COMPLEXES OF NITRO COMPOUNDS WITH OTHER ORGANIC SUBSTANCES

B. V. TRONOV and N. D. STREL’NIKOVA

I. Electronic types of chemical activity of atoms and atomic groups

One of the most important achievements of electronic chemistry is the development of the doctrine of the electronic types of chemical activity (reactivity) of atoms and atomic groups—molecules, radicals, ions.

From the standpoint of electronic chemistry, two types of chemical reactions are distinguished: reactions in which electron pairs remain unchanged, and reactions involving the formation and cleavage of electron pairs. This subdivision of chemical processes does not at all coincide with the long-accepted division in chemistry into reactions of addition, decomposition, and substitution. For example, the addition of acids to unsaturated hydrocarbons proceeds without change in the electron pairs:

\[ \mathrm{CH_3{-}C{:}{:}CH_2} \begin{matrix} \\[-0.6em] |\\[-0.2em] \mathrm{H} \end{matrix} \;\longrightarrow\; \mathrm{CH_3{-}\overset{+}{C}{:}\overset{-}{CH_2} + H} \;|\;{:}\mathrm{Br} \;\longrightarrow\; \mathrm{CH_3{-}C{-}CH_3} \begin{matrix} \mathrm{Br}\\[-0.2em] |\\[-0.2em] \\[-0.4em] |\\[-0.2em] \mathrm{H} \end{matrix} \]

The addition of atomic hydrogen (hydrogen at the moment of liberation) must proceed with the formation of new electron pairs:

\[ \mathrm{H_3C{-}C{:}{:}CH_2 + 2H^{\cdot} \;\longrightarrow\; H_3C{-}C{-}CH_3} \begin{matrix} \mathrm{H}\\[-0.2em] |\\[-0.2em] \\[-0.4em] |\\[-0.2em] \mathrm{H} \end{matrix} \]

Two mechanisms of decomposition reactions are possible; for example, formation of a double bond by elimination of a hydrogen halide from an organic halogen derivative (the reaction reverse to the first of those given here) and cracking:

\[ \mathrm{CH_3{-}C{:}C{-}CH_3} \begin{matrix} \mathrm{H}\quad \mathrm{H}\\[-0.2em] |\quad |\\[-0.2em] |\quad |\\[-0.2em] \mathrm{H}\quad \mathrm{H} \end{matrix} \;\longrightarrow\; \mathrm{CH_3{-}C^{\cdot} + {}^{\cdot}C{-}CH_3} \begin{matrix} \mathrm{H}\quad \mathrm{H}\\[-0.2em] |\quad |\\[-0.2em] |\quad |\\[-0.2em] \mathrm{H}\quad \mathrm{H} \end{matrix} \]

Exchange too can proceed in different ways [4]: ionic exchange

\[ \mathrm{A{:}B \longrightarrow \overset{-}{A}{:} + \overset{+}{B}} \]

and an ionic one, \(A : B \to A\cdot + B\cdot\), for example,

\[ [\mathrm{OH}]^{-} + \mathrm{R}:\ddot{\mathrm{Br}}: \to \mathrm{HO}\cdot \mathrm{R} + [:\ddot{\mathrm{Br}}:]^{-} \]

\[ \mathrm{R}\cdot + \mathrm{R}' : \mathrm{H} \to \mathrm{R} : \mathrm{H} + \mathrm{R}'\cdot \]

Bimolecular reactions of the first electronic type (with preservation of electron pairs) begin with electrostatic attraction between the electron pair of one of the reacting atoms or groups and the positive nucleus of the other component of the reaction. If both atoms or groups are ions with charges of opposite sign, they most often simply combine into a molecule.

If one component is an ion and the other is a neutral molecule, then the ion may simply act by its charge, while the neutral molecule is attracted because the electric charges in it are distributed nonuniformly: in some parts a negative charge predominates (the electron density is greater), in others a positive charge (lower electron density). It is very important here that sometimes one charge is more concentrated than the other (while the neutrality of the whole molecule as a whole is preserved). Thus, in the ammonia molecule the negative charge is more concentrated than the positive.

In one corner of the tetrahedron described around the nucleus of the nitrogen atom, a strong negative field is created; on the three other sides the molecule acts with a positive electric field, considerably weaker. Therefore, in ammonia reactions proceed most readily through the unshared (non-delocalized) pair of electrons of the nitrogen atom

\[ \begin{array}{c} \ \ \mathrm{H} \\ \ \ | \\ \mathrm{H}-\mathrm{N}: + \mathrm{H}-\mathrm{Cl} \to \left[ \begin{array}{c} \ \ \mathrm{H} \\ \ \ | \\ \mathrm{H}-\mathrm{N}-\mathrm{H} \\ \ \ | \\ \ \ \mathrm{H} \end{array} \right]^{+} [\mathrm{Cl}]^{-} \\ \ \ | \\ \ \ \mathrm{H} \end{array} \]

In the molecule of boron fluoride, on the contrary, the more concentrated positive field is in one of the corners of the tetrahedron around the nucleus of the boron atom, while the negative charge is dispersed over the fluorine atoms.

\(\mathrm{BF}_3\) reacts more readily with a positive charge:

\[ \begin{array}{c} \ \ \mathrm{F} \\ \ \ | \\ \mathrm{F}-\mathrm{B}^{+}-\mathrm{F}^{-}-\mathrm{H}^{+} \to \left[ \begin{array}{c} \ \ \mathrm{F} \\ \ \ | \\ \mathrm{F}-\mathrm{B}-\mathrm{F} \\ \ \ | \\ \ \ \mathrm{F} \end{array} \right]^{-} \mathrm{H}^{+} \\ \ \ | \\ \ \ \mathrm{F} \end{array} \]

Ammonia and boron trifluoride also readily combine with one another:

\[ \begin{array}{ccccc} & \mathrm{H} & & \mathrm{F} & \\ & | & & | & \\ \mathrm{H}-\mathrm{N}: & + & \mathrm{B}-\mathrm{F} & \to & \mathrm{H}-\mathrm{N}:\mathrm{B}-\mathrm{F}. \\ & | & & | & \\ & \mathrm{H} & & \mathrm{F} & \end{array} \]

There are also substances in which opposite charges of considerable intensity are concentrated in different parts of the molecule. Such is hydrogen fluoride, capable of reacting both with fluorine and with hydrogen. To a weaker degree this is characteristic of water. Its oxygen and hydrogens are active, but more often water adds to other substances.

with its oxygen atom. Compounds that have, in their molecules, simultaneously active centers with charges of opposite sign are especially prone to association. Hydrogen fluoride gives rather stable molecules

\[ \mathrm{H-F\cdots H-F} \]

Water in the liquid state consists of molecules of average composition \((\mathrm{H_2O})_4\). Among organic compounds, alcohols are associated to a considerable extent. Strong dimers, which persist even in the vapor phase, are formed by carboxylic acids:

\[ \begin{array}{c} \mathrm{CH_3-C-O-H\cdots O} \\ \qquad \| \qquad\qquad \| \\ \mathrm{O\cdots H-O-C-CH_3} \end{array} \]

Substances which, in the reactions under consideration by us, are acted upon by a negative electric field exhibit, as it is said, electron-donor or nucleophilic reactivity. Substances in which the action of a positive field predominates possess electron-acceptor or electrophilic reactivity. Finally, there are substances with a mixed type of chemical activity; moreover, the electron-donor activity must be concentrated in one part of the molecule, and the electron-acceptor activity in another.

The concept of electron-donor and electron-acceptor reactivities is a substantial step forward in elucidating the dependence of the course of reactions on the chemical structure of the reacting substances. This concept makes it possible to understand why bonds of one and the same structural type (single, double, triple) behave differently depending on which atoms are connected by the given bond, and also on the composition and structure of the other parts of the molecule. Since the electronic type of chemical activity can, at least in part, be derived from the principles of electronic and quantum chemistry, here the achievements of theoretical physics bring concrete benefit in solving chemical problems. Thus, the easy variability of the motion (“mobility of the electron cloud”) of the \(\pi\)-electrons of the carbon double bond leads to the fact that, in hydrocarbons of the ethylene series, the electron-donor function predominates. Therefore they readily react with electron-acceptor reagents, for example, free halogens, boron fluoride, etc., but are not capable of reacting with substances of electron-donor character, for example, with ammonia, sodium bisulfite, and others. Aldehydes and ketones, on the contrary, exhibit chiefly electron-acceptor ability due to the positive charge of the carbonyl carbon. They react with ammonia and its derivatives, with hydrocyanic acid, with the CN ion adding to the carbon.

The electron-donor or electron-acceptor activity of a given atom or atomic group also depends on the other parts of the molecule; sometimes it depends so strongly that the given atom or group acquires the opposite electrochemical character. For example, the ethylenic double bond, under the influence of phenyl rings, loses the ability to add bromine, but adds sodium (something simple olefins are not capable of):

\[ (\mathrm{C_6H_5})_2\mathrm{C=C}(\mathrm{C_6H_5})_2 + \mathrm{Br_2} \]

does not react.

\[ (\mathrm{C_6H_5})_2\mathrm{C=C}(\mathrm{C_6H_5})_2 + 2\mathrm{Na} \rightarrow (\mathrm{C_6H_5})_2\mathrm{CNa-CNa}(\mathrm{C_6H_5})_2 \]

This is explained by the fact that the phenyl ring is, in general, capable of drawing off electrons from the atoms bonded to it (with which many features of aromatic compounds are connected, for example, the low activity of haloge-

in many reactions that remove halogen, the weakening of basic and strengthening of acidic properties in phenols as compared with ordinary alcohols, and the weakening of basic properties in aromatic amines). Owing to this influence of the phenyl groups, the group \(> \mathrm{C}=\mathrm{C}<\) becomes electron-unsaturated and no longer capable of attracting a positive halogen ion, as happens in ethylene. This group must now more readily attract an electron-deficient pair from some reagent and can add two separate electrons from sodium atoms. An anion is formed:

\[ \left[(\mathrm{C}_6\mathrm{H}_5)_2\mathrm{C}-\mathrm{C}(\mathrm{C}_6\mathrm{H}_5)_2\right]'' \]

In reactions that proceed with the formation and cleavage of electron pairs, a different type of electronic interaction—apparently the principal one—has great significance, one connected with the wave nature of electrons and atomic nuclei. According to wave theory and quantum mechanics, it is precisely wave interaction that explains the strength of an electron pair with opposite “spins” and the special stability of the eight-electron system (octet) in the valence shell of an atom. The energy of formation of electron pairs, with which the ease of occurrence of reactions of this type (atomic-radical) is connected, is also determined not only by the nature of the directly reacting atoms, but also by the composition and structure of the whole molecule. Since the laws of wave interaction are less known to us than the laws of electrodynamics, in this field physical theory as yet permits less explanation and prediction.

