Prof. V. I. Minaev

On Some Derivatives of Anthracoumarin

(Laboratory of Chemical Technology of Dyes and Fibrous Substances, Tomsk Technological Institute).

Introduction. As early as the time when the first chapters of the chemistry of anthraquinone and its substituted type of mordant dyes were being developed, i.e. the alizarins (polyoxy-anthraquinones), syntheses had already been found for certain complex “condensed” nuclei, consisting of cyclic anthraquinone complexes with new rings fused onto them. In this period of the alizarin direction in the chemistry of anthraquinone, which was created under the influence of the well-known “Liebermann and Kostanecki rule,” which predicted in advance the technically valuable mordant properties of polyoxy-anthraquinones, there were synthesized, for example, anthracene-dicarboxylic anhydride (I) and Isomer (II)—naphthanthraquinone.¹

\[ \text{[structural formula of anthracene-dicarboxylic anhydride]} \qquad (I) \]

\[ \text{[structural formula of naphthanthraquinone]} \qquad (II) \]

The first isomer is constructed according to the type of linear-external, and the second according to the type of angular-external, fusion of the benzene ring to anthraquinone.

To this second type of angular fusion of a new ring belong the historically much earlier obtained dioxo-anthraquinone-quinolines (III–β and III–α),² i.e. compounds with a heterogenic N-containing ring.

\[ \text{[structural formula of dioxo-anthraquinone-quinoline]} \qquad (III\text{–}\beta) \]

\[ \text{[structural formula of dioxo-anthraquinone-quinoline]} \qquad (III\text{–}\alpha) \]

Here, finally, must be assigned an entirely different type of condensed anthraquinone nuclei, in which (in contrast to the types of external annelation just indicated) new rings of aromatic and heterocyclic character are built up in the inward direction, i.e., are formed with the participation of the CO groups of anthraquinone (one or, sometimes, both). In this case, chronologically, the first to become known were: (IV)—anthracoumarin, obtained for the first time by Kostanecki¹ from boric acid and m-oxybenzoic acid under the influence of conc. sulfuric acid, acting simultaneously in an oxidizing and condensing manner; and (V)—yellow benzanil²:

\[ \begin{array}{cc} \text{(IV)} & \text{(V)} \end{array} \]

![structural formulas IV and V]

Of the derivatives of anthracoumarin, only hydroxy-substituted compounds have been obtained: (VI)—styrogallol or anthradaphnetin³ and (VII)—oxystyrogallol⁴, both known as mordant dyes, although without technical application:

\[ \begin{array}{cc} \text{(VI)} & \text{(VII)} \end{array} \]

![structural formulas VI and VII]

Subsequently, and precisely in the last period of the greatest, most abundant development of the chemistry of complex derivatives of the anthraquinone series⁵, 10 new types of rings were condensed onto the anthraquinone complex in the direction of ring closure onto the CO group. First of all, in this series of compounds one must mention (VIII)—benzanthrone, and then compounds with nitrogen-containing five-membered rings: pyridine-anthrone, (IX)—pyridone-anthrone, or pyrimidine-anthrone, pyrimidone-anthrone, pyridazone-anthrone, and

oxazone-anthrone, and also compounds with five-membered rings: thiazole, pyrazole, and thiophene rings (X).¹

\[ \text{(VIII)} \qquad \text{[structural formula: anthraquinone nucleus fused with a five-membered ring containing } \mathrm{CH},\mathrm{CH},\mathrm{C},\mathrm{CO}\text{]} \]

\[ \text{(IX)} \qquad \text{[structural formula: anthraquinone nucleus fused with a five-membered ring containing } \mathrm{CO},\mathrm{CH},\mathrm{NH},\mathrm{C}\text{]} \]

\[ \text{(X)} \qquad \text{[structural formula: anthraquinone nucleus fused with a thiophene ring bearing } \mathrm{C_6H_5}\text{ and } \mathrm{S}\text{]} \]

