The Chemistry of Ethylene Oxide

Public Channel / Chemical Fact Sheets

0 0

Share on Social Networks

Share Link

Use permanent link to share in social media

Share with a friend

Please login to send this document by email!

Embed in your website

Select page to start with

6. C..INADIAiV JOURNAL OF CHEAIISTRE'. 1'01,. 30 Fig. 5. The acid-catalyzed hydrolysis of ethylene oxide. 0 =HC104 alone; .=HC104+NaC104; X = HClOd-tLiC101. I'T Fig. 6. Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

5. E/ISTHAbl AND LA TXEAIOUILLE: CHEMISTRY OF ETHYLENE OXIDE. 1'. 173 I I I I I I 0 I -2 -3 -4 -5 SALT CONCENTRATION (m/L.) Fig. 3. Reaction of ethylene oxide with iodide ion in neutral solution. .0031 .0032 .0033 'IT Fig. 4. Activation energies for the uncatalyzed addition of halide ions to ethylene oxide. Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

1. Canadian Journal of Chemistry Issued by THE NATIONAL RESEARCH COUNCIL OF CANADA VOLUME 30 MARCH. 1952 NUMBER 3 THE CHEMISTRY OF ETHYLENE OXIDE V. THE REACTION OF ETHYLENE OXIDE WITH HALIDE IONS IN NEUTRAL AND ACID SOLUTION1 ABSTRACT The rates of reaction of halide ions with ethylene oxide in neutral aqueous solution and the rate of hydrolysis of ethylene oxide in acid solution have been measured and the activation energies determined. From these data and from the ratio of glycol to chlorohydrin formed when ethylene oxide reacts with excess aqueous hydrogen halide, the rates of the acid-catalyzed addition of halide ions to ethylene oxide at 25°C. have been estimated. In aqueous solution ethylene oxide reacts with halide ions to form halo- hydrins. These reactions were studied for a number of substituted oxides at 20°C. by Bronsted, Kilpatrick, and Icilpatrick (1) and shown to occur by both uncatalyzed and acid-catalyzed processes. The uncatalyzed reactions CH? - CH2 + X- $- H20 S XCHZCH2OH + OH- \/ 0 were followed by measuring the rate of acid addition required to keep the solution neutral and were usually sufficiently rapid that the uncatalyzed hydrolysis of the oxide did not seriously interfere with the measurements. In acid solution, however, the formation of glycol was an important side reaction and the accurate measurement of the rate of the acid-catalyzed addition of halide ions was therefore difficult. Bronsted et al. succeeded in making such measurements for the bromide-glycid system by measuring the rate of change in the conductivity of the solution, and established that the reaction was of the third order, dependent upon the oxide, hydrogen ion, and bromide ion concentrations. The method was laborious and was therefore applied only to the one reaction, but the authors did record some approximate rate constants for the acid- catalyzed addition of chloride and bromide ions to ethylene oxide. In the course of work in these laboratories it became desirable to extend . some of these data to ethylene oxide itself and accordingly, the uncatalyzed -. ..& Alanzrscribt received Sebtember 17. 1951. Contribution'from the ~iv'ision of Chemistry (Applied), National Research Council. Issrred as N.R.C. No. 2634. Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

4. 172 CANADIAN JOURNAL OF CIIEBIISTRY. I'OL. 30 45°C. NaCL k- (2.15 - 0.40~) l0~'lI./mol/min. 2.0 I I I 2.6 25'C. NaCL k ~(2.55-0.20~) l./mol/rnin. 0 .2 .4 .6 .8 . 1.0 1.2 C1-( m/L.) Fig. 1. Reaction of ethylene oxide with chloride ion in neutral solution. I I I I I I 0 1 .2 .3 .4 -5 -6 Br- ( rn/l.) Fig. 2. Reaction of ethylene oxide with bromide ion in neutral solutio~i. Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

