STUDY OF THE COMPOSITION OF CHLORINATION PRODUCTS OF THE ETHYLENE FRACTION OF COKE OVEN GAS

A. M. MIROSHNIKOV, V. V. KLESHCHEVA, E. I. SIDOROVICH

(Presented by the scientific seminar of the Faculty of Chemical Technology)

At the Kemerovo Nitrogen Fertilizer Plant, for the production of dichloroethane (DCE), an ethylene fraction obtained by separation of coke oven gas at low temperatures by the Linde method is used. The commercial product thereby met only the requirements for the lowest grade of DCE.

Improving the quality of commercial DCE raised the question of the precise determination of the qualitative and quantitative composition of impurities in the reaction mass and the study of the pathways of their formation, with the aim of subsequently separating them from the main product.

The authors carried out analyses of the raw material, commercial DCE, production wastes, and unreacted residues. For this purpose, a set of classical methods of analytical control was used in combination with gas chromatography (gas–liquid and gas–adsorption). The results of the analyses are summarized in Table 1.

The conditions of the chromatographic analyses are given in Table 2. The chromatograms were calculated by the method of internal normalization using derived sensitivity coefficients [9, 10].

The requirements for improving the quality of commercial DCE and increasing its yield could not be met without knowledge of the mechanism of the chlorination reactions of the ethylene fraction. Sufficiently complete literature data on the chlorination of ethylene-containing raw materials similar in composition to the ethylene fraction are lacking. However, there are indications [3, 6] that higher homologues of ethylene are chlorinated first. The chlorination of pure ethylene has been studied many times. It has been proved that the addition of chlorine to ethylene can proceed by reactions of two types [1]:

Ionic type of reaction

$$ \mathrm{CH_2{=}CH_2} \ \xrightarrow{+\mathrm{Cl}^{+}}\ \mathrm{CH_2{-}CH_2^{+}} \ \xrightarrow{+\mathrm{Cl}^{-}}\ \mathrm{CH_2{-}CH_2} $$

$$ \begin{array}{ccc} & \big| & \big| \\ & \mathrm{Cl} & \mathrm{Cl} \end{array} $$

Radical type of reaction

a) in the absence of peroxides

$$ \mathrm{CH_2{=}CH_2} \ \xrightarrow{+\mathrm{Cl_2}}\ \mathrm{CH_2{-}CH_2\cdot} \ \xrightarrow{+\mathrm{Cl_2}}\ \mathrm{CH_2{-}CH_2} $$

$$ \begin{array}{ccc} & \big| & \big| \\ & \mathrm{Cl} & \mathrm{Cl} \end{array} $$

Table 1

Analyses of feedstock and products in dichloroethane production, %

*) Qualitative determination.

b) in the presence of peroxides

\[ \mathrm{CH_2{=}CH_2} \ \xrightarrow{+\mathrm{Cl}\cdot}\ \mathrm{CH_2{-}CH_2} \ \xrightarrow{+\mathrm{Cl}\cdot}\ \mathrm{CH_2{-}CH_2} \]
\[ \begin{array}{c} \phantom{\mathrm{CH_2{-}}}\big| \\ \phantom{\mathrm{CH_2{-}}}\mathrm{Cl} \end{array} \qquad\qquad \begin{array}{cc} \big| & \big| \\ \mathrm{Cl} & \mathrm{Cl} \end{array} \]
\[ \begin{array}{c} \mathrm{CH_2{-}CH_2} \\ \big| \\ \mathrm{Cl} \end{array} \ \xrightarrow[\ -\mathrm{HCl}\ ]{+3\mathrm{Cl}\cdot}\ \begin{array}{cc} \mathrm{CHCl{-}CH_2} \\ \big| \quad \big| \\ \mathrm{Cl} \quad \mathrm{Cl} \end{array} \ \xrightarrow[\ -\mathrm{HCl}\ ]{+\mathrm{Cl}\cdot}\ \begin{array}{cc} \mathrm{CHCl{-}CHCl} \\ \big| \quad \big| \\ \mathrm{Cl} \quad \mathrm{Cl} \end{array} \]

Along with the addition reaction, a substitution reaction also proceeds. None of the authors gives a clear distinction between the types of reactions and fully takes into account the presence of peroxides in industrial chlorination [1].

Table 2

Conditions of chromatographic analyses

This circumstance is apparently the basis for the differing views of a number of authors on the chemistry of obtaining DCE and on the design of the technological process. At the same time, it is recommended that the process be carried out under conditions that are often quite contradictory. For example, according to some data the chlorination temperature should be no higher than 20—30°C [3], while according to others it should be 80—120°C (in a homogeneous medium) and even 150°C (in a heterogeneous medium) [7, 8]. It is recommended to carry out the reaction with an excess of ethylene [3, 4, 6] and with an excess of chlorine [3, 4], in an anhydrous medium [8] and in the presence of moisture [3], without the addition of catalysts [3] and in their presence [3, 7, 8]. In our study, in the DCE feedstock a series of products was found whose formation does not fit into the scheme of ethylene chlorination (Table 3). For example, the formation of chlorinated methanes was observed, although the initial feedstock contained no methane. This fact indicates that chlorinolysis reactions proceed even at low temperatures. Water apparently accelerates this process, since when the reaction is carried out in an aqueous medium (hypochlorination), chlorinated methanes are formed in amounts an order of magnitude greater.