The similarity of structures (V) and (X) is self-evident, since in the present case the question concerns only the furan ring in structure (V) and the thiophene ring in (X), etc.²

Some of these complex condensed anthraquinone nuclei are starting materials for obtaining acid dyes for wool from them; in this case the arylamido-substituted derivatives of these are subjected to sulfonation. An example of dyes of this type is alizarin-rubinol R [By], formed upon sulfonation of 4-n-toluidino-1-methyl-anthrapyridone (XI).³ In the same way, these same compounds, or similar compounds with two anthraquinone nuclei and connected with one another also through the intermediary of an NH group, give coloring substances of a blue character; here one may assign, first of all, the compound found by Tomaszewski (XII)—red alolein [By].

\[ \text{(XI)} \qquad \text{[structural formula: anthraquinone nucleus fused with an } N\text{-methyl pyridone ring and bearing an } \mathrm{NH}\text{-aryl substituent with } \mathrm{CH_3}\text{]} \]

\[ \text{(XII)} \qquad \text{[structural formula: two anthraquinone nuclei, one fused with an } N\text{-methyl pyridone ring, connected through } \mathrm{NH}\text{]} \]

It should also be noted here that the low coloration of these complex compounds, which we must ascribe, firstly, to the strong bathochromic properties of the carbonyl chromophore inherent in pyridone-anthrone and, secondly, to attribute these same properties especially to the CO group of the pyridone ring, which in this sense occupies a very favorable position, being situated in relation to the anthra-

¹ Cf. V. I. Minaev, “On certain derivatives of 1,4-chloro-methyl-anthraquinone and 1,4-dichloro-anthraquinone-carboxylic acid,” Izv. Kiev. Politekhn. Inst., 1912, I—Dissertations.
² Cf. furan, thiophene, pyrrole—Berichte, Kr. uch. organ. khimii, 3rd ed. (1906), p. 365.
³ DRP 192,201 (cf. Frdl., IX, 732). Chemistry of dyestuffs, Georgievich and Grandmougin, p. 324.

… to the quinonoid nucleus also, as it were, in the carbethoxy group arranged [[unclear: words obscured]] (XIII). We note here, for the establishment of the constitution and for comparison, that in an entirely similar position the CO group of the newly formed ring is found in the pyrimidone-anthrone—(XIV) and anthra-coumarin—(XV) complexes.

\[ \begin{array}{ccc} \text{(XIII)} & \text{(XIV)} & \text{(XV)}\\ \text{structural formula} & \text{structural formula} & \text{structural formula} \end{array} \]

That such a position of the new CO group in the [[unclear: word obscured]] ring does indeed affect the lowering of color in anthraquinone condensed nuclei is confirmed by the fact that the dye (XVI), constructed from pyrimidone-anthrone in complete analogy with the above-mentioned red algol dye, likewise found by Tomaszewski, gives on cotton only colors of a weak reddish color, whereas [[unclear: several words obscured]] the complex, despite the presence of the CO group in the newly formed ring in the same position, possesses weakened dyeing properties, and the \(\beta\)-anthraquinonyl-4-amino-phenyl-pyrazol-anthrone (XVII), obtained by me together with Ullmann,\(^{1}\) gives from an alkaline hydrosulfite vat colors on cotton only of a reddish-brownish color, while the isomeric \(\alpha\)-anthraquinonyl-4-amino-phenyl-pyrazol-anthrone\(^{2}\) gives quite weak, technically low-value dyeings.

\[ \begin{array}{cc} \text{(XVI)} & \text{(XVII)}\\ \text{structural formula} & \text{structural formula} \end{array} \]

\(^{1}\) R. Ullmann and W. Minajeff, A. 338 (1912), 217—221.
\(^{2}\) Ibid., 220.