3. EASTHAM AND LATREMOUILLE: CHEMISTXI' OF ETHYLENE OXIDE. V. 171 concentration was varied from about 2 to 10 times that of the oxide. The reaction mixture was then analyzed for halohydrin by acid-base titration or for glycol by periodate oxiclation, or both, using methocls very similar to those previously described (3). The analyses were carefully tested on synthetic mixtures of acid, glycol, and halohydrin before use. No difficulty was exper- ienced with chloride mixtures and bromides gave good results when pro- cedures were standardized and blanks carefully determinecl. The determination of glycol in the iodicle solutions by means of periodate, however, was only accomplisl~ed after first precipitating the excess iodide with silver nitrate and then rapidly removing the excess si!ver with sodium chloride; under these conditions, reproducible and apparently accurate analyses were obtained. No appreciable amount of free iodine was liberated during the reaction of oxide with the iodide. Reactions with hydrofluoric acid were conducted in polystyrene containers. The glycol/chlorohydrin ratios as cletermined by glycol analysis were in reasonable agreement with those obtained by acid-base titration at the lower halide concentrations but above about 0.5 inolar discrepancies were observed (Fig. 7). The deviations may have been due to the formation of polymeric products such as diethylene glycol, etc. Altering the HX/NaX ratio did not appear to influence the deviation. Measurements were made at 25" and at 0" but the ratios were not sufficiently sensitive to temperature to permit an estimate of the activation energies. The data were plotted (Fig. 7) and the rate constants estimated from the smoothed curve. RESULTS The rates for the uncatalyzed addition of halide ions to ethylene oxide are shown in Figs. 1, 2, 3, and the activation energies for these reactions are shown in Fig. 4. In estimating the activation energies, rates were compared at a concentration of halide ions common to the three temperatures rather than at zero concentration to avoid extrapolation errors. It is perhaps of interest to note that lithi~im chloride seems to give a positive salt effect whereas sodium chloride gives the usual negative salt effect in these reactions (Fig: 1). This difference may simply be due to the fact that the activity of lithium chloride increases while that of sodium chloride decreases in the concentration range of these experiments. It is however of interest to note that some recent experiments (to be published) in pyridine solution have led us to believe that lithium ions may indeed have some catalytic activity towards ethylene oxide reactions. Figs. 5 and 6 show the rates, salt effects, and activation energy for the acid- catalyzed hydrolysis of ethylene oxide. The data appear to be in good agree- ment with those of Bronsted et al. Fig. 7 shows the ratio of glycol to halohydrin obtained when ethylene oxide is allowed to react with acid solutions of halide ions. From the smoothed curves the acid-catalyzed rate constants were calculated and are shown in Table I; it is evident that the bromide and iodide constants show a consider- Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

7. EASTHAM AND LAZ'REBIOUILLE: ClIEhIlSTRY OF ETIfYLENE OXIDE. 1'. 175 o FLUORIDE- - BY GLYCOL DETERMINATION 0 BY ACID-BASE TITRATION -0-0- 25OC. A- 0-c. I- a: W 60 $- BROMIDE 6a- I I I ! I I I I 0 .2 .4 .6 .8 1 .O 1.2 X-CONCENTRATION (m/l.) Fig. 7. able drift towards lower rates as the concentration increases and the recorded average values must therefore be regarded as very approximate. The approxi- mate values reported by Bronsted et al. are included for comparison. The results of all these experiments, together with the corresponding data for a number of other reactions of ethylene oxide, are shown in Table 11. TABLE I Halide ion k,+ at 25" for the formation of: concentration, -- moles per liter I UIycoI 1 Bromohydrin , ' (l.n~I.~mii.~) Chlorohydrin (1.~n1ol.-%in.~~)1 Iodohydrin .025 .050 .075 .10 .15 .20 .30 .40 .50 Average Bronsted el al. Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

