Dehydrochlorination of 2-chloroethanol, 2-chloro-1-propanol, 1-chloro-2-propanol, 2-chloro-2-methyl-1-propanol and 1-chloro-2-methyl-2-propanol

The reactions between a few 1,2-chlorohydrins and sodium hydroxide have been studied and shown to involve a two-step nucleophilic elimination of hydrogen chloride. The data are given for the slow rate-determining step of 2-chloroethanol 1 , 2-chloro-1-propanol 2 , 1-chloro-2-propanol 3 , 2-chloro-2-methyl-1-propanol 4 and 1-chloro-2-methyl-2-propanol 5 . Compounds 4 and 5 gave 2-methyl-1,2-propanediol as the final product instead of oxiranes given by compounds 1 − 3 . In contrast to some earlier reports the mere water reaction was shown to be almost negligible. In constant ionic strength the base concentration had no effect on the rates whereas at different base concentrations (0.050 − 0.250 mol dm − 3 ) alone the rate of alkaline dehydrochlorination of 1 clearly decreased (10 3 k 2 , dm 3 mol − 1 s − 1 : 10.0 − 8.7, respectively). The rate of 2 at constant base concentration (0.010 mol dm − 3 ) and at different ionic strengths (dm 3 mol − 1 : 0.010 − 0.500) decreased also (10 3 k 2 , dm 3 mol − 1 s − 1 : 76 − 65, respectively) indicating that the decrease is mainly due to the change in the ionic strength also in the former case.


Introduction
−6 Most of the reports deal with the preparation of hydrogen chloride 7 or different oxiranes. 8,9The best report is that of Ballinger and Long 10 on the dehydrochlorination of 2-chloroethanol 1 which also summarizes some early studies.They proved that the reaction between 1 and sodium hydroxide is a specific hydroxide ion catalysis via a two-step mechanism including the prior equilibrium and the slow ratedetermining step: ClCH2CH2OH + HO − ⇌ ClCH2CH2O − + H2O (fast) k2 −13 This is why we decided to study the alkaline dehydrochlorination of 2-chloroethanol 1, 2-chloro-1-propanol 2, 1-chloro-2propanol 3, 1-chloro-2-methyl-2-propanol 4 and 2-chloro-2-methyl-1-propanol 5 as well as their neutral water reactions in some detail.

Results and Discussion
Based on the first-order rate constants of the alkaline dehydrochlorination of 1−3 it can be concluded that the dehydrochlorination caused by water reported earlier 11 is in fact practically negligible.The studies of Radulescu and Muresanu 12 on the neutral hydrolysis of 2-chloroethanol 1 pointed out that at the temperatures used in our study the dehydrochlorination caused by water is insignificant as shown also by our study on the neutral reaction.The second-order rate constants reported in literature for the latter are not significant in comparison with the rate constants for the alkaline dehydrohalogenation. 13−17 This is also true for 4 and 5 as shown for 4. Our measurements carried out within the base concentration 0.010−0.100mol dm −3 at a constant ionic strength (0.100 mol dm −3 adjusted by NaNO3, Table 1) do not support a conclusion that the second-order rate constant increases along with decreasing base concentration 18−21 supposing that the ionic strength is kept constant; obviously this observation was due to a small specific salt effect.Instead the hydroxide-ion concentration exhibits a primary salt effect in larger than 0.100 mol dm −3 concentrations (Table 2).The same observation has been made by Ballinger and Long.The rate constants for the alkaline dehydrochlorination of 1−3 are shown in Table 3 at 35, 25 and 10 °C, respectively.In the table k1' is the pseudo first-order rate constant and k2 the corresponding second-order rate constant.Also the average values of the latter are shown.Table 4 lists the recalculated average second-order rate constants for 4 and 5 at different temperatures.The Arrhenius parameters, A and Ea, together with the thermodynamic functions of activation, H  , S  and G  derived from the data in Tables 3 and 4 for 1−5 are collected in Table 5.
When inspecting the activation parameters of 1−3 shown in Table 5 it appears that ca 12 kJ mol −1 lower activation enthalpy explains the slowest dehydrochlorination rate of 1 (Table 3).Also the frequency factor of 1 is clearly higher than any of those for the other 1,2-chlorohydrins (2−5) studied.As to isomeric 2 and 3 both the enthalpy and entropy terms favor the clearly faster dehydrochlorination of the latter.Compounds 4 and 5 gave 2-methyl-1,2-propanediol as the predominant product whereas in the case of 1−3 the products were oxiranes.The dehydrohalogenation rates of 4 and 5 (at 10 °C, Table 4) are very close to each other as stated also earlier. 22The structural difference between 4 and 5 is reflected in their activation enthalpies and entropies (Table 5) in such away that they largely compensate each other the enthalpy being much smaller and entropy very negative for 4 as compared to those of 5.