Theme of the work. The available factual material concerning the formation, from the complex condensed anthraquinone nuclei listed above, of new derivatives having the character of acid dyes for wool or vat dyes, belonging to the general class of “indanthrone” dyes, permits one to discern, in the original condensed nucleus, the chromogen, in whose structure is found, if not the sole, then the principal chromophore, predetermining the color both of the chromogen and of the future dye. Undoubtedly also an auxochromic \(\mathrm{NH_2}\)-group, free or arylated—\(\mathrm{-NH.C_6H_5}\), \(\mathrm{-NH.C_6H_4.CH_3}\), \(\mathrm{-NH.C_6H_3.CH_3.SO_3H}\), etc., down to \(\mathrm{-NH-(C_6H_3<^{CO}_{CO}>C_6H_4)}\), …, exerts an influence on the color, shifting the absorption bands toward the red and farther.

The aryl substituent, e.g., especially frequently encountered here—\((\mathrm{NH})\)-\(\mathrm{C_6H_4.CH_3}\) \((n)\), serves as the site for the introduction of the \(\mathrm{SO_3H}\)-group upon subsequent sulfonation, which ultimately converts the synthesized product into a dye sulfonic acid, whose color, owing to the aryl substituent at the \(\mathrm{NH}\)-group, is lowered according to the general rules slightly toward the red, while the \(\mathrm{SO_3H}\)-group has no appreciable influence on the color.

On the basis of the considerations expressed here (and above), one may a priori suppose that from anthracoumarin, since in this respect it has not yet been investigated by anyone, acid dyes of the condensed anthraquinone series should be obtained, for example, of the following general type:

\[ \begin{array}{c} \begin{array}{ccccc} & & \mathrm{CO} & & \mathrm{O} \\ & \diagup & & \diagdown & \vert \\ \mathrm{CH} & & & & \vert \\ \Vert & & & & \vert \\ \mathrm{C} & & & & \mathrm{C_6H_2} \\ \diagdown & & & \diagup & \alpha \\ \mathrm{H_4C_6} & \diagdown & \mathrm{CO} & & \vert \\ & & & & \mathrm{HN{-}Aryl{-}SO_3H} \end{array} \\[0.5em] \text{(XVIII).} \end{array} \]

Moreover, it is of course possible to place the entire imido-aryl substituent at another position of the right or left benzene ring in the anthracoumarin complex; consequently, here not only homologues but also isomers are possible. Since sulfonic acids of this kind in the anthracoumarin series have not yet been described by anyone, we set their condensation, the investigation of their properties, and the verification of the theoretical assumptions as one of the further tasks suggested here.

Secondly, it is also a priori beyond doubt for us that from anthracoumarin there may be obtained dye substances with vat properties, the color of which should be similar to the color of red alizarin dyes. The synthesis of such dyes and the verification of the theoretical assumptions constitute our second task. In the present case, instead of a simple \(\mathrm{NH{-}Aryl}\) substituent to the anthracoumarin chromogen, one must have in view a more complex imido-anthraquinonyl substituent, which itself carries a new quinonoid chromophoric grouping, influencing the color as a whole, apparently, not very significantly,\(^{1}\) but having the essential purpose of imparting to the entire complex the property of being reduced to a leuco product soluble in aqueous alkalis, i.e., of giving a “vat.” As an example one may point to

\(^{1}\) Cf. structures XI and XII,—both substances are red in color, despite the fact that the former does not possess a second quinonoid chromophore.

such a compound (XIX)—β-anthraquinonyl-4-amino-1-anthracoumarin, and (XX)—the probable product of its reduction in an alkaline vat:

\[ \text{(XIX) — structural formula of β-anthraquinonyl-4-amino-1-anthracoumarin} \quad \longleftrightarrow \quad \text{(XX) — structural formula of its alkaline-vat reduction product} \]

\[ \text{(β) Anthraquinone} \]

It goes without saying that in this series of vat dyes isomers may also be constructed, when anthraquinone is attached to 4-amino-anthracoumarin in the α-position, or when the entire imino-anthraquinonyl substituent is joined to anthracoumarin not in position 4, but in a new position, for example in 2, 5, or 6, etc.

