Organic Compounds in Hydroiodic Acid Under High Pressure (Part 2)* -The Addition Reaction of Carbon Monoxide with Ethylene

119

Studies on the Kinetics of Addition Reactions of Carbon Monoxide with Organic Compounds in Hydroiodic Acid under High Pressure (Part 2)* -The Addition Reaction of Carbon Monoxide with Ethylene-

by Hiroshi Teranishi**, Kumao Hamanoue**, Shinichi Hori** and Toshiharu Takagi**

Summary: The addition reaction of carbon monoxidewith ethylene in aqueous hydroiodic acid was investigatedunder high pressurein the range of 30 to 90kg/cm2. Theyield of propionic acid was fairly high and the overall rate equation was:

Namely, the apparent reaction was a pseudofirst order with respect to carbon monoxide and

ethylene, and the apparent rate constants were 3.0×10-5, 13.1×10-5 and 26.8×10-5 sec-1 at

100, 120 and 140℃, respectiaely. From these values, K/PCO's calculated were 0.05×10-5,

0.21×10-5 and 0.43×10-5kg-1・cm2・sec-1 comparable to 0.08×10-5, 0.20×10-5 and

0.30×10-5kg-1・cm2・sec-1 obtained in the reaction of carbon monoxide with ethyl iodide. The overall activationenergy has beenfound to be 16.6kcal/mole, as opposedto that of the latter reaction, i.e., 13kcal/mole. From these results the authors concludedthat ethyl iodide was formed as an intermediateproduct in the reactionof carbon monoxidewith ethylene.

1)17). 1 Introduction From these considerations, we have already Since 1933 when several patents1) on the pro- studied the kinetics of addition reaction of carbon duction of carboxylic acids from olefins and carbon monoxide with ethyl iodide in hydroiodic acid monoxide were published, many works have been under high pressure, and the result shows that the carried out in various media, such as concentra- reaction proceeds via C2H5+I-H3O+I- as an inter- ted sulfuric acid2)～7), phosphoric acid8), anhydro- mediate, and the reaction rate is proportional us hydrogen fluoride4),7), monohydroxyfluorboric to the concentrations of ethyl iodide and carbon acid and its mixtures with phosphoric or sulfuric monoxide (loc. cit.). In the present work, the acid9) and aqueous boron trifluoride10),11) Those authors will report the results of kinetic experiments olefins higher than C4 reacted quite readily with on the addition reaction of carbon monoxide with carbon monoxidc12)～14). On the other hand, those ethylene. olefins lower than C3 reacted with carbon monoxide only under more severe conditions15). Eight 2 Experimental years ago Kanbara et al. obtained propionic acid Commercial ethylene was used and a small in good yields through the reaction of ethyl fluoride amount of ethane was detected by gas chromato- in aqueous hydrogen fluoride16) graphy. In these reactions, it might be expected that Carbon monoxide, kindly supplied by the Ins- the activity of acidic catalysts is directly related to titute for Chemical Research, Kyoto University, the acid strength. However, from the consider- was prepared from formic acid and hot concentra- ation of acidity function, we have concluded that ted sulfuric acid, and its purity was better than even a strong acid, such as conc. HCl or HBr, 97% as determined by gas chromatography. is insufficient to act as a catalyst under atmospheric Hydroiodic acid (G.R. grade, 57wt% aqueous pressure, and HI is the strongest acid among HF, solution of HI), silver carbonate (C. P. grade) and HCl, HBr as shown in our previous paper (Part sodium hydroxide (E. P. grade) were all purchased from Nakarai Chemicals, Ltd., and were used * Received May 26, 1977. without further purification. ** Department of Chemistry , Kyoto Institute of Techno- logy (Matsugasaki, Sakyo-ku, Kyoto 606) The reaction was carried out in a glass vessel