The neutral reactions were tested for 2-chloroethanol 1 at 100 °C in 0.100 mol dm −3 NaClO4 solution.−17 The neutral reaction was also tested for 1-chloro-2-methyl-2-propanol 4 at 100 °C both in water (10 5 k = 3.2 s −1 ) and in 1−6 mol dm −3 HNO3 solution (10 5 k = 2.4−2.9 s −1 , respectively).These rates are practically negligible in comparison with the rates in NaOH solutions.Table 5. Activation parameters for the dehydrochlorination reactions of 2-chloro-1-ethanol (1), 2-chloro-1-propanol (2) and 1-chloro-2-propanol (3) Compound a 1 2 3 4 5 A, dm 3 mol −1 s −1 24.7x10 13 1.6x10 13 3.3x10 13 1.8x10 11 3.6x10  The effect of total ionic strength I on the second-order rate of 2 in alkaline dehydrochlorination was studied at a 0.010 mol dm −3 NaOH varying I from 0.010 to 0.500 mol dm −3 with NaNO 3 .A slight decrease of the second order rate constant was observed (10 3 k 2 , dm 3 mol −1 s −1 : 76.1−65.0).The results are shown in Table 6.Table 6.Pseudo first-order (k1´) and second-order rate constants (k2) for 2 at different ionic strengths in a constant NaOH concentration (0.010 mol dm −3 ) at 25 °C.Equation for the linear dependence (log k2/k2° vs I ½ ) shown under the table and the calculated k2 values in parentheses [NaNO3], mol dm −3 I, mol dm −3 10 3 k1´, s −1 10 3 k2, dm When presenting the pseudo first-order rate constants as function of the hydroxide-ion (base) concentration a straight line is obtained where the slope is equal to the second-order rate constant k2 (dm 3 mol −1 s −1 ) and i (s −1 ) is the intercept which should be = 0 if OH − is the only catalyst.Table 8 indicates that indeed i is practically zero in each case (1−3) and k2 values remain, as expected, almost equal to the values in Table 2 shown in parentheses in Table 7. products whereas in the case of 1-chloro-2-methyl-1-propanol 4 and 1-chloro-2-methyl-2propanol the predominant final product was proved to be 2-methyl-1,2-propanediol besides a small amount of 2-methyl-1-propanal.The mere water reaction was shown to be almost negligible.In constant ionic strength the base concentration had no effect on the rates whereas at different base concentrations the rate of alkaline dehydrochlorination somewhat decreases.The rate at constant base concentration and at different ionic strengths decreased also indicating that the decrease is mainly due to the change in the ionic strength also in the former case.
2-Chloro-1-propanol 2 was prepared by reducing 2-propionylchloride with LiAlH4-Et2O mixture. 23The boiling point was 51 °C at 2.5 kPa, yield 82%, n D 20 1.4362 (lit. 24b.p. 70.4 °C at 10 kPa, nD 25 26 which actually gave also some isomeric 1chloro-2-methyl-2-propanol 4. 2,2-Dimethyloxirane was treated with standard HCl at 5−10 °C and then diluted with fourfold amount of water and saturated with (NH4)2SO4.The isomer mixture was extracted with ether and separated with distillation.The boiling points of 4 and 5 are so close to each other (132−3 and 128-9 °C, respectively) that their separation was not very successful.Because especially 5 tends to decompose at atmospheric distillation, it ought to be done in reduced pressure which further decreased their boiling point difference.The purest fraction of 5 still consisted 12.8 % of 4.
Solutions.KCl (extra pure), KNO3 (pro analysis), NaCl (pro analysis), NaNO3 (cryst.extra pure), NaOH-ampoule (Titrisol), HNO3 (pro analysis) and NaClO4 .H2O (pro analysis) were commercial products from E. Merck AG.The ionic strength of the calibration solution was adjusted by 5 M NaNO3 solution (ion analyzer) from Orion Research.Distilled water was used as the solvent and the solutions were prepared at 25 °C.
Calibration solutions.These were prepared from NaCl.The chloride ion concentration in these solutions varied from 9 .10 −6 to 1 .10 −2 mol dm −3 .At first 0.010 mol dm −3 NaCl solution was prepared by weighing 0.2923 g of the salt into 500 mL of distilled water.Other solutions were prepared by diluting this solution.