In the systematic investigation outlined, it appears interesting to couple two molecules of anthracoumarin at position (4) into di-anthracoumarinyl.^1)

Theoretically, a comparison of the structural formulas of anthracoumarin (or of its di-oxy-substituted citropyronol) with cærulein is also very interesting; both are mordant dyes, while the latter, in addition, is capable of passing from hydrosulfite in a strongly alkaline medium into a vat and of giving, on cotton, beautiful green-colored dyeings. This latter fact was noted long ago, and even then the supposition was expressed that cærulein should be regarded as a derivative of anthraquinone. At the same time, the structural formula of cærulein had not been precisely established, and, as the most probable, the formula of Orndorff and Brewer—(XXI)^2)—was accepted. On the basis of comparisons and of all the facts expressing the chemical relations of cærulein, we believe that the structure (XXII) should be assigned to it:

\[ \text{(XXI)} \qquad \text{structural formula of cærulein according to Orndorff and Brewer} \]

\[ \text{(XXII)} \qquad \text{proposed structural formula of cærulein} \]

^1) In addition, all this is proposed to be extended to the derivatives of anthrathio-coumarin, which should condense from quinaldine and m-mercapto-benzoic acid by means of conc. \( \mathrm{H_2SO_4} \). Anthracoumarins should be regarded as compounds analogous in their structure to anthapyridones and anthracoumarins; the significance here of the replacement of the links \( {-}\mathrm{NH}{-} \), \( {-}\mathrm{O}{-} \), \( {-}\mathrm{S}{-} \) by one another may be seen from the analogous replacement of the \( ({-}\mathrm{NH}{-}) \) group in indigo blue by \( ({-}\mathrm{S}{-}) \) in thioindigo red and, finally, by \( ({-}\mathrm{O}{-}) \) in coumaran-indigo.

^2) See Chemistry of Dyes, Georgievich and Grandmougin, p. 278.

The first of these formulas clearly reflects only the origin of cerulein from gallein; the second, in addition to this latter circumstance, establishes the simultaneous affiliation of cerulein with phthalein (carbo-quinoid chromophore) and anthraquinone dyestuffs; at the same time, here there occurs a kind of superposition of two carbo-quinoid chromophoric groupings upon one another, from which it is natural to expect a strong lowering of the absorption bands of this compound toward the red (and cerulein, indeed, is colored the complementary green color).

Structural formula (XXII) represents, as it were, anthracoumarin (more precisely—strictly speaking, i.e., di-oxy-anthracoumarin), to which a new quinoid ring has been fused.\(^1\)

For the realization of the tasks indicated above, it was first of all necessary for us to synthesize the hitherto unknown chloro- or amido-substituted anthracoumarins. In the development of methods for obtaining these latter, my pupils took a most active part: scientific associate K. M. Ripper\(^2\) and the students S. N. Ignat’ev, A. V. Samoilov, and G. R. Shul’ts, which I consider it my pleasant duty to note here.

Experimental Part.

Communication I. Chloro-substituted anthracoumarins.

1) 4-Chloro-anthracoumarin,\(^3\) is obtained by condensation of corianic acid and 6-chloro-3-oxy-benzoic acid,\(^4\) according to Kostanecki.\(^5\)

\[ \begin{gathered} \text{corianic acid} + \text{6-chloro-3-oxy-benzoic acid} \ \xrightarrow[\ -2H_2O\ ]{\ +O\ }\ \text{4-chloro-anthracoumarin} \\ = C_{16}H_{7}O_{3}Cl. \\ (\text{mol. wt.} = 282.5) \end{gathered} \]

Corianic acid (1.5 g) and 6-chloro-3-oxy-benzoic acid (1.8 g) were dissolved in conc. sulfuric acid (28 g), to which was added a small excess of 18/4 fuming \(H_2SO_4\); the reaction mixture was slowly heated; in so doing, darkening of the mass occurred, and at \(45^\circ\) a noticeable evolution of \(SO_2\) gas began; at \(65^\circ\) this already became very vigorous. After 8 hours’ heating at this temperature the reaction mass was poured into water, whereby a yellow precipitate separated. After suction filtration this

\(^1\) Investigations in this direction, with the aim of proving the structure of cerulein and of similar compounds, must constitute a special task.