However, even in the study of reactions of the first type we cannot expect physical theories to provide a complete explanation of the facts of chemistry. The chemical form of motion of matter is a special form, qualitatively different from other forms of motion, and its laws cannot be reduced to electronic and quantum physics.

II. Methods for Determining the Electronic Type of Chemical Activity

Of the three types of chemical activity—electron-donor, electron-acceptor, and atomic-radical—in the present work we shall consider only the first two, since the last is based on a different type of electronic interaction and requires other methods of investigation.

A physical constant characterizing the distribution of electric charges in a molecule is the dipole moment. This constant is often used in electronic chemistry. Thus, some investigators have calculated the yields of ortho-, meta-, and para-derivatives in the aromatic series, taking precisely the dipole moments of substituents as their basis. The results of such calculations sometimes agree quite well with experimental data.

Refraction is also of importance for judging the course of reactions. The dipole moment is a static constant. It expresses the distribution of charges in the molecule of the substance under consideration by itself, without any external influences. Refraction is connected with the polarizability of the molecule, with its ability to change the distribution of charges under the influence of external effects, for example, in the presence of a reagent acting on the given substance.

Comparing, for example, ethane with ethylene, we find that both hydrocarbons are nonpolar, i.e., their electric charges are arranged symmetrically. Under the action of external agents, the electrons of a single bond, \(\sigma\)-electrons, are little susceptible to this action and only with difficulty change the orbits of their motion. In a double bond we have two pairs of electrons: \(\sigma\)-electrons

and \(\pi\)-electrons. Owing to the mobility of the latter, the distribution of electron density in the ethylene molecule changes readily; the molecule undergoes considerable polarization. In accordance with this difference in polarizability, the refraction of a simple carbon bond for the sodium D line is 1.21; the refraction of a double bond is 4.16 [1].

The physical constants indicated are of great use in solving questions concerning the electronic type of reactivity of substances. However, both constants—the dipole moment and refraction—give a general characterization of the distribution of charges and of changes in this distribution, without sufficiently taking into account details that are important for the chemist. Thus, the dipole moment is equal to the product of the magnitude of the charge and the distance between the centers of the charges. But with the same magnitude of charges in parts of a molecule (or in parts of a group of atoms constituting the molecule of the substance under study) and with the same distance between the centers of the charges, different concentrations of these charges are possible. The charge may be distributed over a rather large space or concentrated in a smaller volume, forming a very active electron-donor or electron-acceptor center.

Moreover, there may be several electron-donor or electron-acceptor centers in a molecule. For example, the dipole moment of dioxane is equal to zero because of the symmetrical distribution of charges, but its oxygen atoms exhibit a certain electron-donor activity characteristic of simple ethers.

The dipole moment of p-dinitrobenzene is also equal to zero, which likewise does not prevent this compound from exhibiting a definite electronic activity in its nitro groups.

Furthermore, in determining the electronic type of chemical activity, we are already passing from electrical phenomena to chemical ones; and chemical interaction is connected with electrical and wave phenomena, but is not reducible to them, has its own specific regularities and, consequently, must be studied not only by physical but also by chemical methods. The Conference on the Theory of Chemical Structure in Organic Chemistry, held on June 11–14, 1951, by the Institute of Organic Chemistry of the Academy of Sciences of the USSR, quite correctly outlined the paths for the further development of organic chemistry. The resolution of the conference [3] states:

“The conference notes that the most important task of theoretical organic chemistry is the creative development of Butlerov’s doctrine, the deepening and expansion of its basic principles in the light of modern ideas about the structure of matter.

The development of theoretical organic chemistry, and first of all the development of the theory of chemical structure, must proceed in a resolute struggle against idealistic and mechanistic theories in chemistry on the basis of the worldview of dialectical materialism.

Concerning the relationship between chemical and physical methods of investigation, we read in the resolution:

“The leading direction of organic chemistry is the synthetic direction and the studies of the structure and properties of organic substances, including natural ones, associated with it, by the methods of synthetic organic chemistry. In chemical synthesis the connection between theory and practice is manifested most vividly. At the same time, in work on synthesis the theory of structure is constantly tested and deepened, and the area of its application is broadened.

In the synthetic direction, the theory of structure also finds new stimuli for development.”

And further:

“To solve the problems of the theory of chemical structure, it is necessary to make broader use, alongside the brilliantly justified chemical pathways, …”

and methods of investigation, all the achievements of related disciplines and, first of all, of modern physics with all the richness of its experimental and theoretical methods.”

We have already spoken about the basic physical methods for determining the type of chemical activity, i.e., reactivity. The chemical determination of the capacity for reactions, obviously, must first of all consist in the study of these reactions. In doing so it is necessary to ascertain with which reagents the given substance interacts, to investigate the reaction products (the properly synthetic method), to measure the rate of the reactions, to clarify their mechanism, to determine the dependence of the direction and rate of the reactions on the structure of the reacting substances themselves, on various physical factors, on the medium in which the process takes place, and on catalysts.

The experience accumulated by chemists in the investigation of diverse reactions indicates that by no means are all reactions equally suitable for establishing the type of chemical activity. One must take reactions with a known mechanism and, moreover, simple ones, not complicated by subsequent change of the initially formed product. Otherwise we may obtain an entirely erroneous idea of the type of activity.

Chemists, for example, long believed that the halogen in acid halides of carboxylic acids is very active, because acid halides readily react with water, alcohols, ammonia, etc.

$$ \mathrm{CH_3{-}C(=O){-}Cl + HOH \rightarrow CH_3{-}C(=O){-}OH + HCl.} $$

$$ \mathrm{CH_3{-}C(=O){-}Cl + HO{\cdot}CH_3 \rightarrow CH_3{-}C(=O){-}O{\cdot}CH_3 + HCl} $$

$$ \mathrm{CH_3{-}C(=O){-}Cl + NH_3 \rightarrow CH_3{-}C(=O){-}NH_2 + HCl} $$

These reactions usually proceed much more readily than does the interaction of halogenated hydrocarbons with the same reagents:

$$ \mathrm{CH_3{-}CH_2Cl + HOH \rightleftarrows CH_3{-}CH_2OH + HCl} $$

$$ \mathrm{CH_3{-}CH_2Cl + HO{\cdot}CH_3 \rightleftarrows CH_3{-}CH_2{-}OCH_3 + HCl} $$

$$ \mathrm{CH_3{-}CH_2Cl + NH_3 \rightleftarrows CH_3{-}CH_2{-}NH_3Cl.} $$

However, experiments with the action of iodide salts led to the opposite result:

$$ \mathrm{CH_3{-}C(=O){-}Cl + KI \rightarrow CH_3{-}C(=O){-}I + KCl} $$

$$ \mathrm{CH_3{-}CH_2Cl + KI \rightarrow CH_3{-}CH_2I + KCl.} $$

It turned out that here acetyl chloride reacts more slowly than ethyl chloride. Meanwhile, precisely in the action of iodide salts we should expect direct exchange by atoms or ions of halogens; whereas in the other reactions given above for acid halides, an entirely different process is possible.

another mechanism [5], with preliminary rupture of the double bond in the carbonyl group:

\[ \begin{aligned} &\mathrm{CH_3{-}C(=O){-}Cl + HOH \longrightarrow CH_3{-}C(OH)_2{-}Cl \longrightarrow CH_3{-}C(=O){-}OH + HCl} \\ &\mathrm{CH_3{-}C(=O){-}Cl + HOCH_3 \longrightarrow CH_3{-}C(OCH_3)(OH){-}Cl \longrightarrow CH_3{-}C(=O){-}OCH_3 + HCl} \\ &\mathrm{CH_3{-}C(=O){-}Cl + NH_3 \longrightarrow CH_3{-}C(NH_2)(OH){-}Cl \longrightarrow CH_3{-}C(=O){-}NH_2 + HCl} \end{aligned} \]

The new explanation is in very good agreement with the facts of chemistry. It has long been known that a halogen and a hydroxyl cannot be attached to the same carbon atom. $\alpha$-Halogen-substituted simple ethers give up their halogen so readily that it is not even possible to determine with any accuracy the rate of their reactions, for example, with amines or alcoholates [9]. M. F. Shostakovskii [10] even found that in these ethers the halogen is in the ionic state:

\[ \mathrm{CH_3{-}O{-}CH_2Cl \longrightarrow [CH_3{-}O{-}CH_2]^+\,Cl^- .} \]

Thus, a halogen atom indeed acquires very great activity if it is located at a carbon atom connected by a single bond with oxygen, and there is no reason to think that a halogen attached to a carbonyl carbon should be active. Chlorine in acid chlorides is not active in itself (under the influence of the carbonyl group), but becomes active only after the carbonyl group has added something to itself.

Consequently, in order to determine the true activity of some atom or group of atoms in a molecule, one must take a reaction in which the given atom participates while being part of the molecule of precisely the substance under study, and not of an intermediate product arising with the direct participation of another atom or group. Only under this condition shall we obtain results that agree well with those to which the application of physical methods of experiment and theoretical calculation leads. Thus, in the example of acid halides considered, the propositions of electronic chemistry made one expect not an increase but a decrease in the activity of the halogen under the influence of the carbonyl group, and the great activity of acetyl chloride toward water, alcohols, ammonia, etc., seemed strange. Meanwhile, the correct application of electronic and quantum theories is important for deriving the regularities of the mutual influence of atoms in molecules. Correctly derived regularities, in turn, contribute to the success of synthetic work to a considerable degree.

