1632
D.
S.NOYCE,P. A. KING,F. B. KIRBSAND W. L. REED
follows that any modification of the phenolate hybrid which serves to stabilize electron density a t the para carbon atom should facilitate ionization. Whether by electrostatic effects alone, as in XVII, or by covalent bond formation as well, as in XIV, the benzamido carbonyl may serve as an electrophile t o promote anion solvation a t sites other than the phenolic oxygen. Such an argument is supported by the spectral data of Fig. 1, I X being more acidic than VI and both more acidic than the corresponding cresol XVIII. The fact that the ester VIa is more readily ionized than the free acid VI supports the argument used to relate the hydrolysis rates of VI and IX. Ionization constants for a variety of unhindered phenols with appropriate side chains were determined spectrophotometrically; in each case, the pK was found to lie within the range 10.0 to 10.1. Thus, i t seems unlikely that the variations shown in Fig. 1 need
[CONTRIBUTIONFROM
THE
Vol. 81
” Ph XIX
be ascribed to inductive effects resulting from sidechain modification. We, therefore, attribute the enhanced ionization of VIa to the existence of an additional locus for charge solvation, as in X I X or its dipolar analog.
DEPARTMENT O F CHEMISTRY, UNIVERSITY
OF
CALIFORNIA, BERKELEY, CALIF.]
The Kinetics of the Acid-catalyzed Isomerization of cis-Cinnamic A ~ i d l - ~ B Y DONALD s. NOYCE,
PL4ULA. KING,4 FRED B.
KIRBYAND WILMERL. REED
RECEIVED SEPTEMBER 29, 1961 cis-Cinnamic acid is smoothly isomerized to trans-cinnamic acid by moderately concentrated sulfuric acid.
I n the region
45-7570 sulfuric acid, the rate of isomerization follows Hammett’s acidity function ha with unit slope. Evidence is presented t o show that the course of the reaction follows an addition-elimination sequence, with 8-phenyl-P-hydroxypropionic
acid being formed a s an unstable intermediate.
Introduction For some time we have been interested in a variety of acid-catalyzed reactions. Most recently, detailed studies of the isomerization of cisb e n z a l a c e t ~ p h e n o n e shave ~ ~ ~been ~ ~ reported. Two different mechanisms were found to be operative depending upon the nature of the substituents in the benzene rings. One of these mechanisms involved addition of water to form the hydroxyenol, while the other mechanism involved the direct rotation aboct the carbon-carbon double bond in the oxonium salt of the substituted benzalacetophenone. The present paper presents a study of the acidcata!yzed isomerization of cis-cinnamic acid to trans-cinnamic acid. This study was undertaken to obtain additional information regarding reaction mechanisms to which the use of the acidity function IT0 might be applied. The cinnamic acid system offers several advantages for a study designed to test proton addition mechanisms’ for acid-catalyzed reactions. It is (1) This research was supported i n p a r t b y a g r a n t f r o m t h e Petroleum Research F u n d administered b y t h e American Chemical Society. Grateful acknowledgment is hereby made t o t h e donors of this fund. ( 2 ) Presented in p a r t a t t h e 8 t h Conference o n Organic Reaction RIechanisms, Princeton, ST.J., September, 1960. 131 Paper XlTI in t h e Series Carbonyl Reactions; previous paper, D.S. Noyce and M. J. Jorgenson, J . A m . Chem. Soc., 88, 2525 (1961). ( 4 ) Union Carbide a n d Carbon Fellow, 1956-1957. ( 5 ) D. S S o y c e , W. A. P r y o r a n d P. A. King, J . A m . Chem. Soc., 81, 6423 (1959). (6) D. S. h’oyce, G . I,. Woo a n d M. J. Jorgenson, ibid.. 83, 1160 (1961).
known from the thermal isomerization of methyl cinnamates that the cis-trans isomerization proceeds to completion. In considering possibly useful related compounds, @-phenyl-/3-hydroxypropionic acid is well known and has also been resolved. Further, the stereomutation reaction will be free of any complications of a-P,P-y equilibria, which would not be the case with many aliphatic compounds. We desired to obtain a body of information, which would establish with a high degree of certainty the mechanism for the isomerization of cis-cinnamic acid; and then finally to consider the acidity deor “non-Ho”) of the reaction in the pendence (Ha light of these results. In this and the immediately following papers, we present our experimental results and then finally the conclusions which we have drawn from these studies.
