A. K. OLSONAND P. V. SOULE
2468
[CONTRIBUTION FROM THE CHEMISTRY DEPARTMENT,
Vol. 73
UNIVERSITY OF CALIFORNIA]
The Hydrolysis of p-Butyrolactone BY A. It. OLSONAND P.V. Y O U L E ~ ~ Previous work has shown that at least two mechanisms are involved in the hydrolysis of 8-butyrolactone producing the two optical forms of ,&hydroxybutyric acid. The work is here extended so as to permit a determination of two catalytic coefficients for each of the following bases: carbonate ions, phosphate ions and tetraborate ions. The separate catalytic coefficients, but not their sums,fit Brgnsted diagrams. It is suggested that the occurrence of simultaneous multiple reactions may contribute to the curvature that has been observed in the Br6nsted diagrams of other reactions.
@-Butpalactone reacts with water and its ions to form hydroxybutyric acid. The rate of hydrolysis can be expressed by the equation
These results were confirmeda by hydrolyzing the lactone in heavy oxygen water, decomposing the product and analyzing the water formed by the decomposition for isotopic oxygen content by the method of Cohn and Urey.4 According to Brqinsted the specific velocity constant (ka, kb) for an acidic or basic catalyst is related to the acid or base constant ( K a , K b ) of the catalyst by the equation
If the solution contains ions of weak acids or bases, a fourth term (which may be complex) must be added to (1). By employing optically active lactone, Olson and Miller2 were able to show that the reaction with molecular water produced only ks.h = G G b that acid that rotates the plane of polarized light in the same direction as did the original lactone where G and x are empirical constants. Following a reaction that involves an inversion of the alcoholic Brqinsted and Pedersenjb many investigators have carbon and that the reactions involving either obtained data in agreement with this equation. H + or OH- produced only the enantiomorph. However, Olson and Miller found that that portion They further showed that a base like COa- pro- of the rate corresponding to catalysis by carbonate duced a mixture of the two optical forms. In ion was complex. We address ourselves therefor postulating mechanisms for these reactions, they to the questions: (1) Are the reactions involving found it necessary to assume that a molecule of other basic catalysts similarly complex? (2) water is involved, in addition to each reactant For any one catalyst does the Brginsted relationship specifically indicated in the terms in the right apply to the over-all reaction or to its separate hand side of equation (l),resulting in reactions parts? The study of the hydrolysis of p-butyrolactone now familiarly known as “push-pull” reactions. For carbonate ions the results can be represented as might be complicated by (1)regeneration of lactone from hydroxybutyric acid, (2) decomposition of H CHI hydroxybutyric acid into water and crotonic acid \/ or into propylene and carbon dioxide, (3) ester H--0i!-H+ 0-C formation with organic bases. Within the limits I - I-- i I O=C-C--H --+ HO-C + 1*20 of error the titration results in Table I show the 1 -t COI’ t i hydrolysis to be irreversible. The data of PressO=C--c H-0 H man and Lucass indicate that the second reaction I,‘ I ,#‘i OH should be unimportant near room temperature. H .. However, some of our results, particularly those cosat 35’) are marred by interfering reactions. Ester and formation with organic anion bases is important. It will be discussed in a subsequent paragraph. The preparation of optically active @-butyrolactone and the study of its hydrolysis by polarimetric experiments, followed exactly the procedure outlined by Olson and Miller. In those experiments where the reaction was followed by titrations, carbon dioxide was excluded from the reaction and titration vessels by sweeping them out with a stream of purified nitrogen. The reaction with water was unimolecular and the reaction with hydroxide ion bimolecular over its whole course. For the rapid hydroxide ion reaction, 200 cc. of COz-free water was placed in a stoppered H bottle containing a handful of glass beads. An (1) The work that is reported in this article was performed over ten ampoule containing a weighed amount of lactone years ago. Its publication has been delayed partly by the war and was added and then an amount of barium hypartly by the hope that more complete and more precise data would I
be obtained. Since this hope bas not been realized, the material is now being released. (la) Commonwealth Fund Fellow. Present address: Alkrington, Middleton, Manchester, England. (2) A. R.Olson and R . J \Idler, ’I if15 J U I J X N A I , , 66, 2657 (1935).
(3) A. R. Olson and L. J. Hyde, ibid., 6S, 2459 (1841). (4) hlildred Cohn and H. C. Urey, ibid., 60,678 (1938). ( 5 ) J N Br#nsted and K Pedersen, Z physik Chcm , 108, 185 (1924) ( 6 ) D Pressman and H J Lucds, ‘J
HIS J O U K N A I
, 61, 2271 (1039)
June, 1951
' h E IJYDROLYSIS OF
droxide equivalent to this amount of lactone. The reaction could be started at a given instant by shaking the bottle. At appropriate intervals 50 cc. of the mixture was added to a known quantity of 0.1 N acid to stop the reaction. Titration with barium hydroxide solution gave the amount of lactone decomposed. The results are summarized in Table I. TABLE I HYDROLYSIS BY WATER (a) Typical titration: initial concentration of lactone 0.0185 M , temp., 20' Minutes
M1. 0.104 N Ba(0H)z
k X 104
400 1271
1.52 3.81
4.69 4.38
(b) Summary of rates and activation energies k X 10' E 20 4.6 25 8.2 20.8 30 14.4 19.8 35 24.8 19.7
Temp., 'C.
