Kinetics of Bisulfite Addition to α,β-Unsaturated Compounds1,2

The kinetics of the addition of bisulfite in aqueous solution to acrylonitrile, methacrylonitrile, methyl acrylate and methyl methacrylate were studie...
0 downloads 0 Views 424KB Size
July 20, 1952

KINETICS OF BISULFITE ADDITION TO a,UNSATURATED COMPOUNDS

[CONTRIBUTION FROM THE RUBBER RESEARCH LABORATORY O F THE UNIVERSITY

3323

OF A K R O N ]

Kinetics of Bisulfite Addition to qo-unsaturated Compounds1~2 BY MAURICE MORTON AND HAROLD LANDFIELD RECEIVED JULY 25, 1951 The kinetics of the addition of bisulfite in aqueous solution to acrylonitrile, methacrylonitrile, methyl acrylate and methyl methacrylate were studied. The reactions were found to be second order over a range of concentrations from 0.01 to 0.10 M , a t PH values ranging from 4 to 8. Increase in PH resulted in a marked increase in reaction rates, in a manner which indicated that the sulfite ion concentration coritrolled the rate of the reaction. On this basis, the following rate constants were obtained for three of the compounds investigated: methyl acrylate, k l = 2.0 X lO*,exp(-12000/RT) l./mole/sec.; methyl methacrylate, Ki = 1.1(h0.3) X 106 exp( -12000/RT)l./mole/sec.; methacrylonitrile, kl = 4.0 (*1.3) X lo9 exp( 18000/RT)l./mole/sec. An anomalous trend was obtained in the activation energy of the fourth compound, acrylonitrile.

-

Introduction It is well known that bisulfite salts add readily to aldehydes and ketones, but only with great difficulty to ordinary ethylenic linkages. However, this addition reaction apparently occurs quite readily a t the ethylenic bond in a,@-unsaturated compounds such as aldehydes, ketones, acids and esters. The present Fork arises from a polymerization study of water-soluble monomers, such as methyl methacrylate and acrylonitrile, in aqueous solution. In polymerizations of this type the bisulfite ion apparently has a catalytic effect in the presence of oxidizing agents. Since monomers of this type are capable of adding bisulfite, the rate of polymerization is sensitive to conditions which may affect the extent of such a side reaction. This study was undertaken to throw some light on the mechanism of this addition reaction, which has received only scant mention in the literature. No actual rate studies have been reported for the addition of bisulfite to a,8-unsaturated systems. Stewart and Donnally3 observed that the rate of decomposition of the bisulfite-aldehyde product increased with lower PH (down to 1.8) and then decreased again. On the basis of the two ionization constants of sulfurous acid, they concluded that in the pH range of 3 to 13 this reaction involves the sulfite ion rather than the bisulfite ion. Bacon' observed an exothermic reaction between acrylonitrile and sodium sulfite, accompanied by a rise in pH, according to the equation CHt=CHCN

+ NazSOI + HzO + CH&OaCH&N

+ NaOII

The kinetic study of Gubarova' on the bisulfite addition to acetone has shown that the rate decreases with a decrease in pH. The activation energy for this reaction was found to be 6 to 8 kcal. In the present work some preliminary experiments showed that sodium sulfite or bisulfite added very rapidly and quantitatively to such compounds as acrylonitrile or methyl acrylate in aqueous solution, even a t concentrations of 0.01 M. Furthermore the rate of the reaction was very sensitive to (1) Presented a t the 119th Meeting of the American Chemical Society in Cleveland, Ohio, April, 1951. (2) This investigation was carried out under the sponsorship of the Office of Rubber Reserve, Reconstruction Finance Corporation, in connection with the U S. Government Synthetic Rubber Program. (3) T. D. Stewart and L. H. Donnally, THISJOURNAL, 64, 2333, 3559 (1932). (4) R. G. R. Bacon, T r o w s . Faraday Soc., 42, 140 (1946) ( 5 ) M. A. Gubarova, J. Gen. Chcm. (U.S.S.R.),16, 238 (1948).

pH and unaffected by the presence of a free radical inhibitor such as hydroquinone. Hence an ionic mechanism is indicated. Since free hydroxyl ions are released by this reaction, an adequate hydrogen ion buffer system is required for any kinetic study. The four compounds studied were acrylonitrile, methacrylonitrile, methyl acrylate and methyl methacrylate. The study comprised an investigation of the effect of pH and concentration on the order of the reaction, as well as an evaluation of the activation energy. Experimental

A.