\(^2\) The portion of the work carried out jointly with K. M. Ripper was published in Monatshefte f. Chemie (1921), nos. 3 and 4: W. I. Minajeff u. K. Ripper, “Über Anthracumarinderivate.” This same part of the work was presented by K. M. Ripper to the University of Vienna as his “Inaug. Dissertation” for the doctoral degree, which was awarded to him.

\(^3\) The synthesis of 4-chloro-anthracoumarin was carried out jointly with K. M. Ripper (1919–20 academic year).

\(^4\) See V. I. Minaev and K. M. Ripper, 6-chloro-3-oxy-benzoic acid, Izv. T. T. I., vol. 6, p. 15.

\(^5\) B. 20 (1887), 3143.—Anthracoumarin, recrystallized from benzene and acetic acid, is described as a substance in the form of yellowish needles, m.p. \(260^\circ\), subliming in the form of beautiful, lustrous yellow needles, sparingly soluble in alcohol with fluorescence, similar to that of gallein solutions. Anthracoumarin dissolves readily in benzene and glacial acetic acid on heating, in alkalis and \(H_2SO_4\)—with green fluorescence.

the precipitate was washed many times with hot water, ammonia water, and then again with boiling water. The ammoniacal wash waters fluoresced, as in the preparation of anthracoumarin, dark green in reflected light and, in transmitted light, intensely red. After drying at \(110^\circ\), 0.85 g of a yellow crystalline substance was obtained, which corresponded to a 29% yield of the theoretical amount.

\(\alpha\)-Chloroanthracoumarin is readily crystallized from glacial acetic acid and nitrobenzene; it dissolves in conc. sulfuric acid, from which, on dilution with water, it again precipitates. \(\alpha\)-Chloroanthracoumarin is colored pale yellow, much more intense than the color of \(\alpha\)-chloroanthraquinone.

In its chemical properties \(\alpha\)-chloroanthracoumarin resembles \(\alpha\)-chloroanthraquinone, but the Cl of the former is much more mobile than the Cl of the latter; this phenomenon apparently must be explained by the presence, in the para-position to chlorine, of a negative residue\(^{1}\). Whereas introduction of chlorine into anthraquinone at the \(\alpha\)-position lowers the melting point by \(111^\circ\) and, respectively, at the \(\beta\)-position by \(69^\circ\), \(\alpha\)-chloroanthracoumarin has m.p. \(274^\circ\) C, i.e., \(14^\circ\) higher than the melting point of pure anthracoumarin\(^{2}\).

\(\alpha\)-Chloroanthracoumarin sublimes in the form of golden-yellow, long needles. After two recrystallizations from glacial acetic acid, in the form of golden-yellow felted needles, it shows a constant melting point \(=274^\circ\). Thus purified, sublimed, and brought to constant weight, the product on analysis gave figures in very good agreement with the formula given above.

Analysis.
I. 3.796 mg of substance gave 9.451 mg \(CO_2\) and 0.932 mg \(H_2O\) (by Pregl).
II. 3.903 mg » » 9.702 mg » » 0.943 mg »
III. 0.2131 g » » 0.1086 g AgCl (by Carius).

Calculated for \(C_{16}H_7O_3Cl\) (282.5) . . . C 67.85%; H 2.47%; Cl 12.57%.
Found . . . . . . . . . . . . . . . . . . . . . . C 67.90%; H 2.73%; Cl 12.62%.
Found . . . . . . . . . . . . . . . . . . . . . . C 67.68%; H 2.69%.