Table 3

Results of analyses of dichloroethane feedstock

Content (wt. %) of components in samples

It is also interesting to note the appearance of chlorobenzene in the reaction products, although benzene was absent from the initial feedstock. Apparently, chlorobenzene is formed via ethylene chlorohydrin, since experiments we carried out on the chlorination of ethylene oxide in an EDC medium confirmed the formation of chlorobenzene.

In many samples the content of tetrachloroethane is greater than that of trichloroethane; consequently, tetrachloroethane is formed not only through the metalepsis reaction. Since the acetylene present in the feedstock does not participate in the reaction (its total content in the fraction does not change), the source of tetra- (as well as penta- and hexa-) chloroethanes is apparently the radicals $\mathrm{CHCl_2\cdot}$ and $\mathrm{CHCl_3\cdot}$, which may also be responsible for the formation of chlorinated methane derivatives. It is characteristic that an increase in the content of the light fraction (predominantly chlorinated methanes) in the chlorination products occurs simultaneously with an increase in the amount of highly chlorinated hydrocarbons (Table 3, samples 5, 7, 11). Within the known limits, additions of iron salts (0.02–0.085 g/l) and water (0.02–0.058%) do not exert any noticeable influence on the results of chlorination. The ethylene content in the off-gases within the range of 3–18% is not reflected in the composition of the chlorination products. It was also observed that an increase in the electrical conductivity of the EDC feedstock is associated with an increase in the amount of addition products.

The data obtained in our studies allow one to assume that, in the chlorination of the ethylene fraction, both mechanisms take place, and the composition of the reaction products depends on them.

Since the development of an ionic type of addition reaction is promoted by an increase in the concentration of ions in solution and by an increase in electrical conductivity, we examined the effect of additions of ferric chloride and water. The experiments showed that the activity of ferric chloride increases in the presence of water, which is also reflected in an increase in the electrical conductivity of the system (Fig. 1). When ferric chloride is added together with water, the specific gravity of the EDC feedstock usually decreases, i.e., the amount of polychlorinated compounds diminishes; in the absence of water with $\mathrm{FeCl_3}$, however, an increase in the specific gravity of the feedstock is observed.

Fig. 1.

The addition of water within the limits of solubility increases the electrical conductivity, but with a further increase in its content it separates into a separate phase, hydrolyzes ferric chloride, and extracts ions from the organic phase, which leads to a decrease in the activity of ferric chloride. It was observed that zinc chloride has no catalytic activity.

The development of the radical mechanism of the chlorination reaction is favored by an increase in temperature, as well as by the presence of polychlorides, which decompose more readily into radicals than mono- and dichloro derivatives. Metals of variable valence, including iron, may also be sources of radicals. Since it has been established that chlorination reactions may be branched-chain reactions [2], the main task in the chlorination of the ethylene fraction is the prevention-

... the occurrence (or at least the weakening) of the development of a branched chain reaction, in which ethylene is consumed not only for the formation of dichloroethane but also for other products, for example, chlorinated methanes. This can be done either by accelerating the ionic process, which we have verified using ferric chloride and water as an example, or by inhibiting chain reactions through the addition of inhibitors. We have established that, for example, the addition of phenol increases the yield of dichloroethane based on reacted ethylene. Taking into account that metals of variable valence can serve as initiators of both ionic and radical reactions, it is necessary to maintain the catalyst concentration in the state of greatest electrical conductivity of the system.

Conclusions

1. The composition of the feedstock, the reaction products, and the unreacted residues in the production of dichloroethane by chlorination of the ethylene fraction of coke-oven gas has been investigated.

2. It has been shown that the composition of the reaction products depends on the ratio of the ionic and radical reaction mechanisms and can be regulated by the composition of the feedstock and by catalyst additives.

References

1. O. A. Reutov. Theoretical Foundations of Organic Chemistry. Moscow State University, 1964.

2. B. A. Krentsel. Chlorination of Paraffinic Hydrocarbons. Nauka, 1964.

3. A. I. Brodovich. Ethylene from coke-oven gas as a raw material for organic synthesis. Metallurgizdat, 1963.

4. F. Asinger. Chemistry and Technology of Monoolefins. Gostoptekhizdat, 1960.

5. M. S. Malinovskii. Olefin Oxides and Their Derivatives. Goskhimizdat, 1961.

6. Yu. Rudenko. Dichloroethane. Oborongiz, 1939.

7. U.S. Patent 2929852, 1960.

8. British Patent 960083, 1961.

9. V. G. Martur et al. Zavodskaya laboratoriya, No. 9, 1964.

10. Z. O. Smolyan, N. D. Demina. Proceedings on Chemistry and Chemical Technology, No. 2, 433, Gorky, 1964.