The indicated requirement for choosing simple reactions, in which only that part of the molecule whose type of activity we want to determine participates, is very well satisfied by complex-formation processes that are not accompanied by a deeper change in the substance.

To find out whether a given substance acts in complex formation as an electron donor or as an electron acceptor, we must add to it a reagent of unquestionably electron-acceptor or electron-donor character. Electron-acceptor reagents are strong acids, for example, $\mathrm{HClO_4}$, $\mathrm{H_2SO_4}$, $\mathrm{C_6H_5SO_3H}$ (picric acid has strong acidic properties, but can also react through its nitro groups). Salts of many metals also belong here. The same...

the effect predominates also in carboxylic acids, as well as in boric acid esters capable of giving complexes with alcohols having clearly acidic properties:

\[ (\mathrm{C_2H_5O})_3\mathrm{B}+\mathrm{C_2H_5OH}\rightarrow[\mathrm{B}(\mathrm{C_2H_5O})_4]\mathrm{H}. \]

A definitely pronounced electron-donor character is possessed by amines of the saturated series, nitrogen heterocycles with strong basic properties, and, in general, a large proportion of amines. Weaker, yet still electron-donor, reagents are ethers—in any case, ethers of the saturated series.

Having studied the action on the compound under investigation of several reagents of both types, we learn what type of electronic activity the given compound has. If this compound enters into complex-formation interaction with reagents of both electronic types, it means that it possesses mixed activity; it has both positive and negative poles of sufficient intensity.

One of the most important immediate tasks of organic chemists at the present time should be regarded as the systematic investigation of all the most important classes of organic compounds with respect to their ability to form complexes. In doing so, it is necessary not only to identify qualitatively one or another type of electronic activity of a substance, but also to give a quantitative assessment of this activity. The electronic type of activity is mainly determined by some most active group of atoms, the so-called functional group (sometimes by several such groups), but it also depends on the other parts of the molecule. It is therefore of interest to identify and quantitatively assess the influence on the activity of the principal (predominant) group, as well as of other atoms and atomic groups. Many examples are known in which a given functional group may fail entirely to give the reactions characteristic of it if the molecule contains many atoms or groups that weaken the chemical activity of that group. For example, the amine nitrogen of triphenylamine has almost no basic properties; this amine does not react even with hydrochloric acid. Tetraphenylethylene does not give the bromine-addition reaction characteristic of ethylenic hydrocarbons, etc.

Reactions of complex formation may be studied by various methods. The preparative method should be placed first, i.e., the preparation of complex compounds and their isolation with subsequent analysis. However, many complexes are too unstable. In such cases one has to be content with methods of physicochemical analysis, determining some physical index for systems containing the compounds under study in different ratios. Thermal analysis is used most often; sometimes good results are given by the determination of viscosity, electrical conductivity, specific gravity, colorimetric analysis, etc. The degree of chemical activity of the complex-forming centers of a molecule can be judged from certain data of physicochemical analysis (for example, from the form of melting diagrams, from vapor elasticity), and also from the energy of formation of the complex compounds and from their stability; reactions of mutual displacement of the components of a complex are convenient in this respect.

In the present work the results are given of an investigation of complexes formed by nitro compounds of the aromatic series with many other organic substances. Nitro compounds are of interest in this respect because the nitro group, as could be judged from the literature data, is a complex former of a mixed type, functioning both as an electron donor and as an electron acceptor.

Studying the complex formation of various nitro compounds and noting the strengthening or weakening of electron-donor or electron-acceptor

abilities under the influence of various changes in the composition and structure of molecules, much can be clarified in the question of the mutual influence of atoms in molecules.

III. Review of the Literature on Complex Formation of Nitro Compounds

1. Complexes of Aromatic Nitro Compounds with Mineral Substances

The complexes of nitro compounds with aluminum halides have been studied rather well. Thus, for nitrobenzene the following are known [11, p. 241; 2]:

\[ \begin{aligned} &\mathrm{C_6H_5NO_2\cdot AlCl_3}; \qquad &&2\mathrm{C_6H_5NO_2\cdot AlCl_3},\\ &\mathrm{C_6H_5NO_2\cdot AlBr_3}; \qquad &&2\mathrm{C_6H_5NO_2\cdot AlBr_3}. \end{aligned} \]

Metadinitrobenzene gives [11, 261] \(\mathrm{C_6H_4(NO_2)_2\cdot AlCl_3}\). Analogous complexes [11, 32, 325] have been obtained for ortho- and para-nitrotoluenes.

Compounds are known with antimony trichloride [2] and pentachloride [11, 241], with mercuric chloride [11, 325], for example,

\[ p\text{-}\mathrm{CH_3C_6H_4NO_2\cdot HgCl_2}, \]

\[ p\text{-}\mathrm{CH_3C_6H_4NO_2\cdot HgCl_2}. \]

Let us also note complexes with lithium iodide [11, 241, 321]:

\[ \mathrm{C_6H_5NO_2\cdot LiI}; \qquad o\text{-}\mathrm{CH_3C_6H_4NO_2\cdot LiI}. \]

A crystalline compound of nitrobenzene with sulfuric acid, having the composition

\[ \mathrm{C_6H_5NO_2\cdot H_2SO_4}, \]

has been obtained [13].

In one of our works [6], using the kinetic method of physicochemical analysis, the existence of two compounds of \(\mathrm{C_6H_5NO_2}\) with nitric acid was demonstrated. The rate of nitration of nitrobenzene with nitric acid was determined at different ratios of the reacting components. Within the limits of the first gram-molecule of acid the reaction rate increased according to the law of mass action. With the second gram-molecule of \(\mathrm{HNO_3}\), the rate of nitration of \(\mathrm{C_6H_5NO_2}\) increased 150-fold; with the third, by another 20-fold. Finally, the addition of the fourth gram-molecule of acid again led to a comparatively small increase in the rate. These facts can be explained only by the formation of the compounds \(\mathrm{C_6H_5NO_2\cdot HNO_3}\) and \(\mathrm{C_6H_5NO_2\cdot 2HNO_3}\), of which the first is more stable than the second. The first molecule of \(\mathrm{HNO_3}\) is rather firmly bound by nitrobenzene and can nitrate the benzene ring only to an insignificant extent.

The second molecule is more easily split off and acts more freely; the third is completely free and reacts with nitrobenzene normally (of course, more slowly than with benzene, since the nitro group strengthens the benzene nucleus).

In all the complex-formation reactions listed, the reagents acting on nitro compounds are strongly pronounced electron acceptors; consequently, the nitro group must function as an electron donor, i.e., through its oxygen atoms. With \(\mathrm{HNO_3}\), compounds should be formed:

\[ \mathrm{C_6H_5{-}N(=O)(=O)\ldots HO\,NO_2} \quad \text{or, more precisely} \quad \mathrm{C_6H_5{-}\overset{+}{N}(=O){-}\overset{-}{O}\ldots HONO_2}. \]

\[ \mathrm{C_6H_5—N{=}O\ldots HONO_2} \quad\text{or}\quad \mathrm{C_6H_5—\overset{+}{N}—\overset{-}{O}\ldots HONO_2} \]
\[ \begin{array}{c} \mathrm{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ O\ldots HONO_2,} \end{array} \]
and it may be,

\[ \mathrm{C_6H_5—\overset{+}{N}—\overset{-}{O}\ldots HO—\overset{+}{N}—\overset{-}{O}\ldots HONO_2} \]
\[ \begin{array}{cc} \mathrm{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ O} & \mathrm{O} \end{array} \]

Electrolytic dissociation of these compounds gives the cations

\[ \left[ \begin{array}{c} \mathrm{C_6H_5N—OH}\\ \mathrm{\ \ \ \ \Vert}\\ \mathrm{\ \ \ \ O} \end{array} \right]^{+} \quad\text{and}\quad \left[ \begin{array}{c} \mathrm{C_6H_5—N—OH}\\ \mathrm{\ \ \ \ |}\\ \mathrm{\ \ \ \ OH} \end{array} \right]^{++} \]

The complex with sulfuric acid must have the structure

\[ \mathrm{C_6H_5—\overset{+}{N}—\overset{-}{O}\ldots HOSO_2OH} \]
\[ \begin{array}{c} \mathrm{\ \ \ \ \ \ O} \end{array} \]
or
\[ \left[ \begin{array}{c} \mathrm{C_6H_5—N—OH}\\ \mathrm{\ \ \ \ |}\\ \mathrm{\ \ \ \ OH} \end{array} \right]^{++} \left[\mathrm{SO_4}\right]^{--}. \]

In the chlorides and bromides of aluminum, antimony, and mercury, the most active complex-forming center is the metal ion, which also has an electron-acceptor character.

In lithium iodide, complex formation is possible both through lithium and through iodine; but if iodine were acting here, the complexes could also be obtained with other iodide salts.

The electron-acceptor function of the nitro group in inorganic complexes of nitro compounds cannot be noted with certainty. Its possible manifestation is indicated, for example, by the fact that metadinitrobenzene dissolves in liquid ammonia, forming a solution of the color of an aqueous solution of \(\mathrm{KMnO_4}\) [11, 259]. It will be shown below that colored complexes are formed precisely from nitro compounds with amines, and that the formation of these complexes proceeds through the positive field of the nitrogen of the nitro group.

2. Organic–inorganic ternary complexes of nitro compounds

In the work of B. V. Tronov, L. N. D’yakonova, and E. A. Zonova [7], ternary complex compounds are described, whose composition includes:

a) polynitro compounds of the benzene series;
b) iodide salts;
c) compounds with a double or triple carbon bond to other nonmetallic elements, for example, aldehydes, ketones, esters, nitriles.