2) 2—Chloroanthracoumarin\(^{3}\).

\[ \begin{array}{c} \mathrm{CO}\\[-2pt] /\ \backslash\\[-2pt] \mathrm{CH}\quad \mathrm{O}\\[-2pt] \Vert\quad |\\[-2pt] \mathrm{C}\quad |\\[-2pt] \text{anthracoumarin skeleton with } \mathrm{Cl}\text{ substituent}\\[-2pt] \mathrm{CO} \end{array} \]

For the synthesis of this chloro-substituted anthracoumarin, 4-chloro-3-hydroxybenzoic acid\(^{4}\) was first prepared from \(m\)-hydroxybenzoic acid by chlorination with \(SCl_2\) in a carbon disulfide solution.

Brown acid (3 g) and the resulting 4-chloro-3-hydroxybenzoic acid (4 g) were dissolved in conc. \(H_2SO_4\) (25 g) mixed with 18% oleum (15 g). The reaction mass was heated for \(1/2\) hour on a water bath at \(50\)—\(60^\circ\) C. The solution, at first olive-colored, gradually turns brown, while remaining olive-colored in thin layers on the walls of the flask. \(SO_2\) is evolved in the form of gas bubbles of a specific odor.

Work-up of the reaction mass, carried out analogously to the preceding case, gave 0.4 g of crude product, soluble in pyridine and nitrobenzene. Recrystallization from the latter gave only 0.1 g of pure product. The reaction for chlorine (according to Beilstein) is quite distinct. The yield of the substance is very small. Apparently this 2-chloroanthracoumarin has not yet been investigated.

\(^{1}\) 4-Chloro-1-anthracoumarin was (see below) converted into 4-anilidoanthracoumarin—a reaction which here proceeds much more readily than with 1-chloroanthraquinone.

\(^{2}\) M.p. of anthraquinone—\(273^\circ\) (Beilstein, III, 406); m.p. of \(\alpha\)-chloroanthraquinone—\(162^\circ\) [A. 381, 1 (1911)]; m.p. of \(\beta\)-chloroanthraquinone—\(204^\circ\) [A. 283, 24 (1886)]. M.p. of anthracoumarin—\(260^\circ\).

\(^{3}\) The synthesis of 2-chloroanthracoumarin was first carried out by me together with A. A. Sanin as early as 1913, but was not published at that time.

\(^{4}\) Friedländer, III, 847, DRP. 74,493.

3) 5-Chloro-anthracoumarin\(^1\) was obtained by the general reaction from m-chloroboric and m-oxybenzoic acids.

\[ \text{[structural formula of 5-chloro-anthracoumarin]} \]

m-Chloroboric acid was obtained from benzaldehyde in the following series of consecutive transformations known in the scientific literature:

\[ \ce{ C6H5CHO ->[\text{nitration}] m{-}NO2{-}C6H4CHO ->[+\ce{CH3COONa}][\text{Perkin synthesis}^{2}] m{-}NO2{-}C6H4CH=CHCOOH ->[\text{reduction}] m{-}NH2{-}C6H4CH=CHCOOH } \]

\[ \ce{ m{-}NH2{-}C6H4CH=CHCOOH ->[\substack{\text{diazotization and subsequent}\\ \text{conversion of the hydrochloric diazo compound}\\ \text{through copper diazo salt into the chloro product}\\ \text{(heating with KOH, HCl}^{3}\text{)} }] m{-}Cl{-}C6H4CH=CHCOOH } \]

Of several experiments on the condensation of 5-chloro-1-anthraquinone with the greatest yield the following was accompanied: 3.6 g of m-chloroboric acid and 2.6 g of m-oxybenzoic acid in a solution of fuming sulfuric acid (30 g \(\ce{H2SO4}\) + 12 g of 18% oleum) were heated on a water bath for \(2 \tfrac{1}{2}\) hours at a temperature up to \(65^\circ\) during the first hour and at \(70^\circ\) during the last \(1 \tfrac{1}{2}\) hours. The most vigorous evolution of \(\ce{SO2}\) was in the middle of the condensation process and toward the end diminished noticeably. From the reaction mass, on pouring into ice water, there precipitated a sediment insoluble in ammonia solution.

