STUDY OF THE FORMALDEHYDE METHOD FOR PURIFYING BENZENE FROM THIOPHENE. FIRST COMMUNICATION

N. M. SMOLYANINOVA, M. I. ALEKHIN, I. A. MYKOLNIKOV,
V. N. SITARKINA

(Presented by the scientific-methodological seminar of the KhTF)

One of the numerous types of raw material for the chemical industry is benzene, the consumption of which in the coming years will increase sharply, chiefly in connection with the production of caprolactam based on cyclohexane.

Despite the great prospects for the use of petroleum raw materials for the production of aromatic hydrocarbons, at present the main supplier of benzene is the coke-chemical industry. However, the commercial products produced until recently by coke-chemical enterprises cannot meet the increased requirements for the quality of these products as raw material for the production of modern synthetic materials. These requirements particularly concern the content of carbon disulfide and thiophene in benzene. Thus, the new standard for coal-tar benzene (GOST 8448-61) limits the content of carbon disulfide and thiophene in benzenes of the higher categories: “especially pure” to not more than 0.0001% of each, and “pure for synthesis,” grade 1, to not more than 0.0001 and 0.0005% of carbon disulfide and thiophene, respectively.

Obtaining such a product under the technological conditions existing at coke-chemical plants is extremely difficult and is associated with large expenditures of raw materials and reagents.

The required degree of purification of benzene from carbon disulfide can be achieved by increasing the efficiency of the carbon-disulfide column; therefore, the problem of obtaining deeply purified grades of benzene is reduced mainly to the removal of thiophene.

Since catalytic hydropurification of crude benzene, which provides a high degree of desulfurization of pure products, proves economically justified only for feedstock of a definite quality and for large-scale production [1, 2], researchers’ efforts have been directed toward finding chemical methods for the de-thiophenation of benzene.

Analysis of literature data and exploratory studies carried out by us [3] showed that the required degree of purification of benzene from thiophene is achieved when the formaldehyde method is used. Its essence consists in the more pronounced ability of thiophene, as compared with benzene hydrocarbons, to condense in the presence of an acid catalyst with formaldehyde, with the formation of polymers containing methylene bridge bonds [4, 5].

In the works of Soviet [6, 7, 8] and foreign authors [4, 5, 9], commercial grades of coal-tar benzene were used mainly as the starting material.

Of great interest from an economic point of view is the process of removing thiophene with formaldehyde at the stage of chemical purification of the benzene–toluene (BT) or benzene–toluene–xylene (BTX) fractions of crude benzene. This would make it possible to reduce substantially the cost of thiophene-free benzene by eliminating secondary rectification of the product and reducing the consumption of sulfuric acid, as well as by making it possible to carry out the process on the ordinary equipment of the washing section.

There is almost no information in the literature on the formaldehyde method for desulfurizing BT or BTX fractions.

The present work is devoted to studying the process of removing thiophene from benzene by formaldehyde purification of the unwashed BT fraction.

The work was carried out with the BT fraction from the coke-chemical production of the Kuznetsk Metallurgical Combine; its characteristics are given in Table 1.

Table 1

Characteristics of the BT fraction of the KMK

Content of pure products, %

Group composition of sulfur compounds, %

As reagents, 93% sulfuric acid (GOST 4204-48) and technical formalin containing 35.6% formaldehyde and 9.3% methanol were used. The thiophene content in benzene was determined by the photocolorimetric method according to GOST 8448-61.

The experiments were carried out at room temperature in a two-necked round-bottom flask equipped with a stirrer and a dropping funnel for adding reagents.

In the first experiments, a procedure was used that provided for the simultaneous addition of formalin and acid to the product being purified. It turned out that in this way it was impossible to purify the unwashed BT fraction from thiophene, even with a very large excess of reagents. The reason for the negative effect lay in the particular features of the chemical composition of the BT fraction as compared with pure benzene, the purification of which by the same procedure proceeded easily and rapidly. As is known, unwashed BTF contains a considerable amount of unsaturated hydrocarbons. The simultaneous introduction of formalin and sulfuric acid leads to the preferential occurrence of reactions of formaldehyde with unsaturated compounds, owing to their greater reactivity, with the formation of 1,3-glycols, unsaturated alcohols, and functional derivatives of these compounds [4].