In these complexes, one molecule of a nitro compound and three molecules of a ketone, etc., correspond to one molecule of iodide. On the structure of complex compounds of this type, the following considerations may be expressed, differing somewhat from the assumptions given in the cited article.

It is known that sodium iodide with acetone forms a compound of composition \(\mathrm{NaI\cdot 3C_3H_6O}\). The complex bonds here could be provided either by iodine ions or by the sodium ion, although the latter is less capable of complex formation than lithium. However, the second assumption is excluded, since \(\mathrm{NH_4I}\) and \((\mathrm{CH_3})_4\mathrm{NI}\) behave analogously to \(\mathrm{NaI}\), and the tetramethylammonium ion gives no complexes at all. The iodine anion can combine only with

positively charged atom of the acetone molecule, i.e., with the carbonyl carbon of the semipolar ketone form. The negative poles, created by the unshared electron pairs of the oxygen atoms, remain free and are strengthened. These negative poles are probably attracted to the positive charges of the nitro-group nitrogens. The latter is consistent with the particular capacity of trinitro compounds to take part in such complexes, and specifically with nitro groups in the 1, 3, 5 positions (with such an arrangement of the $\mathrm{NO_2}$ groups, the positive charges of the nitrogen atoms should especially reinforce one another). In general we obtain the scheme:

$$ \begin{array}{cccccc} & \overline{\mathrm{O}} & \mathrm{CH_3} \\ & \diagdown & + & \diagup \\ & & \mathrm{C} \\ & \diagup & + & \diagdown \\ \mathrm{CH_3} & & & \cdots & \overline{\mathrm{I}} & \cdots \mathrm{CH_3}\!-\!\overset{+}{\mathrm{C}}(\overline{\mathrm{O}})(\mathrm{CH_3}) \cdots \overset{+}{\mathrm{C}}(\mathrm{CH_3})_2\!-\!\mathrm{O} \cdots \overset{+}{\mathrm{N}}(\overline{\mathrm{O}})(=\mathrm{O})-\mathrm{C_6H_3(NO_2)_2} \end{array} $$

It is possible that in crystals, and perhaps partly also in solution, all three oxygens of the carbonyl groups surrounding the iodine ion are saturated by the nitrogens of nitro groups—though from different molecules of trinitrobenzene. Such a structure has not been proved, but is quite probable; in its favor speaks, for example, the coincidence of the number of carbonyls and nitro groups associated with one iodine atom.

3. Complexes of Nitro Compounds with Organic Substances

Older literature sources contain information on purely organic complexes only for compounds with two or three nitro groups. Thus, a crystalline complex of paradinitrobenzene with naphthalene of composition $\mathrm{C_6H_4(NO_2)_2 \cdot C_{10}H_8}$ has long been known [11, 261].

In the form of this compound, paradinitrobenzene is recommended for isolation from a mixture with the meta isomer; 1, 3, 5-trinitrobenzene crystallizes from a benzene solution with one molecule of $\mathrm{C_6H_6}$.

Many complex compounds are known for nitrophenols, especially for picric acid. The latter gives molecular compounds [12, 271] with pinene, benzene, mesitylene, tetramethylbenzene, pentamethylbenzene, hexamethylbenzene, indene, naphthalene, its homologues and derivatives, anthracene, phenanthrene, etc., and with phenols [12, 274]. Picric acid and other nitrophenols give crystalline complexes with numerous amines. However, these complexes are of two types. They may arise through attraction of the hydroxyl hydrogen of the phenol to the nitrogen of the amine, i.e., have the structure of ammonium salts, or through attraction between the amine nitrogen and the nitro group. As Pfeiffer [15] notes, salt-like complexes of the first type are close in color to nitrophenols, i.e., only weakly colored (picrates, for example, are yellow); formation of nitro complexes, however, leads to a strong deepening of the color. Even cases of isomerism of amine complexes of nitrophenols are known.

Amine complexes of nitro compounds that do not contain phenolic hydroxyl are also known. Such complexes, in all cases with aromatic amines, were discovered by I. I. Ostromyslensky [14], and then studied by E. V. Biron and Morguleva and by other authors. Some

from these complex compounds are readily obtained in crystalline form (for example, the stable compound of metadinitrobenzene with benzidine, precipitating as well-formed black crystals); the formation of other complexes was detected by the authors cited by the colorimetric method. These include, for example, the simplest representative of this group—the complex of nitrobenzene with aniline

\[ \mathrm{C_6H_5NO_2 \cdot C_6H_5NH_2}. \]

Nitro compounds of the aromatic series are also capable of interacting with amines of the saturated series. Thus, an increase in electrical conductivity was observed in systems composed of nitrobenzene with diisoamylamine and triisoamylamine. In the latter case, the amine molecule contains no parts capable of complex formation other than the nitrogen atom, which can act only by means of its unshared electron pair. Here, therefore, it must be considered beyond doubt that, on the side of the nitro compound, the positively charged nitrogen of the nitro group participates in the reaction:

\[ \begin{array}{c@{\qquad\text{or}\qquad}c} \begin{array}{c} \mathrm{O}\\[-2pt] \overset{+}{\mathrm{C_6H_5{-}N}}\!{=}\!\mathrm{O}^{-}\\[-2pt] \vdots\\[-2pt] \mathrm{NR_3} \end{array} & \begin{array}{c} \mathrm{O}^{-}\\[-2pt] \overset{+}{\mathrm{C_6H_5{-}N^{+}{-}O^{-}}}\\[-2pt] \vdots\\[-2pt] \mathrm{NR_3} \end{array} \end{array} \]

and electrolytic dissociation probably proceeds with the formation of ions

\[ [\mathrm{C_6H_5 \cdot NO_2}]^{-}\;[\mathrm{NR_3}]^{++}. \]

In our laboratory a large number of systems have been studied that contain a nitro compound on the one hand and representatives of various other classes of organic substances on the other. In all the works already published on this question, a new electrochemical method of physicochemical analysis was used, developed by one of us jointly with L. P. Kulev [8]. The essence of the method consists in measuring the electromotive force of a galvanic cell composed of two metallic electrodes and the liquid under investigation between them. As the active metal directly reacting with the substance under study, sodium was most often used, more rarely zinc or some other metal. The second electrode was almost always a platinum crucible. The measuring instrument was a voltmeter or millivoltmeter, representing a galvanometer with a high resistance. This instrument can serve only for an approximate determination of the electromotive force, since the readings of the voltmeter depend also on the electrical conductivity of the system, but for our purpose this method proved more suitable than accurate potentiometric measurements. For example, in determining the activity of hydrogen in alcohols we obtained results in very good agreement with those given by determining the rate of reaction with sodium; meanwhile, potentiometric measurements showed a considerably smaller difference between primary, secondary, and tertiary alcohols. This is apparently explained by the fact that structural influences in the alcohol molecule, which alter the reactivity of the hydroxyl hydrogen, act in the same direction on both the electromotive force and the electrical conductivity. As a result, the complex physical quantity read on the voltmeter of our apparatus proves more sensitive and reveals better the difference in chemical activity of the compounds being compared.

If a compound, more or less active with respect to metals—for example, one possessing active hydrogen—enters into a complex-forming reaction, then its chemical activity, in particular the activity of the hydrogen atom entering into its molecule, must change. As numerous experiments have shown, our method for determining chemical activity makes it possible to detect this change readily. In principle, changes in different directions are possible, but in our experiments so far only a change toward increased activity with respect to metals has been reliably observed. Moreover, comparing the results of our determinations with those obtained by other research methods, we found that the maximum of the electromotive force value determined in our apparatus corresponds (exactly or approximately) to the composition of the complex compound formed.

The electrochemical study of systems containing nitro compounds led to the following results.

1. Nitro compounds with alcohols

For primary and secondary alcohols of the saturated series, the addition of nitro compounds did not produce an increase in the e.m.f. For trimethylcarbinol with nitrobenzene, a maximum e.m.f. was observed at 20–30 molecular percent of alcohol. The same was true for cyclohexanol.

2. Nitro compounds with phenols

Phenols proved to be better complex-forming agents. \(C_6H_5OH\) showed e.m.f. maxima, and quite considerable ones (an increase sometimes by hundreds of times in comparison with phenol itself), at the following molecular percentages of phenol:

\[ \begin{array}{ll} C_6H_5NO_2 \ldots \ldots \ldots \ldots 30\text{—}40\%; & \alpha C_{10}H_7NO_2 \ldots \ldots \ldots 30\%;\\ mC_6H_4(NO_2)_2 \ldots \ldots \ldots 30 \text{ and } 70\%; & mClC_6H_4NO_2 \ldots \ldots \ldots 30\text{—}40\%;\\ pC_6H_4NO_2(OCH_3) \ldots \ldots 50\%. & \end{array} \]

3. Nitro compounds with carboxylic acids

With nitrobenzene, e.m.f. maxima were observed at the following molecular percentages of acids:

\[ \begin{array}{ll} CH_3COOH \ldots \ldots \ldots 60\%; & CH_3CH_2COOH \ldots \ldots 60\text{—}70\%;\\ CH_3CH_2CH_2COOH \ldots \ldots 60\text{—}70\%; & (CH_3)_2CHCOOH \ldots \ldots 50\%;\\ (CH_3)_2CHCH_2COOH \ldots \ldots 50\%. & \end{array} \]

with \(\alpha\)-nitronaphthalene:

\[ \begin{array}{ll} CH_3CH_2COOH \ldots \ldots 70\%; & CH_3CH_2CH_2COOH \ldots \ldots 70\%;\\ (CH_3)_2CHCH_2COOH \ldots \ldots 50\%. & \end{array} \]

In these experiments it was especially interesting that acids with a straight chain of carbon atoms—acetic, propionic, and butyric—upon addition of nitro compounds gave only a slight increase in e.m.f., whereas acids with a branched chain—isobutyric and isovaleric—gave an increase by hundreds of times (125 and 700). This phenomenon may be connected with the composition of the complex compounds. In acetic, propionic, and butyric acids, the e.m.f. maximum is reached at a molecular ratio of nitro compound to acid equal to \(1:2\); for acids of iso-structure, the composition of the most active mixture is \(1:1\). It is possible that acids of normal structure—

…enter the complex in the form of dimers with weakly active hydrogen atoms, whereas isobutyric and isovaleric acids enter in monomolecular form.