The yield of the product was brought up to 20%; in the preceding experiment 1.1 g had been obtained.

The reaction for Cl (by Beilstein) is expressed quite distinctly.

5-Chloro-anthracoumarin is soluble in the cold in nitrobenzene, glacial acetic acid, methyl alcohol, chloroform, also in benzene with a yellow-brown coloration; it does not dissolve in alkalis, even on heating. Reprecipitated twice from hot nitrobenzene or 99–100% acetic acid, the brown powder has m.p. \(241^\circ\); recrystallized four times from acetic acid it showed m.p. \(242^\circ\). It sublimes readily in golden-yellow needles.\(^3\)

\(^1\) The synthesis of this chloro-substituted compound was carried out with the participation of student (now engineer-technologist) S. N. Ignat’ev.

\(^2\) B. XVI, 2036; XI, 1782; XII, 2660. See also Gattermann, p. 273.

\(^3\) B. XV, 2294; XVI, 2036. Gattermann, p. 217.

\(^3\) Note. The sublimed product apparently melts above \(280^\circ\); i.e., at this temperature melting was not observed. The reaction for chlorine is distinct and conclusive. Owing to the failure of the experiment, the yield and loss of material were not investigated more closely.

4) 6-Chloro-anthracoumarin.¹) For the synthesis of this chloro-substituted product, by preliminary nitration there was obtained p-nitrocinnamic acid; separated by the method of alcohol fractionation from the o-isomer and purified by conversion through the ethyl ester, the p-nitrocinnamic acid was reduced to p-amidocinnamic acid, which was then converted into p-chlorocinnamic acid.

Another route for the separation of the two p- and o-amido acids was also tried, by converting them into their copper salts.

p-Chlorocinnamic acid condensed with m-hydroxybenzoic acid under the usual conditions. The yields of the condensation product, as in the other cases, were not large. Chlorine in the purified product was clearly demonstrated by Beilstein’s reaction.

The syntheses of the 2-, 5-, and 6-chloro-substituted anthracoumarins will be repeated and studied more closely at the first favorable opportunity, which, we hope, should arise for our Siberian laboratories; for the present, only preliminary data are recorded concerning the investigation begun and suspended in 1920 (because of the absence of reagents, vessels, test tubes, etc.), the aims and tasks of which have been outlined above.

Communication II. Derivatives of 4-chloro-1-anthracoumarin.

(Together with K. M. Ripper).²)

1) 4-Anilido-1-anthracoumarin. 0.5 g of 4-chloro-anthracoumarin and 5 g of freshly distilled aniline, together with 0.25 g of freshly fused CH₃COOK and 0.05 g of cupric acetate salt, are heated to boiling in a round-bottomed Jena flask with an ascending glass tube as a reflux condenser; the contents of the flask at once become colored dark violet. After 1½ hours of heating, in order to remove the excess aniline, the reaction mass is boiled with dilute hydrochloric acid; the residue is filtered off and repeatedly boiled out with water, giving a crystalline powder of deep violet color, which is dried in a drying oven at 110°.

4-Anilido-anthracoumarin crystallizes well from alcohol and glacial acetic acid, and also sublimes in the form of violet-red crystals. It dissolves in concentrated sulfuric acid with a black-brown color; from the sulfuric-acid solution, on dilution with water, it again separates in the form of violet-colored flakes. On boiling 4-anilido-anthracoumarin with caustic potash alkali, dissolution occurs with a deep dark-red color and aniline is split off, which is readily detected by the characteristic odor of the latter.

After recrystallization of the crude crystalline product obtained from 96% alcohol, 0.33 g of substance was obtained, corresponding to a yield of 67% of theory. The product, recrystallized from glacial acetic acid and obtained in the form of red-violet plate-like crystals, showed a not especially sharp melting temperature between 184–186° C.

¹) The synthesis was carried out with the participation of student A. V. Samoilov in the winter of 1918–19, and, owing to the circumstances of the time, all records were lost.