Experimental Preparation of Materials.-Methyl cis-cinnamate was prepared by hydrogenation of methyl phenylpropiolate using a Lindlar catalyst. Hydrolysis of methyl cis-cinnamate afforded cis-cinnamic acid, which was crystallized to constant m.p. from ligroin; m.p. 67.568.0” (lit.9 68”), neut. equiv. 147.2 (calcd. 148.15). Kinetic Procedures.-The progress of the reaction was followed by the increase in ultraviolet absorption a t an appropriate wave length, usually 300 mp, using a Beckman DU spectrophotometer. A weighed amount of cis-cinnamic (7) F. A. Long a n d M. -4.Paul, Chem. Revs., 67, 935 (1957), review much of t h e literature relative t o such mechanisms. (8) G . B. Kistiakowsky a n d W, R . S m i t h , J . A m . Chem. Soc., 67, 269 (1935). (9) C . Liebermann, B e y . , 2S, 2510 (1880).
X a y 5, 1962
ISOMERIZATION OF CZ'S-CINNAMIC ? A C I D
acid was dissolved in water and, to initiate reaction, a small aliquot of this solution mas diluted with sulfuric acid of the desired concentration. Two slightly different procedures were used a t 45". For faster runs, the optical density readings were made on a single sample maintained in a thermostated cell block. For slower runs, aliquots of a large volume of solution maintained a t temperature were withdrawn and read with the Beckman. .Zt 70", the conventional sealed ampule technique was used. Reactions were, in general, allowed to proceed to completion. The kinetic data were plotted in the form of apparent percentage of remaining cis-cinnamic acid; Le. O.D., O.D.,
- 0.D.t x - O.Do
100
The infinity absorbances were reasonably stable and, in all cases, corresponded closely to those expected for complete conversion to trans-cinnamic acid. Upon standing for long extended periods, a gradual decrease in absorbance occurred. This is due to slow acidcatalyzed decarboxylation of trans-cinnamic acid.1° Results of a typical kinetic run are given in Table I.
1633
anhydrous sodium sulfate and evaporated to yield 3.4 g. of trans-cinnamic acid-a-dl (85% yield), m.p. 133". The infrared spectrum showed that the doublet a t 10.2 p , characteristic of ordinary cinnamic acid, had almost completely disappeared. Analysis of the spectra indicated that this product is 88% deuterated in the a-position. Isomerization of cis-Cinnamic Acid in Sulfuric Acid-&.A suspension of 300 mg. of cis-cinnamic acid in 100 ml. of 9.2 M sulfuric acid-dn in D?O was maintained a t 45' for 36 hours. Work-up in the usual fashion, followed by crystallization from water and then from pentane, afforded 100 mg. The inof trans-cinnamic acid-a-d, m.p. 133.2-134.4'. frared spectrum of this material compared favorably with authentic cinnamic acid-a-dl, but the doublet a t 10.2 p was more pronounced, indicating that the material was 80% deuterated.
Results Acidity Dependence.-The isomerization of ciscinnamic acid to trans-cinnamic acid proceeds smoothly in moderately concentrated sulfuric acid. At 45' in the region 60-757* sulfuric acid, each TABLE I kinetic run is smoothly first order from 5-907, reaction, and proceeds to completion within the preciTYPICAL KIXETICRUN of the spectrophotometric method Temp. 45.00°, HzSOc62.80%, initial concn. cis-cinnamic acid sion (*2%>. 3.85 x I O - 5 ~ The isomerization is smoothly acid catalyzed. Time, Optical k X to,", sec. X 10-2 density % Rxa sec. The half-life a t 45' decreases from nearly a week in 45% sulfuric acid to only 10 minutes in 75% sulfuric 0 0.219 0 .... acid. The rate constants over this range of acidity 6.2 ,236 4.04 (6.64) are given in Table 11. Also given are the sums log k 5.95 12.0 ,248 6.89 Ho, using both the Ho values of Paul and LongI2 18.0 ,263 10.45 6.15 a t 25', and corrected Ho values, applying the tem36.0 ,302 19.71 6.09 perature corrections determined by Gelbstein, 49.2 ,326 25.42 5.96 Shcheglova and Temkin.13 60.0 ,349 29.93 5.93 It is to be noted that the slope of log k v s . Ho is i3.2 ,367 35.15 5.9% very nearly unity throughout the entire range. 