4. Other complex-forming systems

By the same electrochemical method, certain systems were investigated that included compounds having tautomeric forms with active hydrogen. Ethyl acetoacetate and nitromethane were taken:

\[ \mathrm{CH_3{-}C(=O){-}CH_2{-}COOC_2H_5} \rightleftarrows \mathrm{CH_3{-}C(OH){=}CH{-}COO{\cdot}C_2H_5} \]

\[ \mathrm{CH_3{-}N(=O){=}O} \rightleftarrows \mathrm{CH_2{=}N(=O){-}OH}. \]

Ethyl acetoacetate with nitrobenzene did not give a positive result. With nitromethane, an increase in emf from 0.1 mv to 6.6 mv was achieved.

IV. Experimental Part

In the experimental part of the present work, systems were studied which, in addition to nitro compounds, included phenols, carboxylic acids, and amines. Of the nitro compounds, nitrobenzene, p-nitrotoluene, α-nitronaphthalene, and m-dinitrobenzene were taken; of the acids—acetic and benzoic; phenols—carbolic acid, α-naphthol, β-naphthol, resorcinol; amines—aniline, p-toluidine, dimethylaniline, α-naphthylamine, β-naphthylamine, benzidine, diphenylamine, tribenzylamine, piperidine, diisoamylamine. The systems taken were studied by methods of thermal and colorimetric analysis.

1. Results of thermal analysis

1) $\mathrm{C_6H_5NO_2 + CH_3COOH}$
  (fig. 1). Eutectic.

2) $\mathrm{C_6H_5NO_2 + C_6H_5COOH}$
  (fig. 2). Eutectic.

3) $\alpha\mathrm{C_{10}H_7NO_2 + CH_3COOH}$
  (fig. 3). A slight break at a molecular ratio of about 50:50.

4) $\alpha\mathrm{C_{10}H_7NO_2 + C_6H_5COOH}$
  (fig. 4). The maximum melting temperature was observed at 70 mole percent of the nitro compound and a break in the curve at 30%.

1. $\mathrm{C_6H_4(NO_2)_2 + C_6H_5COOH}$
     (fig. 5). Eutectic.

2. $\alpha\mathrm{C_{10}H_7NO_2 + C_6H_5OH}$
     (fig. 6). A break in the curve at a molecular ratio of nitro compound to phenol of about 70:30.

3. $\alpha\mathrm{C_{10}H_7NO_2 + \alpha C_{10}H_7OH}$
     (fig. 7). Eutectic.

4. $\alpha\mathrm{C_{10}H_7NO_2 + \beta C_{10}H_7OH}$
     (fig. 8). Eutectic.

5. $\alpha\mathrm{C_{10}H_7NO_2 + m C_6H_4(OH)_2}$
     (fig. 9). Maximum at a ratio of 70:30.

6. $\mathrm{C_6H_4(NO_2)_2 + C_6H_5OH}$
     (fig. 10). Eutectic.

7. $\mathrm{C_6H_4(NO_2)_2 + \alpha C_{10}H_7OH}$
     (fig. 11). Maximum at a ratio of 50:50.

8. $\mathrm{C_6H_4(NO_2)_2 + \beta C_{10}H_7OH}$
     (fig. 12). Eutectic.

Fig. 1

Labels in the figure:

• \(CH_3COOH\)
• \(C_6H_5NO_2\)
• “Acetic acid—nitrobenzene”
• \(16.6^\circ\)
• \(-5.4^\circ\)
• \(9^\circ\)
• \(10^\circ\)

Fig. 2

Labels in the figure:

• \(CH_3COOH\)
• \(C_6H_5COOH\)
• “Acetic acid—benzoic acid”
• \(16.6^\circ\)
• \(2^\circ\)
• \(120^\circ\), \(100^\circ\), \(80^\circ\), \(60^\circ\), \(40^\circ\), \(20^\circ\), \(0^\circ\)

Fig. 3

Labels in the figure:

• \(CH_3COOH\)
• \(\alpha C_{10}H_7NO_2\)
• “Acetic acid—\(\alpha\)-nitronaphthalene”
• \(16.8^\circ\)
• \(37.2^\circ\)
• \(70^\circ\), \(60^\circ\), \(50^\circ\), \(40^\circ\), \(30^\circ\), \(20^\circ\)

Fig. 4

Fig. 5

Fig. 6

Figure 7.
In-figure labels: \(\alpha\)-\(C_{10}H_7NO_2\); \(\alpha\)-\(C_{10}H_7OH\); “\(\alpha\)-nitronaphthalene—\(\alpha\)-naphthol.”
Visible point labels include: \(61.0^\circ\), \(33^\circ\), \(96^\circ\).

Figure 8.
In-figure labels: “\(\alpha\)-nitronaphthalene—\(\beta\)-naphthol”; \(\alpha\)-\(C_{10}H_7NO_2\); \(\beta\)-\(C_{10}H_7OH\).
Visible point labels include: \(61.0^\circ\), \(41.0^\circ\), \(122^\circ\).

Figure 9.
In-figure labels: \(\alpha\)-\(C_{10}H_7NO_2\); \(m\)-\(C_6H_4(OH)_2\); “\(\alpha\)-nitronaphthalene—resorcinol.”
Visible point label: \(2:1\).

Fig. 10

Labels in the figure:
\(m\,C_6H_4(NO_2)_2\); \(C_6H_5OH\); dinitrobenzene–phenol.

Fig. 11

Labels in the figure:
\(C_6H_4(NO_2)_2\); \(\alpha\,C_{10}H_7OH\); \(1:1\), \(50\%\).

Fig. 12

Labels in the figure:
\(C_6H_4(NO_2)_2\); \(\beta\,C_{10}H_7OH\); dinitrobenzene–\(\beta\)-naphthol.

1. $\mathrm{C_6H_4(NO_2)_2 + mC_6H_4(OH)_2}$
   (fig. 13). Maxima at ratios 70:30 and 50:50.

2. $\mathrm{pCH_3C_6H_4NO_2 + pCH_3C_6H_4NH_2}$
   (fig. 14). Eutectic.

3. $\mathrm{pCH_3C_6H_4NO_2 + \alpha C_{10}H_7NH_2}$
   (fig. 15). Eutectic.

4. $\mathrm{pCH_3C_6H_4NO_2 + \beta C_{10}H_7NH_2}$
   (fig. 16). Eutectic.

5. $\mathrm{pCH_3C_6H_4NO_2 + (C_6H_5)_2NH}$
   (fig. 17). Eutectic.

6. $\mathrm{pCH_3C_6H_4NO_2 + (C_6H_5CH_2)_3N}$
   (fig. 18). Eutectic.

7. $\mathrm{pCH_3C_6H_4NO_2 + (C_6H_4NH_2)_2}$
   (fig. 19). Eutectic.

8. $\mathrm{\alpha C_{10}H_7NO_2 + C_6H_5NH_2}$
   (fig. 20). Break at a ratio of about 50:50.

9. $\mathrm{\alpha C_{10}H_7NO_2 + pCH_3C_6H_4NH_2}$
   (fig. 21). Eutectic.

10. $\mathrm{\alpha C_{10}H_7NO_2 + C_6H_5N(CH_3)_2}$
    (fig. 22). Eutectic.

11. $\mathrm{\alpha C_{10}H_7NO_2 + (C_6H_4NH_2)_2}$
    (fig. 23). Eutectic.

12. $\mathrm{\alpha C_{10}H_7NO_2 + C_5H_{11}N}$
    (fig. 24). Weak break at a ratio of about 70:30.

13. $\mathrm{C_6H_4(NO_2)_2 + C_6H_5NH_2}$
    (fig. 25). Break in the curve at a ratio of 50:50.

14. $\mathrm{C_6H_4(NO_2)_2 + pCH_3C_6H_4NH_2}$
    (fig. 26). Eutectic.

15. $\mathrm{C_6H_4(NO_2)_2 + C_6H_5N(CH_3)_2}$
    (fig. 27). Weak break at a ratio of about 50:50.

16. $\mathrm{C_6H_4(NO_2)_2 + (C_6H_4NH_2)_2}$
    (fig. 28). Sharp maximum at a ratio of 50:50, corresponding to the composition of an easily isolated crystalline product.

17. $\mathrm{C_6H_4(NO_2)_2 + (C_6H_5)_2NH}$
    (fig. 29). Eutectic.

18. $\mathrm{C_6H_4(NO_2)_2 + (C_6H_5CH_2)_3N}$
    (fig. 30). Eutectic.

19. $\mathrm{C_6H_4(NO_2)_2 + C_5H_{11}N}$
    (fig. 31). Eutectic.

2. Results of colorimetric determinations

Colorimetric analysis, i.e., determination of the composition of compounds formed in the systems under study by the intensity of coloration, was first applied in the chemistry of organic complex compounds by I. I. Ostromyslensky. Ostromyslensky and, somewhat later, E. V. Biron and Morguleva used this method precisely in studying the reaction of nitro compounds with amines, which is also treated in our work.

We took the compounds under study in benzene solution. The measuring instrument used was a Duboscq colorimeter. For each system, the solution with the maximum coloration was taken as the unit of comparison. A comparison was also made of the intensities of the most strongly colored solutions of several systems. As in investigations by the electrochemical method and by the method of thermal analysis, in colorimetry we took series of samples differing in composition successively by 10 molecular percent of each constituent. The experiments were carried out at a temperature of 18–20°.

We present the results of the determinations.

1. $\mathrm{C_6H_5NO_2 + C_6H_5NH_2}$
   (fig. 32). The color of the solutions is yellow. Maximum intensity at a molecular ratio of 50:50.