Thus, formaldehyde is removed from the sphere of reaction with thiophene, and purification does not occur.

Taking the foregoing into account, the following series of experiments was carried out by a modified procedure—with stepwise addition of reagents: first (while the fraction was being stirred) sulfuric acid was added, thereby ensuring the removal of unsaturated compounds from the reaction sphere with formaldehyde by means of their polymerization; then, after a certain time, formalin was added. The results obtained showed that, with stepwise addition of the reagents, purification of the BT fraction from thiophene proceeds normally at a sulfuric-acid consumption of 2% and a formalin consumption of 1% relative to the volume of the product being purified.

In order to find the optimum conditions for the process, the influence of various factors (sulfuric-acid concentration, formalin consumption, and the duration of stirring with formalin and with acid) on the degree of desulfurization of the BT fraction was investigated.

The influence of temperature was not studied, since the reaction proceeded well under room conditions.

The reagent consumptions were adopted on the basis of previous experiments; the duration of stirring of the BT fraction with acid was 20 min, and with formalin the same.

After completion of the process, the mixture is allowed to settle for 2–3 min. The purified fraction is separated from the acid and the high-molecular-weight polycondensation products, after which the benzene cut (79–85°C) is distilled from it; after drying over calcium chloride, it is submitted for analysis. The thiophene content was determined photocolorimetrically; unsaturation was characterized by the degree of coloration with sulfuric acid according to GOST 2706–54.

The results of the experiments are presented in Figs. 1–4 and in Table 2.

Fig. 1. Dependence of the thiophene content in benzene on the concentration of sulfuric acid. Ratio BTK : H₂SO₄ : CH₂O = 100 : 2 : 1. Stirring time with H₂SO₄—20 min. Stirring time with CH₂O—20 min.

As can be seen from Fig. 1, with an increase in acid concentration the degree of purification of benzene increases. Desulfurization practically does not proceed at acid concentrations below 70%. Purification proceeds most effectively in the presence of concentrated sulfuric acid, which ensures a decrease in the unsaturation of the product being purified to the required limit, at which formaldehyde reacts predominantly with thiophene. At low sulfuric-acid concentrations, polymerization reactions practically do not proceed, just as no interaction of formaldehyde with thiophene occurs; for thiophene, according to the data of N. K. Moshchinskaya [6], an acid concentration of not less than 85% is required (Table 2).

As the results of the experiments presented in Table 2 showed, the optimum formalin consumption, ensuring under these conditions a high degree of desulfurization of the BT fraction, is 1% of the initial

of feedstock, which corresponds to a consumption of 0.36% of 100% formaldehyde, or a 20-fold excess relative to the theoretical amount [4].

Table 2

Effect of the amount of formaldehyde on the degree of desulfurization of benzene

Figures 2 and 3 present curves characterizing the kinetics of desulfurization as a function of the duration of stirring of the BT fraction with acid (Fig. 2) and with formalin (Fig. 3). These data show that, as the contact time of the reagents increases in both the first and the second stages of the purification process, the purification efficiency increases.

It is interesting to note: the shorter the duration of stirring with acid, the longer the contact with formalin must be. Apparently, for the rapid course of the polycondensation reaction of thiophene with formaldehyde, it is necessary to reduce the content of “active” unsaturated components of the BT fraction to some definite limit. As our experiments showed (Fig. 4), for this BT fraction this limit may be taken as the removal of unsaturates in the first purification stage to a coloration of the benzene with sulfuric acid of less than 1 standard-scale unit, which is ensured by a certain contact time with sulfuric acid.

From analysis of the curves in Figs. 2 and 3, the optimum duration of stirring of the fraction with sulfuric acid may be taken as 10 min, and with formalin likewise 10 min, with an overall duration of the purification process of 20 min (all this, of course, at the given intensity of stirring).

Experiments carried out under conditions of more intensive stirring (approximately 1000 revolutions of the stirrer per minute) showed that in this way the overall duration of the purification process can be substantially reduced, down to 4 min.