Materials Used.-The following materials were used : Acrylonitrile, technical grade (American Cyanamid Company). This was freed from inhibitor by washing with a 170 sodium hydroxide solution and then vacuum distilled at 30" under nitrogen. The PH of a 1% aqueous solution of acrylonitrile was 6.5 after washing, as compared with a PH of 9 originally. Methacrylonitrile was received from the Shell Development Company. This methacrylonitrile was inhibited with 0.05% p-t-butylcatechol. To remove the inhibitor, the material was washed first with a sodium hydroxide-sodium chloride solution, then washed with a saturated sodium chloride-water solution, dried and vacuum distilled at 70" under nitrogen. Methyl acrylate was received from Rohm and Haas and contained 0.25% hydroquinone. The sample was purified in the same manner as the methacrylonitrile. Methyl methacrylate was received from Rohm and Haas and contained 0.006% hydroquinone. The same method was used for purifying as in the case of the methacrylonitrile with the exception that before vacuum distilling at 85" the methyl methacrylate was left overnight in calcium sulfate and then filtered. The water was boiled about 15 minutes to remove most of the dissolved oxygen and then cooled under a stream of nitrogen (Linde high purity). The water was stored under nitrogen. Sodium sulfite used was Merck Analytical reagent grade (anhydrous). Iodine Solution.-A 0.01 M iodine solution was used containing 0.02 M potassium iodide and consisting of 4oY0 by volume of ethanol. The iodine was Merck reagent grade and the potassium iodide was C.P. Baker analyzed. Buffer Systems.-The PH of the system was adjusted by means of the following two buffer systems, which were capable of maintaining the required pH values (=!=0.1 unit) during the course of the reaction, in the presence of the reactants at the concentrations used. fiH

6 5 4 bH

8 7 6

Acetate buffer acetic acid, M

Sodium acetate, M

n ,025

0.470

.o33 ,073

,

ntx

.onti

Phosphate buffer NarHPO', M

HiPO4, .\I

0.113

0.0092

.113 .I13

.os0 ,312

B.

Procedure. --’lhe rcactioiis were carrietl out i i i WUvoluinetric flasks under a11 atiiiosphere of Liiide highIurity iiitrogen, the flasks being iininerscd i n a wa.tcr-bath (I iid t herniostated at the tiesircil tern per at u rc without any agitatioii. The following general method of charging the flasks was used: ( a ) The unsaturated coinpound was dissolved in ivster. ( b ) , Buffer system viviis then added. (c 1 Sodium sulfite solution was added last and flasks were made up to correct volume with twter. The reaction time was tneaswed upon the addition of the sodium sulfite solution. At various intervals, samples of 25 t o 50 ml. were removed arid the bisulfite analysis was carried out by iodimetric titratioii using the 0.01 llil iodine solution and starch indicator. The titration was generally conducted by running the bisulfite solution into a known quantity of iodine solution ;IS rapidly as possible, to avoid air oxidation of the bisulfite. For a range of PH values from 6 to 8, a phosphate buffer was used as the acetic-acetate systetn did not work in this ratige. This system was only used t o studv the addition Icaciioiis between 0.05 dl triethyl rtiethacrylatc a r i d sulfite.

(0.01 M REACTAN 1 S )

nil.

of pH and Concentration on Reaction

Effect Order.. rractioii acrylate tlic rate

111 Fig. I , is showti the r f f r c t oi pH 011 rate hrtweeii sodiurii sulfite anti iitetlivl (0.01 11T reactants). I t can lie see11 that increases niarketlly with iiicrwsc: i n p H .

-------

I

I

pH 4

o PH 5 e pH6

w

a

2

6

4

8

/

d .

(pY 4:

30

60 TIME IN HOURS.

90

Fig :i. ’ F y p i d biicolecular plot bisulfite-iiiethyl acrylate ieactioii d t 23”.

range studied. l‘hc bitiiolecular rate coristant for the tiiethyl acrylate reaction is approximately twice that found for acrylonitrile a t a pH of 5 . Changing the initial concentration of the reactants from 0.Oi to 0.10 does not affect the order of the reaction as is shown by Figs. 4 and 5 , both for equal and unequal concentrations. Thus the bisulfite-acrylonitrile reaction (at 23’ in an acetate buffer at pH 3 ) gives an apparent value of 1.30 X IO-‘ liter! rnoldsec. for the rate constant with cquimolecular concentrations and an apparent value of 1.20 X liter/mole/sec. for unequal concentr,rtions of reactants.