8. 176 CANADIAN JOURNAL OF CHEMISTXI'. 1'OL. 30 TABLE I1 - - Reagent 1 Urlca talyzed 1 Acid-catalyzed I- B r- C1- F- OH-* HzO** Most amirlest NH3tt Temp. Activation energy a+ ( 1 (ha,.) 1 (I2nol2min.) Ik 33 7.0 2.2 Very small ,010 Activation energy (kcal.) * Lichtezstein and Twigg (5). ** The experinrentally determined rate constants for the hydrolysis reactions Itave been divided by the concentration of water (55.5 moles per liter) in order to give thenz the same dimen- sions as the other constants. t Eastham el al. (2). tt Ferrero et al. (4). REFERENCES 1. BRONSTED, J. N., KILPATRICK, M., and KILPATRICK, M. J. Am. Chem. Soc. 51: 428. 1929. 2. EASTHAM, A. M., DARWENT, B. DEB., and BEAUBIEN, P. E. Can. J. Chem. 29: 575. 1951. 3. EASTHAM, A. M. and LATREMOUILLE, G. A. Can. J. Research, B, 28: 264. 1950. 4. FERRERO, P., BERBE, F., and FLAMME, L. R. Bull. soc. chim. Belges, 56; 349. 1947. 5. LICHTENSTEIN, H. J. and TWIGG, G. H. Trans. Faraday Soc. 44: 905. 1948. Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.

2. 170 CANADIAN JOURNAL OF CHEBIISI'KY. I'OL. 30 rates of halide ion addition were measured, using a slight modification of Bronsted's method. The rates were sufficiently slow to suggest that an estimate of the acid-catalyzed rates could be made by measuring the ratio of glycol to chlorohydrin formed from ethylene oxide in a large excess of acid. This ratio, as pointed out by Liclitenstein and Twigg (5) is expressed approx- imately by the equation Glycol ki = - (X-) Chlorohydrin kz where kl and kp are the rate constants for the acid-catalyzed formation of glycol and of halohydrin respectively and where (X-) is the concentration of halide ion. The value of k1 may be determined independently and therefore by using a large excess of halide ion to minimize concentration changes, the value of kz nlay be estimated from the glycol/halohydrin ratio, on the assump- tion that these are the only products formed. EXPERIMENTAL (a) The rates of the uncatalyzed addition of halide ions to ethylene oxide were deter~nined as follows. A four-necked flask fitted with a mechanical stirrer, a 5 ml. burette and the external electrodes of a Beckinan model G pH meter was held in a constant temperature bath regulated to O.Ol°C. The burette was graduated in units of 0.01 cc. and was tipped with a small stainless steel hypodermic needle to permit the addition of very small drops of liquid. A solution of alkali metal halide was adjusted to pH 7.0 with the corresponding acid and a lcnown volun~e introduced into the flask and allowed to come to bath temperature. A weighed ampoule of ethylene oxide was broken iinder the solution, the stirrer started, and acid was then added from the burette at such a rate that the pH of the solution was kept between G and 8. Measure- ments were made only in the first 5y0 of reaction and the concentration of acid was so adjusted that the increase in volume of the solution during the period of measurement was not over 2%. The chloride and bromide reactions were titrated with the corresponding acids and calculated as pseudo-first order reactions but, because of the instability of hydrogen iodide, the rate of iodide addition was followed with perchloric acid and calculated by second order methods. Sodium perchlorate was used as a neutral salt in the determina- tion of salt effects. (b) The rate of the acid-catalyzed hydrolysis of ethylene oxide was de- termined dilatometrically according to the method of Bronsted et al. and corrected for the iincatalyzed reaction by means of the data of Lichtenstein and Twigg (5). (c) The ratio of glycol to ~hlorohyd~in formed in the reaction of ethylene oxide with acid solutions of halide ions was determined by the following procedure. A weighed sample of ethylene oxide was broken into a hydrogen halide-sodium halide solution of known volurne and concentration and allowed to react completely; the initial halide ion concentration of the solution was always at least 10 times that of the ethylene oxide but the hydrogen ion Can. J. Chem. Downloaded from by on 08/31/16 For personal use only.


  • 77 Total Views
  • 58 Website Views
  • 19 Embedded Views


  • 0 Social Shares
  • 0 Dislikes

Share count

  • 0 Facebook
  • 0 Twitter
  • 0 LinkedIn
  • 0 Google+

Embeds 1

  • 1