Dehydrochlorination solutions.NaOH-ampoule was used to prepare a 0.500 mol dm −3 solution.This was diluted to give the solutions having NaOH concentrations from 0.01 to 0.100 mol dm −3 .The ionic strength was adjusted with solid NaNO3 to 0.100 mol dm −3 in solutions applied to study the catalytic effect of water.When studying the electrolyte effects the NaOH concentration was 0.010 mol dm −3 and the NaNO3 concentrations varied from 0 to 0.490 mol dm −3 .For potentiometric measurements the NaNO3 concentration was adjusted to 0.500 mol dm −3 .The alkaline dehydrochlorination was studied also at higher hydroxide ion concentrations (0.050−0.250 mol dm −3 ) where the NaNO3 concentration was adjusted to 0.250 mol dm −3 .The solution used to study neutral dehydrochlorination contained 0.100 mol dm −3 NaClO4xH20.

Kinetic measurements.
The progress of alkaline dehydrochlorination of 1,2-chlorohydrines 1−3 was followed by measuring potentiometrically the concentration of the released chloride ions.The direct indication method, based on the measurement of the electromotive force, E, of the cell, was applied.
where E°' [= E°(working electrode)− E(standard electrode)] is the apparent electromotive force of the cell, y is the activity coefficient, c the concentration of the sample and k is the Nernst constant: where R is the common gas constant (J K −1 mol −1 ), T is absolute temperature (K), z is the charge number of the ion and F Faraday's constant.
The potentiometer used was Compensator E 388 Metrohm Herisau.The working electrode was a chloride-selective membrane electrode (94-17A9) from Orion and the standard electrode was a saturated calomel electrode (EA 404), which was equipped with a 3.0 mol dm −3 KNO3 salt bridge solution to avoid diffusion.The chloride selective electrode was measuring the chloride ion concentration accurately until 5 .10 −5 mol dm −3 .By measuring the electromotive force E for several chloride ion concentrations (Table 8) it was possible to create calibration curves where E was plotted according to equation (1) against log c/c° (c° = 1 mol dm −3 ).These curves, the shape of which corresponded to those calibration curves reported in literature 28 and described by Orion for the ion selective membrane electrode, were used to estimate log ci for the chloride ion concentrations at the measured electromotive forces E. Alkaline dehydrochlorination reactions of 1−3. 100 mL of the reaction solution was pipetted with a 100 mL full pipet into a 100 ml Erlenmeyer bottle which was tightly closed by a cork and then thermostated at least for 30 minutes in a water bath with Lauda Thermostat.Thereafter 5.00 mm 3 of the chlorohydrine studied was added from a piston pipet with vigorous mixing and this was taken as the zero time for the reaction.However, 1-chloro-2-propanol 2 at 25 °C was thermostated in a methanol bath with a cryostate.Then 5.00 mL of the reaction mixture was pipetted with suitable intervals in test tubes which contained 1.1 mL HNO3 solution the concentration of which was chosen in such a way that the final pH of the sample was 4.5.Altogether 18 samples were taken within two half-lives and the final samples after about 10 halflives.For potentiometric measurements the sample was thermostated to 25 °C and the electromotive forces were measured as stated above.The chloride ion concentration ci were obtained from the calibration curves based on the data in Table 8.][30][31][32][33][34][35] Alkaline dehydrochlorination reactions of 4 and 5. Dehydrochlorination was followed by titrating the chlorine released in the reaction potentiometrically with a dilute AgNO3 solution 34 which was prepared by dissolving 1.6987 g of AgNO3 in 1 L of distilled water.The solution so obtained was kept in a dark bottle.The concentration of the AgNO3 solution (ca 0.010 M) was determined by titrating standard NaCl samples.The standard NaCl solution was prepared by dissolving 9.35084 g of NaCl, dried first 2 h at 128 °C and kept over night in a desiccator, in 2.000 L of distilled water.Potentiometric titration was carried out using Ag-and Hg2SO4electrodes.The sample volumes were adjusted to 16 mL by adding ethanol, acid (if the solution was still alkaline), and water.Potentiometric titrations occurred automatically and gave curves which enabled the evaluation of equivalent points at various times.When studying the alkaline dehydrochlorination of 4 and 5 the substrate concentration of the reaction solution was ca 0.020 mol dm −3 and the NaOH concentration varied from 0.02 to 0.06 mol dm −3 .The reaction solutions were prepared by mixing known alkaline and substrate solution by pouring the former into the latter.These solution were kept in glass-stoppered glass bottles which were immersed in a thermostated water bath the temperature of which was controlled within 0.1 °C.The reaction rates were determined at three temperatures: 5, 7.5 and 10 °C (Table 6).At each temperature the rates were determined in 3-5 different base concentrations.The NaOH solutions were prepared by dissolving a weighed amount of solid NaOH in boiled distilled water.The solution concentration was determined with a known kaliumbipthalate solution using phenolliftalein as an indicator.
