that the closure of five- and six-membered rings (by analogy with alizarin lakes) must entail a considerable increase in the strength of the intracomplex salt. This should apply first of all to metal-containing catalysts, for example zinc chloride, which nevertheless, in its activity, proved to be very close to boric acid (Table 6).

Table 6

It remains to assume that zinc chloride here does not give compounds of the alizarin-lake type, while its catalytic activity is connected above all with the formation of reactive complexes with amines and with a strong condensing action, accompanied by the elimination of water at the expense of the hydroxyl of the hydroxy derivative and the hydrogen of the amino group.

The tendency toward closure of five- and six-membered rings is probably also shown by the hydroxy derivatives of anthraquinone themselves, for example alizarin, quinizarin, and purpurin, in which the hydrogen of the alpha oxy group is bound to the oxygen of the carbonyl by a hydrogen bond [19].

Although these rings are less strong than in alizarin lakes, fairly severe conditions are nevertheless required to rupture them. Therefore the aramination reaction of hydroxy derivatives of anthraquinone, even in the presence of strong catalysts, proceeds with satisfactory yields only under comparatively strong and prolonged heating.

Conversely, leuco derivatives of hydroxyanthraquinones react under any conditions somewhat more readily and rapidly, since, in the absence of carbonyl groups, they are not capable of forming strong rings.

It is highly important that the leuco compounds themselves also exert an accelerating action on the aramination reaction of hydroxy derivatives of anthraquinone. Thus, if a mixture of quinizarin, aniline, and leucoquinizarin is heated in boiling alcohol, the arylamino derivative is obtained in a much higher yield than with quinizarin alone.

Table 7

For a noticeable acceleration of the reaction, very small amounts of the leuco compound are sufficient. For example, to obtain an anilide derivative with a yield of up to 50%, only about 0.0023 mole of leucoquinizarin is required. Here, it must be assumed, oxidation–reduction processes take place, owing to which the arylamino derivative of leucoquinizarin, as a result of oxidation, passes into the ketone form, while the quinizarin itself, playing the role of oxidizing agent, is reduced to leucoquinizarin:

\[ \begin{aligned} &\text{[leucoquinizarin]} + \mathrm{Az{-}NH_2} \longrightarrow \text{[leuco arylamino derivative]}\\[4pt] &\text{[leuco arylamino derivative]} + \text{[quinizarin]} \longrightarrow \text{[arylaminoquinizarin]} + \text{[leucoquinizarin]} \end{aligned} \]

The leucoquinizarin formed in this reaction, in turn, undergoes aramination, and the process is repeated in the same order.

The favorable action of leuco derivatives of oxyanthraquinones is also evidenced by the fact that, in the preparation of certain acidic oxyanthraquinone dyes, it proved useful to add a little zinc dust to the reaction mixture; in its presence the oxyanthraquinone is partially reduced to the leuco compound and thereby promotes a considerable acceleration of the aramination reaction (see, for example, acid anthraquinone blue-black [19]).

Conclusions

1. The catalytic action of complex compounds of aromatic amines with mineral acids, salts of mineral acids, and dihydric phenols has been studied in the aramination of hydroxy derivatives of anthraquinone. It has been established that, in the presence of preformed complexes, the aramination reaction proceeds considerably faster than with a mixture of the substances entering into these complexes; moreover, the activity of the catalysts is inversely dependent on the stability of the complex to heating. The least stable complexes with aromatic amines are given by boric acid, which in all cases exhibits the strongest catalytic action.

2. It has been shown that the catalytic action of various substances is primarily due to their complex addition to aromatic amines, as a result of which the bonds between the hydrogen and nitrogen atoms are loosened and the amines more readily enter into interaction with the hydroxy derivatives of anthraquinone.

Literature

1. Bezzubets, M. K.; Rozina, V. S. ZhKhP, 21, 1152, 1948.
2. Chemistry and Chemical Technology, No. 3, 211, 1953.
3. German Patent 36150 (1894)—F., B. 4, p. 308.
4. Orlov, N. N. Advances in the Field of Dyes and Intermediates, State Technical Publishing House of Ukraine, Kiev, 1936.
5. Schmidt, Z. ang. Ch., 41, 80, 1929.
6. Vorozhtsov, N. N. Fundamentals of the Synthesis of Intermediates and Dyes, Goskhimizdat, 1950.
7. O. Dimroth; T. Faust, Ber., 54, 3020, 1921.
8. German Patent 92223 (1866)—F., B. 6, p. 320.
9. Grandmougin, J. pr. Chem., 76, 124, 1907.
10. Zahn; Ochwat, Lieb. Ann., 560, 462, 1928.
11. P. Friedländer; G. Schick, Z. ang. Ch., 3, 218, 1890.
12. German Patent 205149, Zbl., 477, 1909 I.
13. Beilstein’s Handbuch der organischen Chemie, 4th ed., vol. X/II [[unclear: page or section number]], 1929.
14. Rojkow-Schtcherbanow, Zbl., 22, 1854 I.
15. Reddelien, Lieb. Ann., 388, 165, 1918.
16. R. Kremann, Monatsh., 27, 93, 1906.
17. F. Mylius, Ber., 19, 999, 1886.
18. Grinberg, A. A. Introduction to the Chemistry of Complex Compounds, Moscow, p. 124, 1951.
19. Kogan, I. I. Chemistry of Dyes (Synthetic), ONTI, 1938.
20. Z. Spiegel; T. Spiegel, Zbl., 1611, 1904 II.

CORRECTIONS AND NOTES

PUBLISHING HOUSE OF THE ACADEMY OF SCIENCES OF THE USSR, vol. 102.