2. $\mathrm{C_6H_5NO_2 + C_6H_5N(CH_3)_2}$
   (fig. 33). The color is orange. Maximum at a ratio of 50:50.

3. $\mathrm{C_6H_5NO_2 + (C_6H_4NH_2)_2}$
   (fig. 34). The color is orange. Maximum coloration at a ratio of 50:50.

4. $\mathrm{\alpha C_{10}H_7NO_2 + C_6H_5NH_2}$
   (fig. 35). The color is orange. Maximum at a ratio of 50:50.

5. $\mathrm{\alpha C_{10}H_7NO_2 + C_6H_5N(CH_3)_2}$
   (fig. 36). The color is orange. Maximum at a ratio of 50:50.

Fig. 13

“Dinitrobenzene—resorcinol”

Labels visible in the figure:
\(m\,C_6H_4(NO_2)_2\); \(m\,C_6H_4(OH)_2\)

Fig. 14

“Nitrotoluene—p-toluidine”

Labels visible in the figure:
\(p\,CH_3C_6H_4NO_2\); \(p\,CH_3C_6H_4NH_2\)

Fig. 15

“\(\alpha\)-Naphthylamine—nitrotoluene”

Labels visible in the figure:
\(\alpha\,C_{10}H_7NH_2\); \(p\,CH_3C_6H_4NO_2\)

Fig. 16

Labels in the figure: $\mathrm{CH_3 \cdot C_6H_4 \cdot NO_2}$; $\beta\,\mathrm{C_{10}H_7 \cdot NH_2}$; “$p$-nitrotoluene—$\beta$-naphthylamine.”

Fig. 17

Labels in the figure: $\mathrm{CH_3 \cdot C_6H_4 \cdot NO_2}$; $(\mathrm{C_6H_5})_2\mathrm{NH}$; “diphenylamine—$p$-nitrotoluene.”

Fig. 18

Labels in the figure: $\mathrm{CH_3 \cdot C_6H_4 \cdot NO_2}\,p$; $(\mathrm{C_6H_5 \cdot CH_2})_3\mathrm{N}$; “$p$-nitrotoluene—tribenzylamine.”

Fig. 19

Labels in the figure:

• \(CH_3 \cdot C_6H_4 \cdot NO_2\), \(p\)
• \((C_6H_4NH_2)_2\)
• “\(p\)-Nitrotoluene—benzidine”

Fig. 20

Labels in the figure:

• \(\alpha\, C_{10}H_7 \cdot NO_2\)
• \(C_6H_5 \cdot NH_2\)
• “\(\alpha\)-Nitronaphthalene—aniline”

Fig. 21

Labels in the figure:

• \(\alpha\, C_{10}H_7NO_2\)
• \(CH_3C_6H_4 \cdot NH_2\)
• “\(\alpha\)-Nitronaphthalene—\(p\)-toluidine”
• \(16.2^\circ\)

Fig. 22

Fig. 23

Fig. 24

Fig. 25.

System: “Dinitrobenzene–aniline”

Composition axis: from \(C_6H_4(NO_2)_2\) to \(C_6H_5NH_2\).
Temperature scale: \(0^\circ\)–\(100^\circ\).

Fig. 26.

System: “Dinitrobenzene–[[unclear: second component name, possibly beginning with “tetra-”]]”

Composition axis: from \(C_6H_4(NO_2)_2, m\) to \(C_{10}H_7NH_2\).
Temperature scale: \(0^\circ\)–\(80^\circ\).

Fig. 27.

System: “Dinitrobenzene–diphenylamine”

Composition axis: from \(C_6H_4(NO_2)_2\) to \((C_6H_5)_2NH\).
Temperature scale: \(0^\circ\)–\(100^\circ\).

Fig. 28

Fig. 29

Fig. 30

Fig. 31

Labels in the figure: \(C_6H_4(NO_2)_2\); \(C_5H_{11}N\); “dinitrobenzene–meta-piperidine”; \(-18^\circ\).

Fig. 32

Labels in the figure: \(C_6H_5NO_2\); \(C_6H_5NH_2\); “Conductivity of benzene solutions of the nitrobenzene–aniline system”; color yellow; \(t_k 20^\circ\); \(t 20^\circ\).

Fig. 33

Labels in the figure: \(C_6H_5NO_2\); \(C_6H_5C(CH_3)_2\); “nitrobenzene–dimethylaniline”; color yellow-orange; \(t 18^\circ\).

Fig. 34

Fig. 35

Fig. 36

1. $\alpha\mathrm{C}_{10}\mathrm{H}_{7}\mathrm{NO}_{2} + (\mathrm{C}_{6}\mathrm{H}_{4}\mathrm{NH}_{2})_{2}$ (Fig. 37). Color dark yellow (tea-colored). Maximum at the ratio $50:50$.

2. $\alpha\mathrm{C}_{10}\mathrm{H}_{7}\mathrm{NO}_{2} + (\mathrm{C}_{6}\mathrm{H}_{5})_{2}\mathrm{NH}$ (Fig. 38). Color red-orange. Maximum at the ratio $50:50$.

3. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + \mathrm{C}_{6}\mathrm{H}_{5}\mathrm{NH}_{2}$ (Fig. 39). Color orange. Maxima at the ratios $60:40$ and $40:60$, with a slight weakening of the color at $50:50$.

4. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + \mathrm{C}_{6}\mathrm{H}_{4}\mathrm{N}(\mathrm{CH}_{3})_{2}$ (Fig. 40). Color orange-red. Maximum at the ratio $50:50$.

5. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + \mathrm{CH}_{3}\!\cdot\!\mathrm{C}_{6}\mathrm{H}_{4}\!\cdot\!\mathrm{NH}_{2}$ (Fig. 41). Color orange. Maximum at the ratio $50:50$.

6. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + (\mathrm{C}_{6}\mathrm{H}_{4}\mathrm{NH}_{2})_{2}$ (Fig. 42). Color dark red. Maximum at the ratio $50:50$.

7. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + (\mathrm{C}_{6}\mathrm{H}_{5})_{2}\mathrm{NH}$ (Fig. 43). Color dark red. Maximum at the ratio $50:50$.

8. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + (\mathrm{C}_{5}\mathrm{H}_{11})_{2}\mathrm{NH}$ (Fig. 44). Color dark red at the large color maximum corresponding to the ratio $50:50$, and lighter at the small maximum at the ratio $30:70$.

9. $m\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2} + \mathrm{C}_{5}\mathrm{H}_{11}\mathrm{N}$ (Fig. 45). Color dark red, with greatest intensity at the ratio $90:10$.

In the experiments with dimethylaniline, a comparison was made of the color intensities of the maximally colored systems with different nitro compounds. The ratio of the color intensities of the complexes with nitrobenzene, $\alpha$-nitronaphthalene, and metadinitrobenzene was found to be $1:2.5:30$.

V. Discussion of the Results

1. Mechanism of formation of complexes of nitro compounds

The main task of the present work was to elucidate the electronic type of chemical activity of the nitro group in various complex-formation reactions. According to the requirements of quantum mechanics, nitrogen cannot have five covalent bonds. Consequently, the formula I usually adopted for nitro compounds cannot be correct. It must be replaced by formula II, with a tetracovalent positive nitrogen atom:

$$ \begin{aligned} \text{I.}\quad & \mathrm{R{-}N{=}O} \quad \text{or} \quad \mathrm{R{:}N{:}O} \\ & \phantom{\mathrm{R{-}}}\| \\ & \phantom{\mathrm{R{-}}}\mathrm{O} \end{aligned} \qquad \begin{aligned} & \\ & \\ & \mathrm{:O:} \end{aligned} $$

$$ \begin{aligned} \text{II.}\quad & \mathrm{R{-}\overset{+}{N}{-}\overset{-}{O}} \\ & \phantom{\mathrm{R{-}}}\| \\ & \phantom{\mathrm{R{-}}}\mathrm{O} \end{aligned} \quad \text{or} \quad \begin{aligned} & \mathrm{R{:}N{:}O{:}} \\ & \phantom{\mathrm{R{:}}}\ddot{\ } \\ & \phantom{\mathrm{R{:}}}\mathrm{:O:} \end{aligned} $$

The parachor of nitro compounds also argues in favor of the second formula, with a semipolar bond.

The polarity of the nitro group is also confirmed by the considerable magnitude of the dipole moments of nitro compounds.

Thus, a positive charge must be concentrated at the nitrogen atom, and a negative charge at the oxygen atom. Depending on the nature of the reagent acting, the nitro group may exhibit electron-donor or electron-acceptor reactivity. Its activity must, of course, also be influenced by other atoms entering into the molecule of the nitro compound.