Calculation of the rate constants.The alkaline dehydrochlorination of 1,2-chlorohydrines is a specific second-order reaction catalyzed by the hydroxide ion (Equation 1). 10,28It is first-order in relation of both the hydroxide ion and the chlorohydrin.The reaction happens in two steps where the first and fast pre-equilibrium is followed by a rate-determining step in which the chlorine is replaced by negatively charged oxygen.The reverse reaction is negligible. 29The fast step follows the S N i-mechanism described for the intramolecular substitution reaction. 28In fact the final main product in the case of compounds 4 and 5 was shown to be 2-methyl-1,2-propanediol.Since the reaction follows the second-order kinetics, the rate can be obtained from the equation 35 where [A] is the chlorohydrine concentration, [B] the hydroxide-ion (base) concentration and k2 the second-order rate constant.The ratio of the hydroxide ion (base) concentration and that of the chlorohydrine was 20.In other words we could use the Ostwald isolation method to derive the reaction rates 35 where k1´ is so-called pseudo first-order rate constant.Thus The pseudo first-order rate constants were estimated from equation 8 36 where t is the reaction time, co the initial chloride ion concentration (mol dm −3 ) at t = 0, c the chloride ion concentration at the end of the reaction and ct the chloride ion concentration at time t.In each reaction the initial chloride ion concentration was equal to zero.Equation 6gave then the second-order rate constants k 2 .
The rate constants for 1−3 were also measured at 4 to 5 different temperatures (Table 3) to determine the activation parameters (Table 5) for their dehydrochlorination reactions from the equation where A is the frequency factor (dm 3 mol −1 s −1 ), A° = k2° = 1 dm 3 mol −1 s −1 and Ea the activation energy (kJ mol −1 ) from which the values of H  , S  and G  (Table 5) were obtained in the usual manner.Also some earlier reports 11 from our department has dealt with the alkaline dehydrochlorination of 1−3.In these reports the kinetic measurements were made titrimetrically.However, the reaction conditions did not fulfill the concentration ratio of the reagents required by a pseudo first-order reaction 35 since the base concentration was at most only four-fold compared to that of the substrate.The ionic strength was not kept constant either in the reactions carried out at different temperatures.Therefore we recalculated the rate constants for these measurements 11 from equation where a is the initial concentration of 1−3 (mol dm −3 ), b that of NaOH (mol dm −3 ) and x the chloride ion concentration (mol dm −3 ) at time t (s).The recalculated second order rate constants are given together with the present results in Table 2.This treatment also proved that in fact the neutral dehydrochlorination (water reaction) was insignificant in agreement with our test where we kept 10.0 mm 3 2-chloroethanol in 100 mL of boiling (t 100 °C) 0.100 mol dm −3 NaClO4 solution.5 mL samples were withdrawn with a full pipet and cooled at 25 °C for potential measurements.The reaction was followed for an hour but did not proceed practically at all.In an M.Sc.thesis 37 the alkaline dehydrochlorination of 1-chloro- The calculations showed that the percentage proportion of x2 corresponded within an error limit the initial contribution of 5 (87.2%).According to the literature the dehydrochlorination rates of 4 and 5 are very close to each other (cf Table 5). 22e effect of base concentration.The rate constants for 1-3 (Table 1) were also determined at different NaOH concentrations (0.01-0.100 mol dm -3 ) keeping the ionic strength constant (0.100 mol dm -3 ) with NaNO3.For 1 the rate constants were also measured varying the NaOH concentration from 0.050 to 0.250 mol and keeping the ionic strength (0.250 mol dm −3 ) constant with NaNO 3 (Table 2).
The effect of ionic strength (I).The effect of total I on the second-order rate coefficients of 2 in alkaline dehydrochlorination was studied at a 0.010 mol dm −3 NaOH varying the NaNO3 concentration from 0 to 0.49 mol dm −3 .In other words the ionic strength was varied from 0.010 to 0.500 mol dm −3 .The results are listed in Table 6.
Neutral dehydrochlorination (water reaction).This was tested with 2-chloroethanol 1 at 100 °C in 0.100 mol dm −3 NaClO4 solution.10 mm 3 of 1 was placed with a plunger pipet in 100 mL of the reaction solution which was kept boiling with an electrical bath.Thirteen samples were withdrawn at intervals for an hour with a 5.

Table 4 )
were again recalculated from equation 10 since the reaction conditions applied did not obey the pseudo first-order kinetics because the NaOH concentration was at most only 2.6-fold as compared with the substrate concentration.Since 5 consisted 12.8 % of 4 the rate constants for the former were evaluated from equation 2-methyl-2-propanol 4 and 2chloro-2-methyl-1-propanol 5 was studied argentometrically.The second-order rate coefficients (