TABLE I1 HYDROLYSIS BY HYDROXIDE ION ( a ) Typical titration: initial concentration of lactone and of OH- 6.6 X M , temp., 20" Minutes
M1. 0.1 N HCI M1. 0.104 N Ba(OH)z
5 10 15
k
0.17 .18 .15
1.70 I .20 0.90
35.3 34.2 34.8
P-BUTYROLACTONE TABLE IIIb
k
E
20 25 30 35
35.0 51.2 72.5 90
13.3 12.6 7.7
OM)
(Wcor.
P
1 2 3 4 51 52
-6.45 -3.44 -4.35 -2.50 - .81 +1.07
-7.23 -5.95 -7.50 -4.25 -1.39
0.34 .365 .33 .405 .47 .54
TABLE IIIa Buffer
1 2 3 4 51 52
M M COa- HCOI'
2.18 1.46 1.09 0.72 0.20 0.20
k
pH
C
0.60 10.0 0.207 0.40 10.0 ,0485 .0690 0.30 10.0 2.00 9 . 4 .0876 2.48 8 . 5 .0412 2.48 8.5 .125
X 10'
749 529 426 259 80 70
OL
-0.40 - .05 - .09 - .065
-
+
.01
.04
R
1.12 1.72 1.72 1.72 1.72 1.72
+1.84
104
koa
8.5 8.5 8.5 8.5 8.5 8.5
x
(OH-) 104
49 49 49 12 1.5 1.5
of ( k f )to Mloo%gives p , the fraction (+) acid in the product. If k in Table IIIa is multiplied by p , we obtain the total (+) rate. Since water contributes to the (+) rate and OH- to the (-) rate, we subtract the values in the last two columns in Table IIIb to obtain the (+) and (-) contributions of the anions. By varying the relative proportions of bicarbonate and carbonate ions, the catalytic contributions for each should be obtained. Actually no contribution from bicarbonate ion could be detected, carbonic acid being a stronger acid and bicarbonate ion a weaker base than the usually quoted ionization constant indicates.' Consequently to the carbonate ion present were ascribed plus and minus rate contributions, (k+co,)(COS-) and (k-eo,) (COa'), from which the plus and minus specific catalytic coefficients can be calculated. In Table IIIc we have collected these calculations: TABLE IIIc All rate constants have been multiplied by lo4 Buffer
In Table IIIa we have summarized the data for a group of experiments in carbonate buffer solutions a t 25'. In this table, the quantities in the
kAtO
x
Buffer
(b) Summary of rates and activation energies Temp., O C .
2469
1 2 3 4 5*".
k+
k-CO8
k+Coi
(COa-)
(COS-)
255 193 141 105 38
445 287 236 142 35.5
246 184 132 97 29.5 Av.
k-cor 208 197 216 197 178 199
k+coa
kCOa
112 126 121 135 148 128
320 323 337 332 326 328
The total catalytic coefficient for CO3- is surprisingly constant. Throughout this work we have neglected salt effects as well as any catalysis or ester formation by the hydroxybutyrate ion. In Table IV we have collected the experimental data, and in Table V, the calculations, relating to experiments with carbonate buffers at 30 and 3 5 O , as well as experiments using phosphate and borax buffers at all three temperatures. The calculations follow the pattern presented in detail for the car-
TABLE IV second, third and fourth columns are properties of the buffers, C is the initial molal concentration EXPERIMENTAL RESULTSAT OTHERTEMPERATURES AND IN of the lactone, K is the total first order rate constant, OTHERINORGANIC BUFFERS a! is the final rotation of the solution, and R is the Temp., Lactone, k Plus OC. Buffer 9H M X 104 form, % factor by which the rotation should be multiplied 30 0.545 M COS' 10 0.0975 360 25 due to the fact that the lactone was prepared from 3o .15 M HC0,10 an optically impure acid. The change in rotation .545MCO*' 10 .066 416 25 as the lactone hydrolyzes is not large. Further35 .15 MHCOs10 .OS27 595 19 more, emulsion formation limits the concentration 25 .50 MHPO4' 7 .145 26.4 88.5 of lactone. These factors operate to decrease the .33 M HzPO4precision of k as well as a. 30 0 . 5 0 M H P O ~ 7 .117 46.2 85 Table IIIb is a continuation of Table IIIa. 7 35 0.33 M HnPO4.124 80.8 90 ( M ) is the molal rotation calculated from a, 25 .10 M NaeB40, 9.3-8.4 .OS05 29.6 65 is obtained from ( M ) by applying R. The 30 .10 MNad3407 9.3-8.4 .118 60.0 47 molal rotation of optically pure &hydroxybutyric acid, MUM%, in these buffers is 22.4'. The ratio (7) A. R. Olson and P. V. Youle, THISJOURNAL, 64, 1027 (1940). ~
(wCor.
1'
AHLh
\
S P ~ L I IKI%Ci r - C o ~ . s r a \ i sX I O i 1 etnp , 'C.
X'i,?
55
cos-
IIPO,' HPO,' HPOI' BdO7' B107-
:VJ
35 25 30
. o .
12% 1 iil I \)2
,532 880 36
cos-
-4.
, / '
/ I
'lcr 4' 6 78
16 S2
140
1