Fig. 37

• Left component: $\alpha\,\mathrm{C}_{10}\mathrm{H}_{7}\mathrm{NO}_{2}$
• Right component: $(\mathrm{C}_{6}\mathrm{H}_{4}\mathrm{NH}_{2})_{2}$
• System: “$\alpha$-nitronaphthalene—benzidine”
• Annotation: crystals
• Coloration: tea-colored
• Temperature: $t\,20^\circ$
• Horizontal scale: $0$–$100\%$
• Vertical scale: $0$–$1.2$

Fig. 38

• Left component: $\alpha\,\mathrm{C}_{10}\mathrm{H}_{7}\mathrm{NO}_{2}$
• Right component: $(\mathrm{C}_{6}\mathrm{H}_{5})_{2}\mathrm{NH}$
• System: “$\alpha$-nitronaphthalene—diphenylamine”
• Coloration: reddish-orange
• Temperature: $t\,18^\circ$
• Horizontal scale: $0$–$100\%$
• Vertical scale: $0$–$1.2$

Fig. 39

• Left component: $\mathrm{C}_{6}\mathrm{H}_{4}(\mathrm{NO}_{2})_{2}\,m$
• Right component: $\mathrm{C}_{6}\mathrm{H}_{5}\mathrm{NH}_{2}$
• System: “meta-dinitrobenzene—aniline”
• Coloration: orange
• Temperature: $t\,18^\circ$
• Horizontal scale: $0$–$100\%$
• Vertical scale: $0$–$1.5$

Fig. 40

Visible labels in the graph:

• \(C_6H_4(NO_2)_2\,m\)
• \(C_6H_5C(C_6H_5)_3\)
• “m-Dinitrobenzene—dimethylaniline.”
• Color orange-yellow
• \(t\,18^\circ\)
• Composition scale: \(0\%\) to \(100\%\)

Fig. 41

Visible labels in the graph:

• \(C_6H_4(NO_2)_2\,m\)
• \(CH_3.C_6H_4NH_2\)
• “m-Dinitrobenzene—toluidine.”
• Color orange
• \(t\,18^\circ\)
• Composition scale: \(0\%\) to \(100\%\)

Fig. 42

Visible labels in the graph:

• \(C_6H_4(NO_2)_2\,m\)
• \((C_6H_4(NO_2)_2)\)
• “m-Dinitrobenzene—phenidine.”
• Crystals isolated, \(t_{\mathrm{m.p.}}\,128\text{–}129^\circ\)
• At \(50^\circ\), the solutions are burgundy-colored
• Composition scale: \(0\%\) to \(100\%\)

Fig. 43

Fig. 44

Fig. 45

Of the reagents with which nitro compounds can give complexes, metal salts, such as LiI, AlCl₃, etc., function chiefly as electron acceptors. Strong mineral acids, and apparently also carboxylic acids, belong here as well.

Complex formation with amines must most probably proceed through the positive nitrogen of the nitro group and the unshared electron pair of the amine nitrogen:

\[ \begin{array}{c} \ \ \ \ \ O \\ \ \ \ \ \ \| \\ \mathrm{C_6H_5{-}N{-}O} \\ \ \ \ \ \ \ \ \ \mathrm{NR_3} \end{array} \qquad \begin{array}{c} \ \ \ \ \ \overline{O} \\ \ \ \ \ \ \overset{+}{|}\ \ \ \overline{\ } \\ \mathrm{C_6H_5{-}\overset{+}{N}{-}O} \\ \ \ \ \ \ | \\ \ \ \ \ \mathrm{NR_3} \end{array} \]

However, in the case of aromatic amines, another type of complex formation is less probable but not excluded, likewise owing to possible positive poles in the amine molecule. It was therefore important to study systems that include amines of the saturated series or saturated heterocycles, for example piperidine. Since at our disposal there were only amines of this type that froze with difficulty, for determining the ability toward complex formation we used here chiefly not thermal analysis, but the colorimetric method. Metadinitrobenzene with diisoamylamine did indeed give a dark-red solution (in benzene), with a maximum color corresponding to the equimolecular composition

\[ \mathrm{C_6H_4(NO_2)_2\cdot (C_5H_{11})_2NH.} \]

With piperidine a very intense coloration was observed, but with a maximum at a composition not corresponding to a definite stoichiometric ratio. The maximum is shifted strongly toward dinitrobenzene (not less than 90% of the latter). Apparently this is connected with the fact that, in piperidine, which of all the amines has the strongest basic properties, the field of the negative charge of the nitrogen atom has the greatest intensity and can attract a considerable number of nitro-compound molecules. The fact that the appearance of color is observed equally in amines of different series indicates an identical mechanism of complex formation.

Phenols may exhibit a dual character of reactivity. On the one hand, they have clearly expressed acidic properties owing to the presence of the mobile hydrogen of the hydroxyl group. On the other hand, the oxygen atom has unshared electron pairs, which should create a negative field. Negative charges also arise on the carbons of the benzene ring in the ortho and para positions to the hydroxyl.

Since, nevertheless, the acidic properties predominate in phenols, it could be thought that, in the formation of complexes with nitro compounds, the hydrogen of the hydroxyl group participates:

\[ \begin{array}{c} \ \ \ \ \ \ \ \ + \ \ \ \overline{\ } \\ \mathrm{C_6H_5{-}N{-}O\ldots HO.C_6H_5} \\ \ \ \ \ \ \ \ \ \| \\ \ \ \ \ \ \ \ \ O \end{array} \]

However, in such a case the composition of the complex compounds becomes incomprehensible. In systems containing nitro compounds and phenols, the method of thermal analysis most often gave no positive result; mononitro compounds with monohydric phenols almost always gave

only eutectics. The more sensitive electrochemical method indicated the existence of complex compounds of the type

\[ \mathrm{ROH}\cdot 2\mathrm{R}\cdot \mathrm{NO_2}. \]

For one molecule of phenol there are two molecules of the nitro compound. Meanwhile, in acid complexes one molecule of the nitro compound adds one or two molecules of acid. Apparently, the mechanism of complex formation with acids and with phenols is different. The composition of the complexes of nitro compounds with phenols is well explained if it is assumed that here phenol exhibits an electron-donor function, acting by the pair of electrons of the oxygen atom; then the hydroxyl hydrogen is transferred to the oxygen of the nitro group, and the acid thus formed then adds a second nitro group through formation of a hydrogen bond with its oxygen:

\[ \mathrm{C_6H_5{-}N^{+}(=O)(-O^{-}) \cdots O(H){-}C_6H_5} \;\longrightarrow\; \mathrm{C_6H_5{-}N^{+}(-O^{-})(-OH)(-OC_6H_5)} \;\xrightarrow{\ \mathrm{C_6H_5NO_2}\ }\; \mathrm{C_6H_5{-}N^{+}(-O^{-})(-OC_6H_5){-}O{-}H\cdots O^{-}{-}N^{+}(=O){-}C_6H_5} \]

To test this assumption we decided to prepare the complex of nitrobenzene with sodium phenolate. Phenolate, being to a large extent dissociated into ions, should, through the oxygen of its anion, add still more readily to the nitrogen of the nitro group. However, the complex formed

\[ \left[ \begin{array}{c} \mathrm{O^{-}}\\[-2pt] |\\[-2pt] \mathrm{C_6H_5{-}N^{+}{-}O^{-}}\\[-2pt] |\\[-2pt] \mathrm{C_6H_5{-}O} \end{array} \right]\mathrm{Na^{+}} \]

should no longer react with a second molecule of the nitro compound.

We dissolved 2.3 g of metallic sodium in alcohol, added 9.4 g of phenol (1 g-mol per 1 g-atom of sodium) and 12.3 g of nitrobenzene, i.e., 2 g-mol per 1 g-mol of phenol. The alcohol and excess nitrobenzene were then distilled off in vacuo. The dry solid orange-colored substance remaining in the flask (further distillation led to decoloration) was analyzed for nitrobenzene and sodium. 3.294 g of the complex were dissolved in water, whereupon the color disappeared, and extracted with ether. After distillation of the ether, 1.500 g of \(\mathrm{C_6H_5NO_2}\) remained. The aqueous-alkaline solution was acidified with hydrochloric acid, again extracted with ether to remove the phenol, evaporated to dryness, and the remaining sodium chloride was lightly ignited; 0.878 g of it was obtained, which corresponds to 0.345 g of sodium.

\[ \begin{array}{llll} \text{For} & \mathrm{C_6H_5ONa}\cdot \mathrm{C_6H_5NO_2} & \text{calculated:} & \\ & \mathrm{C_6H_5NO_2}\ldots 51.45\% & \mathrm{Na} & 9.62\%\\ \text{Found} & \mathrm{C_6H_5NO_2}\ldots 46\% & \mathrm{Na} & 10.48\% \end{array} \]

The analytical results are fairly close to a complex of composition 1 : 1, i.e., the second molecule of nitrobenzene has not been added here.

The color of the phenolate complex with nitrobenzene, similar to the color of amine complexes, is also consistent with the interpretation set out above for the reaction mechanism. Free phenols with nitro compounds did not give noticeable coloration, evidently because the negative field of the phenolic oxygen is weak and its attraction to the positive pole of the nitro-

the nitro group cannot cause such a change in the motion of the electrons as is caused by the stronger field of the amine nitrogen. In the phenolate ion the negative field attains greater intensity, and the formation of a complex with the nitrogen of the nitro group here has a stronger effect on the motion of the electrons.

Complex formation of nitro compounds with hydrocarbons has as yet been studied too little for it to be possible to say anything definite about the reaction mechanism.

A substantial shortcoming of the available experimental material on this question is that almost all complexes of this type have been obtained with picric acid, while it is at the same time a nitro compound with three nitro groups and a strong acid. Therefore, without additional investigations it is impossible to decide which part of the picric acid molecule enters into the reaction in each particular case.

2. Comparison of Methods of Investigation

The classical method of physicochemical analysis, well developed and yielding reliable results, is thermal analysis. However, in the study of systems that form very unstable complex compounds, this method often proves insufficiently sensitive. It gives indistinct, and sometimes even negative, results in those cases when the compound formed cannot withstand the temperature to which the mixture has to be heated.

As our numerous experiments have shown, the electrochemical method of B. V. Tronov and L. P. Kulev is more sensitive. In this way it has been possible to detect complex compounds in many systems for which thermal analysis indicated only eutectic mixtures. These include a large part of the systems composed of nitro compounds with carboxylic acids and phenols.

Colorimetric analysis proved to be a very useful supplement. In thermal analysis, nitro compounds with amines almost always also gave eutectics in our work, but in all cases they gave a noticeable coloration.

As for determining the composition of complex compounds, previous work in our laboratory has already made it possible to conclude that, in general, thermal analysis and the electrochemical method lead to coincident results in those cases when both methods indicate the formation of complex compounds. The formation of complexes capable of direct isolation is also clearly registered in electrochemical investigation. The same must be said of the colorimetric method.

However, exceptions to the general rule were also observed. Thus, the fusibility diagram of the nitronaphthalene—benzoic acid system gave a maximum melting temperature at 70 mole percent of nitro compound, whereas the electrochemical method never indicated such a composition for complexes of nitro compounds with acids.