²) W. I. Minaeff and K. Ripper, Über Anthracumarin-Derivate, Sitzungsberichte d. Akademie d. Wissenschaften in Wien, 130. Band (1921) and Monatshefte f. Chemie, 42. Band (1921), pp. 83–84. The communication was also presented at the III Mendeleev Congress (May 1922).

The analytical data agree well with the formula \( \mathrm{C}_{22}\mathrm{H}_{13}\mathrm{O}_{3}\mathrm{N} \) (mol. wt. \(=323\)), namely:

\[ \begin{aligned} \text{I. }& 2.486\ \mathrm{mg} \text{ gave (by Pregl) } 0.106\ \mathrm{cm^3}\ \mathrm{N_2}\ (27^\circ,\ 756\ \mathrm{mm});\\ \text{II. }& 2.279\ \mathrm{mg} \quad » \quad » \quad 0.097\ \mathrm{cm^3}\ \mathrm{N_2}\ (27^\circ,\ 758\ \mathrm{mm}). \end{aligned} \]

\[ \begin{aligned} \text{Calculated for } \mathrm{C}_{22}\mathrm{H}_{13}\mathrm{O}_{3}\mathrm{N}\ (323) &: \mathrm{N}\ 4.33\%;\\ \text{Found} &: \mathrm{N}\ 4.48\%,\ 4.52\%. \end{aligned} \]

2) Preliminary experiment for the preparation of 4,4′-di-anthracoumarin.
It was carried out with the aid of molecular copper according to F. Ullmann\(^{1}\). The reaction scheme may be represented by the following equation:

\[ \text{4-chloro-anthracoumarin} + \text{4-chloro-anthracoumarin} + 2\mathrm{Cu} \longrightarrow \text{4,4′-di-anthracoumarin} + 2\mathrm{CuCl} \]

4-Chloro-anthracoumarin (0.4 g) is dissolved in freshly distilled nitrobenzene (2.5 g); molecular copper (Naturkupfer C) (0.12 g) is added to the solution, and the mixture is heated for 2 hours, then diluted with nitrobenzene. On cooling, brown crystals separate from the dark-red solution; these are filtered off, washed with alcohol in order to remove nitrobenzene, and again boiled with dilute nitric acid for dissolution and removal of the excess copper. In this way we obtained 0.13 g of product, which sublimes in the form of brownish crystals, still does not melt at \(290^\circ\), and contains no chlorine. Chlorine could easily be demonstrated in the nitric-acid mother liquor, so that the product obtained undoubtedly represents di-anthracoumarin.

The easy, entirely clear formation of 4-anilido-1-anthracoumarin in good yield through replacement of the Cl atom in the 4-position of the arylated compound, and likewise the undoubted reaction of coupling of anthracoumarin molecules, indicate a quite easy possibility of new condensations in the sense of constructing the acid and vat dyes projected above in the anthracoumarin series. Unfortunately, owing to a lack of reagents and starting materials and similar circumstances, this investigation was interrupted in 1920; but we hope that it will be continued at the first opportunity.

January, 1924.

\(^{1}\) Cf. the preparation of 4,4′-di-\(\mathrm{CH_3}\)-di-anthraquinone and others: F. Ullmann and W. Minaeff, B. 45 (1912), 687.

1. B. 20 (1887), 3143. ↩↩

2. Georgievich and Grandmougin, Chemistry of Coloring Matters, p. 327.—It is possible that yellow benzanil will be similar in composition to the anthraquinone derivatives obtained by condensation of anthrone with aldehydes: Ch. Ztg. 1905, 1253. ↩↩

3. Jakobson and Julius, B. 20 (1887), 2580. See also DRP. 40375 (Frdl. I, 569). ↩

4. F. Slama, Inaug. Diss.; C. 1899, II, 967. ↩

5. Since the discovery in 1900 by R. Bohn of indanthrone and flavanthrone (up to 1917). ↩