On the basis of the results presented, the following conditions for the process of formaldehyde purification of the BT fraction from thiophene can be recommended as optimal:

1) sulfuric acid concentration — 93%, at a consumption of 2% (by volume), or about 4% by weight;

2) consumption of 36% formalin — 1% by volume, or about 0.4% calculated as 100% formaldehyde;

3) total stirring duration — 20 min, including:
a) with acid — 10 min,
b) with formalin — 10 min;

4) temperature — room temperature.

To assess the possibility of carrying out this process in a washing apparatus (melanger), similar in design to those usually used in production for sulfuric-acid purification of the BT fraction, experiments were carried out

scaled-up experiments were carried out in a laboratory glass stirrer with a capacity of 1 liter under optimal conditions.

The experiments showed that, when the formaldehyde method of desulfurizing the BT fraction is carried out, there is difficulty in discharging from the apparatus the high-molecular-weight polycondensation products formed during purification, in admixture with spent sulfuric acid, which have low fluidity.

The polycondensation products, after evaporation of the benzene hydrocarbons entrained by them, neutralized with alkali, washed with hot water to a neutral reaction, and dried in vacuum to constant weight, were a finely dispersed amorphous powder of light-brown color, not melting when heated above 320°C, with the evolution of straw-yellow smoke. The powder is insoluble in any of the widely used solvents; in 93% sulfuric acid it disperses, coloring the acid black. The elemental composition of the condensation product corresponds to the content: C—79.49%; H—7.26%; S—5.78%; O + N 7.47%. Its yield was about 1% of the BT fraction.

The infusibility and insolubility of the polymer obtained, as indicated by a number of authors [4, 5], testify to the spatial structure of this high-molecular-weight compound.

Fig. 2. Dependence of the thiophene content in benzene on the duration of stirring of the BT fraction with sulfuric acid: 1 — stirring time with formalin — 0 min.; 2 — 7 min.; 3 — 5 min.; 4 — 10 min.

Fig. 3. Dependence of the thiophene content in benzene on the duration of stirring of the BT fraction with formalin: 1 — stirring time with sulfuric acid — 1 min.; 2—5 min.; 3—10 min.

To check the experimental data on formaldehyde purification of the BT fraction obtained under laboratory conditions, semi-works tests of the method were conducted at the Kuznetsk Metallurgical Combine.

For carrying out the experiments, a washing apparatus with a capacity of 100 liters was designed and fabricated.

The BT fraction subjected to purification was taken from the intermediate receiver of the rectification department at a temperature of 45–50°C.

Since the quality of the raw material was not constant, the boiling limits and bromine number of the fraction were determined for each experiment (GOST 2706-54).

For purification, technical sulfuric acid with a concentration of 91.7% and 34.2% formalin were used; neutralization of the fraction was carried out with 18% sodium hydroxide.

Since the efficiency of the mixing device in the experimental melangeur was very low, it did not seem possible to make use of laboratory data on the duration of mixing the fraction with the reagents.

Fig. 4. Dependence of the thiophene content in benzene on the unsaturation of the BT fraction. Ratios BTK : H₂SO₄ : CH₂O = 100 : 2 : 1; mixing time—20 min.

Therefore, a procedure for both stages of the process was first worked out. The duration of mixing with sulfuric acid was determined on the basis of the need to ensure that the coloration of the benzene fraction with sulfuric acid was no more than 1 st. of the scale, while the reaction time with formalin was determined by the negative reaction of the selected sample for thiophene (with isatin-sulfuric acid). It was found that mixing with acid required 60 min, and with formalin—40 min.

The time for feeding the reagents was taken in accordance with the production technological instruction for washing the BT fraction. The reagent consumption corresponded to that determined by us under laboratory conditions, i.e., sulfuric acid (91.7%)—3.5% (by weight); 34.2% formalin—1.0% (by weight); 18% alkali—3.4% (by weight).

A single charge of the melangeur amounted to 52.5 kg of the BT fraction.

The washed product obtained after distillation was analyzed qualitatively for thiophene content and for degree of coloration with acid.

The results of the semi-plant tests are given in Tables 3 and 4.

As follows from the data in the tables, the semi-plant tests fully confirmed the possibility of deep desulfurization of benzene by formaldehyde purification of the crude BT fraction when this process is combined with the usual process of sulfuric-acid washing of the same fraction.