Volume 19, No. 2, November 1977 120 Teranishi, Hamanoue, Hori and Takagi: Studies on the Kinetics of Addition Reactions of

inserted into a stainless steel autoclave (capacity: suggesting that stirring speed of 1,000rpm was 250ml) equipped with a magnetic stirrer. Weigh- sufficient to make the liquid phase reaction the ed amounts of hydroiodic acid were charged into rate-determining under the conditions used. the autoclave. After the temperature of the auto- Figure 2 shows the effect of temperature on the clave had become constant at the desired level, reaction. The maximum yield was obtained at a gaseous mixture of carbon monoxide and ethylene near 140℃. Above 140℃, the reaction involved was introduced and stirred. Just after the some side reactions such as polymerization, and introduction of the mixture gas, the concentra- some dark brown oily material was obtained at tion of ethylene in the gas phase was analysed by the end of run. gas chromatography, and from this result the Figure 3 shows the effect of pressure on the yield amount of dissolved ethylene was calculated. At of acid at 140℃, and at 65kg/cm2 the yield of the end of reaction, the autoclave was quickly propionic acid was found to be maximum. cooled with ice water. The aqueous solution was Figure 4 shows the relationship between the neutralized with Ag2CO3 to remove the hydrogen mole ratio of HI to C2H4 and the yield of acid. iodide and filtered, then the filtrate which con- tained the Ag-salt of organic acid was converted into free acid by passing it through a column of exchange resin (Amberlite 120B) and was analysed with a gas chromatography (Hitachi K53) equipped with a flame ionization detector. A column (2m×3mm) with 20% polyethylene glycol or SE-30 on 60/80 mesh Chromsorb W-NAW was used. Nitrogen was used as the carrier gas. The observed acid was propionic acid, and it was quantitatively determined by titration with aqueous 0.1 N-NaOH. Prior to quantitative analysis, Fig. 2 Effect of Temperature on Yield of Propionic several mixtures of propionic acid and hydroiodic Acid at Reaction Time 120min acid were prepared, and they were analysed by the same procedure as mentioned above to obtain the factor of propionic acid yield. The yield of propionic acid was obtained by dividing the total amount of propionic acid by the amount of dissolved ethylene.

3 Results

Figure 1 shows the relationship between the stirring speed and the rate of propionic acid formation. Above 1,000rpm the reaction rate under Fig. 3 Effect of Pressure on Yield of Propionic Acid the conditions given in the figure was not affected,

Fig. 1 Effect of Stirring Speed on Yield of Propionic Fig. 4 Effect of Mole Ratio of HI to C2H4 on Yield Acid of Propionic Acid

Bulletin of The Japan Petroleum Institute Carbon Monoxide with Organic Compounds in Hydroiodic Acid under High Pressure (Part 2) 121

intermediatecation (may be a carbonium cation), and an equilibrium reaction exists. The ionic intermediatewill combine with the dissolvedcar- bon monoxide to produce propionyl iodide,fol- lowed by the fasthydrolyzation reaction to form propionic acid. Taking into account that HI is present in large excess over C2H4 and C2H5I, reactions (1) and (2) are assumed to be pseud first order reactions; reaction (4) is very fast and the rate of formation Fig. 5 Effect of Temperature on Yield of Propionic of propionic acid is equal to that of propionyl Acid iodide. The result indicated that the yield of propionic Assuming the rate determining process to be acid increased with the mole ratio and it became reaction (3) and a steady state condition for the constant at the mole ratios above 60. intermediate cation, the following rate equation From the results described above, the following is derived: conditions were adopted for the kinetic measure- ments: reaction temperatures lower than 140℃; HI/C2H4 mole ratio 45; stirring speed 1,000rpm; total pressure 65kg/cm2. where HCO is Henry's constant for CO in the solu- Effects of temperature on yields examined at tion and PCO the pressure of CO over the reaction 100, 120 and 140℃ are shown in Fig. 5. The medium. results showed that the initial rate of formation In the derivation of the above equation, it is of propionic acid increased with increasing tem- assumed that CO is present in large excess over perature. Compareing the results at 120℃ and C2H5I and its concentration is constant during the at 140℃, the saturation yield is higher at the reaction. lower temperature. And the total pressure was From equation (1), increased at the long run reaction. These results suggest that at the higher temperature some side reactions or decomposition reactions may have Integrating this equation, occurred as the reaction time became longer.

then the rate of disappearance of C2H5I is given by 4 Discussion the following equation: Following the discussion in Part 1, the main reaction mechanism may be considered as follows:

Substituting the steady state concentration of C2H5+I-H3O+I- in the above equation,

and so, The boiling point of aqueous hydroiodic acid was 127℃ at atmospheric pressure, and it might still have remained as an aqueous solution under the experimental conditions: therefore, the re- Integrating this equation, action is a liquid phase reaction. Reaction (1) is a well known addition reaction of hydrogen halides to olefins, and it may be very fast. Reactions (2) through (4) have already been considered in Part 1, namely ethyl iodide in Consequently, the yield of propionic acid is given aqueous HI dissociates partly to produce an ionic by the following equation,