In addition to determining the electronic type of chemical activity of nitro compounds and other substances entering with them into the complex-formation reaction, we were interested in the question of the influence on reactivity exerted directly by the functional group and by the composition and structure of the entire molecule. This influence can be judged both from the data of physicochemical analysis and from the results of direct determinations of the stability of complex compounds.

Sharp maxima on fusibility diagrams and on emf curves indicate a considerable energy of chemical interaction. Comparing various nitro compounds in this respect, one may say that two

nitro groups in the meta position mutually enhance their electron-acceptor activity and weaken their electron-donor activity.

The electronic scheme makes this clear:

$$ \begin{array}{c} \overset{+\,-}{\mathrm{NO_2}}\\ \ \ |\\ -\ \mathrm{C}\\ {}^{+}\!/\ \backslash^{+}\\ \mathrm{HC}\quad \mathrm{CH}\\ \ \ | \quad \|\\ -\mathrm{HC}\quad \mathrm{C}-\overset{-}{\mathrm{NO_2}}\\ \ \ \backslash\!/\quad +\\ \mathrm{CH}\\ + \end{array} $$

The positive charge of the nitrogen atom of each nitro group is strengthened by the positive nitrogen of the second nitro group and weakened by oxygen, but nitrogen is closer and acts more strongly. The negative charge of oxygen, on the contrary, is weakened by nitrogen and strengthened by oxygen, and again, owing to the difference in distances, the influence of nitrogen is stronger. Hence, in comparison with nitrobenzene, metadinitrobenzene should have increased electron-acceptor activity and decreased electron-donor activity. This effect should be expressed still more strongly in 1,3,5-trinitrobenzene (perhaps this is why the latter is capable of giving a red coloration with alkali?).

This conclusion is confirmed by a whole series of diverse determinations. In experiments with ternary complexes, only trinitro compounds gave good results. In thermal analysis of mixtures with phenols, it was precisely metadinitrobenzene that gave a noticeable maximum with $\alpha$-naphthol and two maxima with resorcinol, but a eutectic with benzoic acid, whereas for nitronaphthalene with $\mathrm{C_6H_5COOH}$ a maximum of the melting temperature was observed. It is precisely $\mathrm{C_6H_4(NO_2)_2}$ that forms a stable complex with benzidine. It also gives breaks on the fusibility curves with aniline, dimethylaniline, and piperidine. After metadinitrobenzene, in ability to form complexes with amines, followed $\alpha$-nitronaphthalene. Paranitrotoluene, in which, under the influence of the methyl group, the electron-acceptor function should be especially weakened and the electron-donor function strengthened, gave only eutectics with all amines. Nitrobenzene in this respect has been little studied because of experimental difficulties (liquid).

Determination of the emf in mixtures of nitro compounds with acids, where the nitro compounds exhibit electron-donor activity, showed a weaker interaction for $\mathrm{C_6H_4(NO_2)_2}$ than for $\mathrm{C_6H_5NO_2}$.

From the intensity of the coloration in colorimetric analysis one can also sometimes judge the strength of the influence of the composition and structure of the molecule on the activity of the nitro group. Very clear results in this sense were obtained by comparison of nitrobenzene, nitronaphthalene, and dinitrobenzene in experiments with dimethylaniline. But when different amines are compared, a very confused picture is obtained here. Aniline with dinitrobenzene gave a coloration that was not very dark (orange), considerably lighter than diisoamylamine and piperidine, which have strong basic properties. However, a dark-red coloration was also obtained with the very weak base diphenylamine.

A good means of determining electron-donor or electron-acceptor activity should be measurement of vapor elasticity. For these measurements, systems must be selected whose components differ sharply from one another in volatility. Such determinations have been successfully applied in the chemistry of inorganic complex compounds; for exam—

measure, in the study of crystalline hydrates and crystalline ammoniates. We have so far carried out several reconnaissance determinations.

Cups containing three crystalline nitro compounds—p-nitrotoluene, α-nitronaphthalene, and m-dinitrobenzene—were placed in one desiccator; paratoluidine, more volatile than the nitro compounds taken, was poured onto the bottom of the desiccator. Each nitro compound should correspond to a definite vapor pressure of toluidine, corresponding to the stability of the complex formed by this nitro compound. If the difference in the activity of the nitro compounds is large, then first one of them should become saturated with the amine, then another, and finally the third, since the vapor pressure of the amine established over the more active nitro compound is insufficient for the formation of complexes by the less active compounds. The course of the process here is conveniently followed by the appearance of color.

The experiment showed that dinitrobenzene began to become colored first. When its color reached considerable intensity (probably the surface layer of the crystals had become saturated), the crystals of nitronaphthalene began to redden, and last of all nitrotoluene became colored. The advantage of the method is that no solvent is needed here; consequently, one can observe the formation, and even carry out the synthesis, of such complex compounds as are decomposed by any solvent.

Even with a simplified investigation procedure, which does not give an exact quantitative characterization of the stability of the complexes, it is possible to compile activity series of complex-formers of one type with respect to one and the same complex-former of another type. The application of the method is, of course, limited by certain conditions. In particular, it is necessary that the complex compounds formed not be too stable. Color facilitates observation of complex formation, but the course of the reaction can also be followed by the increase in weight.

In applying this method, the reaction of displacement of one complex-former by another from an already prepared complex compound may also be used.

Conclusions

1. To elucidate the mechanism of reactions in organic chemistry and to study the mutual influence of atoms in molecules, it is very important to determine the type of chemical activity (reactivity) of various functional groups in organic compounds. Electronic chemistry distinguishes three types of reactivity: electron-donor, electron-acceptor, and atomic-radical. In studying, from this point of view, compounds with the first two types of activity, it is best to use complex-formation reactions that are not accompanied by a more profound change in the molecules.

2. According to the principles of quantum physics, the nitrogen of the nitro group can have a covalent double bond with only one oxygen atom. The second nitrogen–oxygen bond must be semipolar. Consequently, the nitro group must have two active centers—a positive one (nitrogen) and a negative one (oxygen)—and must exhibit both electron-acceptor and electron-donor activity.

3. The authors studied systems in which, among the nitro compounds, nitrobenzene, p-nitrotoluene, α-nitronaphthalene, and m-dinitrobenzene were included; the second components were carboxylic acids, phenols, and amines. These systems were investigated by methods of thermal and calorimetric analysis.

4. As shown by the literature data on complexes formed by nitro compounds and by the results of experiments carried out by the authors, aromati-

nitro compounds do indeed exhibit a dual activity: electron-donor and electron-acceptor. Electron-donor reactivity is manifested in complex formation with acids, and electron-acceptor reactivity with amines. A new conclusion is that, in reactions with phenols, nitro compounds function not as electron donors, but as electron acceptors. This explains the difference in the composition of the complexes formed by nitro compounds with acids and with phenols.

For the complex of nitrobenzene with phenol, the following structural formula is proposed

\[ \mathrm{C_6H_5{-}\overset{+}{N}\!\left(\overset{-}{O}\right)(OC_6H_5){-}OH \cdots \overset{-}{O}{-}\overset{+}{N}(=O){-}C_6H_5} \]

The correctness of this formula is confirmed by the fact that sodium phenolate adds only one molecule of nitrobenzene, giving a compound of probable structure

\[ \left[ \mathrm{C_6H_5{-}N\!\left(\overset{O}{\phantom{O}}\right)(OC_6H_5){-}O} \right]^{-}\mathrm{Na}^{+} \]

1. Of the research methods used in the work, thermal analysis proved to be less sensitive. More sensitive were calorimetric analysis and the electrochemical method of B. V. Tronov and L. P. Kulev.

In those cases where thermal analysis nevertheless indicates the presence of chemical interaction between the components of the system, the conclusions concerning the composition of the compounds formed largely coincided with those given by electrochemical and calorimetric measurements.

LITERATURE

1. Report of the Commission of the Academy of Sciences of the USSR, “The State of the Theory of Chemical Structure in Organic Chemistry,” 1951.

2. B. N. Menshutkin. On compounds of aluminum chloride with nitrated benzene hydrocarbons and their substituted derivatives, Journal of the Russian Physico-Chemical Society, 42, p. 59, 1909.

3. Resolution of the meeting on the theory of chemical structure in organic chemistry, Proceedings of the Academy of Sciences of the USSR, Division of Chemical Sciences, No. 5, pp. 652–654, 1951. The same in the journals Advances in Chemistry, Journal of General Chemistry, Journal of Physical Chemistry.

4. A. Remick. Electronic Concepts in Organic Chemistry, p. 293, 1950.

5. B. V. Tronov. On reactions of cleavage and formation of double bonds, Proceedings of the Tomsk Technological Institute, vol. 45, No. 3, pp. 1–8, 1924.

6. B. V. Tronov and A. P. Grigor’eva. Nitration of nitrobenzene with nitric acid under various conditions, Transactions of the Siberian Physico-Technical Institute, vol. 1, No. 4, pp. 7–13, 1932.

7. B. V. Tronov, L. N. D’yakonova, and E. A. Zonova. Complexes of aromatic polynitro compounds with unsaturated compounds and certain salts, Journal of the Russian Physico-Chemical Society, 59, 333–345, 1927.

8. B. V. Tronov and L. P. Kulev. A new electrochemical method for the study of organic compounds and their mixtures, Proceedings of TPI, 64, 3–89, 1948.

9. B. V. Tronov and G. I. Strebkov. Activity of halides in halogen-substituted simple ethers, Journal of the Russian Chemical Society, 61, 2389, 1929.

10. M. F. Shostakovskii. Reports of the Academy of Sciences of the USSR, 16, 124, 1943.

11. Beilsteins Handbuch der Organischen Chemie, 4th ed., vol. V.

12. Beilsteins Handbuch der Organischen Chemie, 4th ed., vol. VI.

13. Cherbuliez, Chem. Zbl., 1923, I/II, 1941.

14. J. Ostromislenski, Ber. Deutsch. Chem. Gesellschaft, 44, 268–273, 1911.

15. P. Pfeiffer. Organische Molekülverbindungen, 1922.