It is characteristic that the process of benzene desulfurization proved insensitive to changes in the quality of the BT fraction: despite considerable fluctuations in the bromine numbers and in the distillation temperatures of 95% of the volume

fraction (Table 3), the removal of thiophene proceeded well in all cases.

The yield of pure products was not determined because of the imperfection of the experimental unit (insufficient tightness of the apparatus, manual charging of the fraction and reagents, etc.), which led to considerable losses of the fraction. However, the yield of sulfur-free benzene was approximately within the same limits as in the usual sulfuric-acid purification of the BT fraction.

Table 3

Characteristics of the distillate after formaldehyde purification as a function of the quality of the feedstock (BT fractions)

The benzene obtained met, in all indices, the “especially pure” grade (Table 4).

Table 4

Characteristics of benzene obtained after formaldehyde purification

The proposed method of removing thiophene from benzene has a number of advantages in comparison with the sulfuric-acid method:

1. Lower acid consumption (3.5–4.0% versus 10–30% in sulfuric-acid afterpurification of pure benzene [10, 12]).
2. Shorter process duration (2–4 hours versus 8–9 hours [11, 12]).
3. Absence of secondary rectification.
4. Smaller benzene losses.

It should be noted that during the semi-industrial experiments there were also difficulties in discharging high-molecular-weight condensation products from the washing apparatus, especially in the case of purification of the BT fraction with bromine numbers above 3.

With a lower degree of unsaturation of the product being purified, less viscous polymers are formed.

To resolve the question of the possibility of discharging polymers from an ordinary mixer, special investigations must be carried out. Undoubtedly, the properties of the indicated products will be determined both by the quality of the BT fraction and by the conditions under which the process is conducted.

Preliminary techno-economic calculations showed that the formaldehyde method for desulfurizing benzene, provided the process is combined with the primary washing of the BT fraction and purification is carried out in an ordinary washing apparatus, has high economic efficiency.

Conclusions

1. The formaldehyde method for purifying the BT fraction from thiophene has been investigated.

2. It has been shown that benzene of the highest grades can be obtained by treating the raw BT fraction with formaldehyde in the presence of sulfuric acid, while combining the process of primary washing of the starting raw material with formaldehyde purification.

3. The optimal process conditions have been determined.

4. It has been established that the duration of purification is determined by the intensity of stirring and may range from 2 hours to 4 minutes. This indicates that the efficiency of the process can be significantly increased by switching to a continuous purification scheme, which will ensure substantially better contact of the fraction with the reagents.

5. A semi-industrial verification of the method has been carried out, showing its high efficiency.

6. A drawback of the proposed purification method is the difficulties associated with discharging polymers from the washing apparatus.

References

1. Ferrous Metallurgy of the Capitalist Countries. Part VI. Metallurgizdat, 251, 1956.

2. V. E. Privalov. Features of the Processing of Chemical Products of Coking in England. Metallurgiya Publishing House, Moscow, 1964.

3. Production of Sulfur-Free Benzene. Report on Topic 21/62, TPI, Tomsk, 1963.

4. J. Walker. Formaldehyde. GKhI, 1957.

5. P. D. Caesar and A. H. Sachanen. Ind. Eng. Chem., 40, 922, 1948.

6. N. K. Moshinskaya, V. N. Nosach, B. S. Bouden. Proceedings of the Dzerzhinsky DKhTI, issue 7, 57, 1960.

7. L. M. Cherkasova and L. K. Gorin. Coke and Chemistry, No. 1, 44, 1963.

8. R. I. Popov, I. Ya. Rashkevich et al. Coke and Chemistry, No. 4, 37, 1964.

9. Badische Anilin- und Soda-Fabrik. German Patent, 211239, 1909.

10. O. A. Semenova et al. Coke and Chemistry, No. 1, 41, 1963.

11. V. N. Sigarkina, S. I. Sukhenko. Production of Sulfur-Free Benzene at KMK. Report of Coke-Chemical Production of KMK, Novokuznetsk, 1962.

12. A. P. Kolesov, V. Z. Sokolov. Coke and Chemistry, No. 5, 1962.