Volume 19, No. 2, November 1977 122 Teranishi, Hamanoue, Hori and Takagi: Studies on the Kinetics of Addition Reactions of

Fig. 7 Arrhenius Plot of Overall Rate Constant

to koverall in Part 1. The values thus obtained are given in Table 1. As was discussed in Part 1, k-1 was very large compared with k2HCO and K/PCO was constant. Therefore, the latter value must be consistent with that of Part 1. All of these values are also listed in the table, and the result obtained is satisfacotry. Figure 7 is an Arrhenius plot of the overall Fig. 6 Plot of log {[C2H4]0/([C2H4]0-[C2H6COOH]} rate constant resulting in a straight line. The s. Reaction Time v slope gives the overall activation energy of 16.6

Table 1 Effect of Temperature on the overall kcal/mole which contains the molar enthalpy Rate Constant change of reaction (2) and the activation energy of reaction (3). From the above discussion one can conclude as follows: Propionic acid can be produced in good yields in the HI catalyzed carbonylation reaction when ethylene is used as the raw material. The rate equation derived from the proposed mechanism involving the liquid phase reaction of dissolved carbon monoxide with ethyl cation as the rate determining, which is a first order reaction with respect to CO pressure and to C2H4, has been shown to elucidate the observed facts satisfactorily. In the gas phase radical reaction of Eq. (1), The overall activation energy is found to be 16.6kcal/mole, which contains the molar enthalpy k0 is of the order of 10-5 sec－1 (at 200℃); however, change of reaction (2), i.e., the heat of dissociation in the present case, the reaction is in liquid phase of C2H5I and the activation energy of reaction and ionic; therefore, k0 must be larger than 10-5 sec-1. On the other hand, K is of the order of (3), namely, the reaction between ethyl cation and carbon monoxide to form propionyl iodide. 10-5scc-1(at 140℃) as shown in Part 1. Thus, k0 can be larger than K, and Eq. (10) becomes Acknowledgement

Eq. (11). The authors wish to express their sincere thanks to Prof. Yoshimasa Takezaki of Institute for Chemical Research, Kyoto University for supply- Figure 6 shows the plot of ln{[C2H4]0/([C2H4]0 -[C2H5COOH])} against the reaction time, and ing carbon monoxide. Thanks are also due to The Asahi Foundation for the Contribution to Industri- it is in good linear relationship consistent with al Technology for financial support. Eq. (11). Thus the slope of this straight line References gives the apparent rate constant, namely K= k1k2HCOPCO/(k-1+k2HCOPCO)showing to be equal 1) Carpenter, G. B., U. S. 1, 924, 762～8 (1933).

Bulletin of The Japan Petroleum Institute Carbon Monoxide with Organic Compounds in Hydroiodic Acid under High Pressure (Part 2) 123

2) Koch, H., Brennstoff-Chem.,36, 321 (1955). (1961). 3) Koch, H., Rev. dei Combustibili, 10, 77 (1956). 13) Friedman, B. S., Cotton, S. M., J. Org. Chem., 26, 4) Koch, H., Belg. 518, 682 (1955); Brit. 743, 597 3751 (1961). (1957); U. S. 2, 831, 877 (1958). 14) Friedman, B. S., Cotton, S. M., ibid., 27, 481 (1962). 5) Koch, H., Haaf, W., Angew. Chem., 70, 311 (1958). 15) Takezaki, Y., Fuchigami, Y., Teranishi, H., Sugita, 6) Koch, H., Haaf, W., Ann. der Chem., 618, 251 (1958). N., Kudo, K., Bull. Japan Petrol. Inst., 8, 31 7) Koch, H., Ger. 972, 315 (1961). (1966). 8) Koch, H., U. S. 3, 061, 621 (1963). 16) Kanbara, M., Sugita, N., Kudo, K., Teranishi, H., 9) Koch, H., Huisken, W., U. S. 2, 876, 241 (1959). Takezaki, Y., ibid., 11, 48 (1969). 10) Koch, H., Haaf, W., Ann. der Chem., 638, 111 (1960). 17) Teranishi, H., Hamanoue, K., Manki, Y., Takagi, 11) Koch, H., Ger. 1, 095, 802 (1962). T., ibid., 18, 62 (1976). 12) Friedman, B. S., Cotton, S. M., U. S. 2, 975, 199

Volume 19, No. 2, November 1977