N

‘0

TIME I N

l usitig acetate buffer. These rates cai: be plotted o i i ‘i liiiiolecular scalc, :i\ showri i i i Figs. 2 and .I. (0.01M R E A C T A N T S )

/

ipH 4 )

~

I

60

30 T I M E IN

2.

dl

,?’

i-

big.

TIME IN HOURS.

Sccond-order plot bisulfite-acryloiiitrile rcaction 2,jo PI1 5 jcqrinioldr concciitratioiis 1.

I?ig i

90

HOURS.

Typical biiiiolecular plot bisulfite-acryloiiitt-ilc reaction at 23”.

The reaction Seems t(J be second order over a large extent of reaction ant1 o\-er t h e whole PH

2

4

6

TIME I N HOURS.

Fig. .5. ---Second-order piot bisulfite-acrylonitrile reaction a t 25”---pH 5 (unequal concentratiotisj.

(0.05~ REACTANTS) Since the apparent rate constants show excellent A p H 6 (ACETATE) agreement over varied concentrations a t the same pH, it would appear that the p H affects the rate constant by changing the concentration of one of the ions involved. The rate constants appear to be inversely proportional to the hydrogen ion concen, tration, which indicates that the rate-governing step in this type of reaction seems to be the addition of sulfite ion. The two ionization constants of sulfurous acid are most reliably reporteds to be as follows: first second ionizaionization constant, 1.i2 X tion constant, 6.24 x The first ionization constant would not be inTIME 1N HOURS. volved in the pH range studied, but the second ioni- Fig. 6.--Effect of PH on rate of sulfite-methyl methacrylate zation constant must be considered. In order to addition a t 25". confirm the effect of this ionization constant, the pH range was extended up to a value of 8 for methyl culated rate coiis~dntsfor the inethyl liiethacry~atc methacrylate, since the reaction rate was slow sulfite reaction a t E o ,based on the second ionizaenough to be studied in that case. For the fiH tion constant of sulfuous acid. The apparent range of G to 8 the phosphate buffer system was rate constants are included for comparison. used, since the acetate buffer system could not It is obvious that the values thus obtained for maintain a constant pH in this region. The rate the rate constants are in sufficiently good agreeof the reaction between sulfite and methyl meth- ment to indicate that the addition of the sulfite ion acrylate (0.05X reactants) is shown in Fig. 6 as a is indeed the rate-governing step in this reaction. bimolecular plot, including the acetate buffer sys- On this basis, rate constants were calculated for the tem a t pH 6 for comparison. Again a second-order reactions of all the other compounds studied, and reaction rate is indicated although only up to the are shown in Table 11. first 40% of the reaction interval, after which the In the case of acrylonitrile and methyl acrylate, rate falls off. I t should be mentioned here that the values for the rate constant show reasonably although both pH 6 buffers were adequate in main- good agreement at different values of concentration taining a constant pH throughout the reaction, this and pH. However, in the case of the methyl niethwas not the case a t higher pH values. Thus a t a acrylate, a concentration effect is noticeable. In a pH of 7 there was no change during the first 40% of later sectiop, it will be seen that there is also a coiithe reaction but the pH rose to 7.5 after 550jo. The centration effect observable for methacrylonitrile. pH 8 buffer system was not very effective since the It should be noted that these two compounds show pH slowly increased to a value of 9.9 after 3oyOof a much lower rate constant for this reaction, hencc the reaction. However, on the basis of the second the rates had to be studied a t higher values of conionization constant of sulfurous acid, the effect of centration and pH. pH on the rate is quite small above a ;PH of 7, hence TABLE I1 lack of buffering power would not be serious. At CALCULATED RATE CONSTANTS FOR ADDITION OF SLLFITE any rate, this cannot account for the decrease in ION AT 25" ISACETATE BUFFER rates noted in the later stages of the reaction. Concn. of Using the linear slopes obtained during the first reacti: ants (I./mole./scc.) part of the reaction, it is possible to calculate a Compouiid $11 (51) K t x 10' x 10% bimolecular rate constant for the reaction between Acrylonitrile 4 0.01 0.8 22 inethyl methacrylate and sulfite ion. Since the 5 .01 .9 16 ionic strength of the solutions used is fairly high, 5 .02 .9 13 it is important to consider the secondary salt ef5 .05 .9 15 fect. Fortunately, the work of Tartar and Gar6 .01 1.3 18 retson6 includes data on the variation of the ionizaMethyl acrylate 4 0.01 0.8 21 tion constant with ionic strength, from which it is 5 .01 .o 33 possible to extrapolate to the values of ionic 5 .02 .9 33 strength encountered here. Table I shows the cal-