84.0 ,385 39.43 5.89 + Lis-cinnamic acid is thus clearly The ~ K B Hfor 97.2 ,404 43.94 5.95 substantially more negative than - 6.5. The 114.0 ,427 49.41 5.98 ~ K B Hof+trans-cinnamic acid is -6.2, somewhat 137.4 ,454 55.82 5.95 more basic than benzoic acid.14 To be consistent 5.92 162.5 ,479 61. i6 with our rate data, cis-cinnamic acid must be less 174.0 ,490 64.37 5.93 basic than trans-cinnamic acid. This is, of course, ,640 ... ... 972.00 ( m ) Av. 5.97 +t0.05 in accord with the steric inhibition of resonance stabilization which manifests itself more strongly in a % R x = (O.D.6 - O.D.o)/(O.D.m - 0.D.a) X 100. the cation than in the free acid. A similar situation Isolation of Cinnamic Acid under Conditions of the Kinetic Experiments.--A sample of cis-cinnamic acid (200 mg.) was exists in o-methylbenzoic acid.14 Temperature Dependence.-Table I11 presents dissolved in 9 M sulfuric acid and maintained a t 45' for a period corresponding to 6 half-lives. Considerable truns- data for the rate of isomerization a t 70.0". cinnamic acid precipitated during this period. Dilution The energy of activation calculated from these with water, extraction with ether and work-up in the usual data is 24 =t 1 kcal. In order to calculate AS*, i t is fashion afforded a 90% yield of trans-cinnamic acid, m.p. necessary to extrapolate to some standard state for 132.4-133.5'. trans-Cinnamic Acid-a-dl.-Benzalmalonic acid, (dec. proton activity, either 1 M sulfuric acid or ha = 1. 195-200') mas prepared by the method of Claisen and Such an extrapolation is fairly severe in this case. Crismer" and converted to the sodium salt by titration to PH 8.3. The solution was evaporated to dryness under re- The extrapolated rate constant a t ho = 1 is 1.20 X set.-' a t 45', giving AF+ = 31.5 kcal. and AS* duced pressure. = -26 =t 3 e.u. Afzal. Calcd. for CloHa04ha?: C, 50.86; H, 2.56. Found: C, 50.56; H, 2.89. It should be noted that the very negative AS* Conversion to di-deuteriobenzalmalonic acid was ef- obtained in this case contrasts with the entropies fected by solution of 5.13 g. of the sodium salt in 10 ml. of of activation associated with most other reactions DzO and precipitation w-ith 10 ml. of D?S04. The precipitate was collected by filtration and placed in a 250-1111. which correlate with the acidity function. Long, ~ Taft and his coround-bottomed flask, 100 ml. of previously dried pyridine Pritchard and S t a f f o ~ d ' and added, and the flask fitted with a condenser and drying tube workers16have applied the entropy criterion to Ha
+
and heated under reflux for 3 hours. The colorless pyridine solution was cooled and poured onto a mixture of 100 ml. of concentrated HCI and 200 g. of cracked ice. The organic material was extracted with four 50-ml. portions of ether which were combined and washed with 5% KHCOI solution. The ethereal solution was separated and dried over (10) Unpublished experiments of hlr. F. B. Kirby. (11) L . Claisen and L . Crismer, A n n . , 218, 129 (1883)
(12) hl. A. Paul a n d F. A. Long, Chem. Revs., 57, 1 (1957). (13) A. I. Gelbstein, G. G. Shcheglova and hI. I. Temkin, Zhur. neom. K h i m . , 1, 506 (1956). (14) R. Stewart a n d K. Yates, J . A m . Chem. Soc., 8 2 , 4059 (1960). (15) F. A. Long, J. G. Pritchard a n d F. E. Stafford, i b i d . , 79,2362 (1957). (16) R. W.T a f t , Jr., ibid.,74, 5374 (1952); R. W.T a f t , Jr., E. L. Purlee, P.Riesz a n d G A. DeFazio, ibid.. 77, 1584 (1955).
D. S . NOYCE,P. A.KING,F. B.KIRBYAND Lir. L. REED
1634
Yol. s4
TABLE I1 AQUEOUSSULFURIC ACIDAT 45.0'
R A T E OB ISOhIERIZAT~OXOP C i S - C I N s A N I C ACID I N
HYSOI,%
Molar
€I,"
AHsb
46.84 51.63 62.80 67.92 68.88 69.88 75.97
6.52 7.43 9.79 10.99
-3.04 -3.56 -4.79 -5.38 -5.50 -5.64 -6.43
0 0.01 .07 .10 .09 .09 .10
11.47
kl
Hm -3.04 -3.55 -4.72 -5.28 -541
-5.55 -6.33
x
101,
sec. -1
log k
0.133 0.448 5.97 22.7 30.3 42.3 254
-5.88 -5.35 -4.22 -3.64 -3.52 -3.37 -2.60
Cor.
-log
k
Uncor.
+ Ho-Cor.