TABLE I CALCULATED RATE CONSTAXTS FOR ADDITION OP SULFITE Methacrylotiitrilc ION TO METHYLMETHACRYLATE IN PHOSPHATE BUFFER Methyl methacrylate fiH

li (acctotc)

6 7 8

Initial concn. of r e a c t a n t s , 0 OLW Kt X 10'

3 0 2.5 2.5 3.0

R a t e constant (I /mole/sec. X 108) Apparent Calcd.

0 28 0.23

o.si

1.21 1.15 1.13

1.03

1.06

( 6 ) H. V. T a r t a r and H. H. Garrctson, THISJ O U R R A L , 63, 808 (1941).

5 G G G

G

.05 .01

0.10 ,05 .10

.9 1.3 3.0 3.0 3.0

31 29 0 . os5 .13 .23

Effect of Temperature.-In Tables I11 and IV are shown values for the activation energy obtained from the temperature coefficient of the rate constant. For acrylonitrile and methylacrylate, the average value of the rate constant obtained a t any given temperature was used, since no concentration effect was present. However, for nieth-

h l A U R I C B R~OR'I'ONAXD HAROLD IANDFIELD

3326

acrylonitrile and methyl methacrylate, the rate constants a t different temperatures were compared at similar concentrations only, as shown in Table ri-. TABLE 111 ENERGY FOR A n D I r I o x O F S U L F I T E ION TO ACRYLONITRILE AND METHYL ACRYLATE(pH 5 ) ki - ki Activation energy

A4CTIVAl'ION

1'z

-

298 308 318

Ti('K.)

(!./rnole/sec. X ICJ')

- 288 - 298 - 308

Acrylonitrile 11 - 4.9 32 - 1 t 49 - 32

(kcal./mole:

18.0 15.3 8.2

titi - 33 Il(i - 66 TABLE

ACTIVATION "ENERGY FOR

(.'ompoiin,!

Tz

- Ti

Concn. (OK

-

308 298 308 - 298 318 - 308 318 - 308

-

308 298 318 - 308 318 - 308 328 - 318 328 - 318 328 - 318

('W

Iv

METHACRYLOYITRILE (PH 8 ) h - kl ( I /mole/sec )

Methyl Methacrylate 0.05 0.26 - 0 12 . 10 43 - .22 ,05 51 - .26 .10 79 - .45 Methacry lonitrile 0 087 - 0 0.L; .05 11 .04(i -10 3 1 - .OR7 -02 L7 - .20 .05 :'Y - .I4 .10 ( $ 0 - . 23

0 . 10

12.7 11.1

A n n I T I o N OF S L Z F I T E I O U TO

h f E T H Y L METHACRYLATE A h D

Activation energy (kcdl mole)

13 4 12 8 1.1 3 11.1 1'; 21 I$ 17 1 ;t 23

related to the tendency of this compound to show hydrogen bonding, which would lead to association a t lower temperatures thus reducing the activity. The energy values for the other three compounds are fairly consistent, the methacrylonitrile showing some spread in values. On the basis that the sulfite ion concentration is the rate-governing factor, the following rate espressions may be written for the three compounds listed, omitting the acrylonitrile because of the dub i o u s valucs for the activation energy R a t e constant (l./rnole,/sec.)

Methyl acrylate 2.0 X lo8 exp(-l2000/KT) Methyl methacrylate 1.1(i - 0 . 3 ) X lo6 exp( - 18000/R7') blethacrylonitrile 4 . 0 ( 5 1 . 3 ) X lo9 exp( - 18000/HT)

hf ethyl acrylate

308 - 298 318 - 308

\Fol. 74

2 2

I t is apparent that the two acrylates show ;t much lower activation energy than the nitrile, while the latter has the lowest steric factor. Substitution of the hydrogen by a methyl group increases the steric factor by about 100, while substitution of the nitrile group by a carbomethoxy group increases the steric factor by about lo3, The higher activation energy of the nitrile might seem surprising a t first glance since a nitrile group would be expected to exert a greater inductive effect on the electrons than the less electronegative carbornethoxy group. However, in these a ,P-unsaturated systems, the resonance effect can be expected to predominate, leading to the polarized forms iii(rile

,c

(.