-8.92 -8.91 -9.01 -9.02 -9.02 -9.01 -9.03
-8.92 -8.90 -8.94 -8.92 -8.93 -8.92 -8.93 d
5
Ref. 12.
* From data in ref. 13.
Av. = -8.99 i 0.04.
Av. = -8.92 f 0.01.
TABLE 111 KATE OF ISOMERIZATION OF &-CIXNAMIC ACID IN AQUEOUSSULFURIC ACIDAT 70.0' H;SO4,
a
7,
He
AH
Cor. Ha
-3.04 -3.38 -3.91 -4.56
0 0 0.03 0.12
-3.04 -3.38 -3.89 -4.44
Molar
6.52 46.84 50.0 7.11 54.82 8.07 60.69 9.32 Av. = -7.72 f 0.05,
k X 106, EeC. - 1
log k
2.20 4.13 12.9 61.5
-4.66 -4.38 -3.89 -3.21
-log Uncor.
k -I-Ho--
Cor.
-7.70 -7.76 -7.80 -7.77
-7.70 -7.76 -7.77 -7.65 0
reactions. We have also considered this question previously. l 7 Evidence for the Formation of an Unstable Intermediate.-While the rate of formation of trans-cinnamic acid shows good first-order kinetics throughout the course of an individual kinetic run a t high acidities, this is not strictly true a t lower acidities. In the slower runs, there is indication of a small induction period, as evidenced by a lag in the increase in optical absorption. In a typical experiment (Table IV), the rate constant for the formation of trans-cinnamic acid shows a gradually rising value for the first 1520% of the reaction. A plot of these data likewise reveals curvature a t low conversions.
kz ki cis +intermediate -+ trans
(1)
Though it is possible in principle to determine both k, and kf from the data of Table IV, the limitations of the spectrophotometric method in this case preclude any accurate determination of kz. However, kl is easily obtained. I t is much more instructive to compare the rate of dehydrationls of 0-phenyl-0-hydroxypropionic acid (111) with the rate of isomerization. The rate of dehydration is uniformly greater and also of greater than unit slope ZJS. Ho. This means that the ratio changes with changing acidity, becoming smallest a t the lower acidities. A comparison of the rates is given in Table V.
TABLEIV
RATE O F 45.00' Time,
IN
see. X 10-8
0 5.88 11.58 20.88 38.88 91.5 132.5 177.6 266.8 ( a)
TABLE V RATE O F DEHYDRATION OF
[email protected] M H&O4 CONTAINING 2% . . ETHANOL~ HYDROXYPROPIONIC ACID WITH KATE OF ISOMERIZATION OF Optical density Reaction,b k X l$," k X loa,$ cis-CINNNAMIC ACID AT 45.00'
ISOMERIZATION, SHOWING INDUCTION PERIOD AT
300 mr
%
0.126 .128 .133 .142 .172 .230 .263 .300 .340 .421
... 0.66 2.4 5.4 15.6 35.2 46.4 59.0 72.5
COMPARISON O F
sec. -1
XC.
1.13 2.04 2.67 4.36 4.75 4.71 5.02 4.84
... (6.32) 5.30 5.09 5 33 5 03
Inductn. oeriod
38.43 51.87 58.05 62.80 fi9.00 75 97
-2.30 -3.58 -4.25 -4.79 -5.53 -6.43
3.4 X 1.07 X 7.31 X 3.2 X 3.0 X 5.3 X
101 2.4 X 10-4 4.69 X 10-4 1.61 X 10-8 5.97 X 10-9 3.25 X 10-1 2.54 X
10-8 10-5 10-6 10-4
10-8
a From ref. 12. b Taken from ref. 18. from the data in Table 11.
14 + 1 23 f 2 45 f 3 54 f 4 92 It 5 -200
.. Yes Quest.
NO NO
No
Extrapolated
Av. 5.19 f.0.12 These comparisons provide very strong evidence Ethanol added to facilitate solubility. &-Cinnamic acid that ,&phenyl-P-hydroxypropionic acid is an interwas dissolved in 2 ml. of ethanol and diluted t o 100 ml. with mediate in the isomerization of cis-cinnamic acid. aqueous sulfuric acid. The HOvalue of this solution was not determined. (O.D.t-0.D.o)/(O.D.,-OO.D.~)X 100. c k Therefore, an addition-elimination mechanism is = ln[(O.D., - O.D.,)/(O.D., - O.D.~)]t-~,startingatzero operating (eq. 2), and the measured rate of isomtime. $ k = ln[(O.D., - O.D.t)/(O.D., O.D.o)]t-', zero C&~)C_.~
