June 20, 1956
XCETOLYSIS OF CYCLOALKYL *ARYL SULFONATES
273.7
Several attempts to isolate aletheine as the oxalate, following the sodium-liquid ammonia reduction of bis- [N(N-carbobenzoxy-~-alanyl)-2-aminoethyl] disulfide were unsuccessful.20 Kinetic Runs.-The kinetic runs were carried out essentially as described previously,4 following the rate of the disappearance of the thioester group by measuring the ultraviolet absorption a t 233 mp with a Beckman DU spectrophotometer. It was shoxn t h a t the thioesters obeyed Beer's law strictly; the E X 10-3 values a t 233 mp for the three thioesters were as follows: 111, 4.51 ic 0.027; I V , ,519 0.082; V, 4.75 & 0.032. Isolation of Products. A. From N,S-Diacetylaletheine (V).-One gram of N,S-diacetylaletheine was hydrolyzed in 1 1. of 0.1 M sodium hydroxide a t room temperature for 5 days, after which air was bubbled through the solution
until no test for the thiol group was given using iodine-potassium iodide. The pH was brought to 3, with 1 N h;drochloric acid, the solution was reduced to dryness at 40 , with a water-pump, and the residue was extracted with 200 cc. of warm anhydrous methanol. The mixture was filtered, and the filtrate was evaporated to dryness with a water-pump. The white crystalline residue weighed 0.85 g. (m.p. 190-200') and after two crystallizations from methanol yielded 0.80 g. (98%), m.p. 208-209", of the expected disulfide, diacetylalethine (the disulfide derlved from IT, R = H); there was no depression on mixed n1.p. w t h a11 authentic sample. B. From ?-Acetamhopropyl Thioacetate (IV).-, ' 3 J . essentially the above procedure, hydrolysis of 1 g. of I\. and 0.1 ?I! alkali yielded 97% of the expected disulfide (corrcsponding to IVb), m.p. and mixed m.p. 88.5-90°.
(20) T E. King, C. J. Stewart and V. H. Cheldelin, THISJ O U R N A L , 75, 1290 (1953).
ROCHESTER, S Y.
[CONTRIBUTION FROM THE
DEPARTMENT OF
CHEMISTRY O F PURDUE UNIVERSITY]
The Effect of Ring Size on the Rate of Acetolysis of the Cycloalkyl p-Toluene and p-Bromobenzenesulfonatesl BY HERBERT C. BROWN AND GEORGE HAM? RECEIVED SOVEMBER 7, 1955 A number of cycloalkyl tosylates (4-, 5-, 6-, 7 - , 8-, 9-, 11-, 12-, 13-, 14-, 15- and 17-ring members) have been synthesized and the rates of acetolysis measured at several temperatures in order t o ascertain the effect of ring size on reactivity. The behavior of several brosylates ( 5 ,6- and 7-) was also examined. In the common rings, the cyclopentyl and cycloheptyl derivatives exhibit an enhanced reactivity attributed to the de-eclipsing of bonds in the ionization stage. On the other hand, the decreased reactivity of t h e cyclohexyl derivatives is attributed t o a n increase in bond opposition in the ionization stage. A further increase in reactivity is observed in the medium rings (8- t o 12-) followed by a decrease in the large rings (13- t o 17.) t o rates of reactions similar t o those observed in open-chain compounds. The maximum in reactivity in the medium rings is attributed t o a relief of strain accompanying the loss of a bond in the ionization stage. For the strained rings ( 5 - t o 11-members) a simple relationship exists between the association constants for cyanohydrin formation and the rate constants for acetolysis of the tosylates.
Ring compounds exhibit a remarkable change in chemical reactivity with ring ~ i z e . ~ I- t~has been proposed that the changes in chemical reactivity can be correlated with the changes in internal strain accompanying the formation or breaking of a bond to the ring atom in the rate-determining stage. -7 The available data are too few to permit a rigorous test of the utility of the proposed explanation. In order to obtain data of this kind we are currently examining the effect of ring size on the reactivity of ring derivatives in a few representative reactions. The present paper reports a study of the effect of ring size on the acetolysis of cyclic tosylatess and certain selected brosylates. (1) Chemical Effects of Steric Strains. XII. (2) Research assistant on a contract supported by the Office of Naval Research and a grant provided by the National 5cience Foundation. (3) V. Prelog, J. Chem. Soc., 420 (1950). (4) J . D. Roberts and V. C. Chambers, THISJ O U R N A L73, , 5034 (1951). ( 5 ) H. C. Brown, R. S. Fletcher a n d R. B . Johannesen, ibid., 73,212 (1951). (6) H . C. Brown and M. Borkowski, ibid., 74, 1894 (1952). (7) P . D Bartlett, Bull. SOC. chim., ClOO (1951). ( 8 ) In t h e course of discussions with Professor V. Prelog during the Fourteenth International Congress of Pure and Applied Chemistry in Zurich, July 21-27, 1955, it was learned t h a t Professor Prelog and Dr. R . Heck had carried out a similar study of t h e acetolysis of t h e cycloalkyl tosylates (6-, 7-, 8-, 9-, IO-, 11-, 12- and 20-ring memhers). These results have since been published: R Heck a n d Ti. Prelog, Helu. Chim. Acta, 38, 1541 (1955). T h e present study includes data on certain tosylates (4-, 5-, 13-, 14-, 15- and 17-ring members) a n d brosylates (5-, 6- a n d 7-) not included in the investigation by Heck
Results The p-toluenesulfonates of cyclobutanol, cyclopentanol, cyclohexanol, cycloheptanol, cyclooctanol, cyclononanol, cycloundecanol, cyclododecanol, cyclotridecanol, cyclotetradecanol, cyclopentadecanol, cycloheptadecanol and the p bromobenzenesulfonates of cyclopentanol, cyclohexanol and cycloheptanol were prepared by treating the alcohol with $-toluenesulfonyl or p-bromobenzenesulfonyl chloride in dry pyridine, essentially according to the method described by T i p ~ o n . ~ The $-toluenesulfonates of cycloheptanol, cyclooctanol, cyclononanol and cycloundecanol were relatively unstable, the cycloheptyl compound undergoing decomposition after several days a t room temperature and the others undergoing deThe composition after several days a t 0-10'. product obtained by treatment of cyclodecanol with 9-toluenesulfonyl chloride in dry pyridine decomposed rapidly a t room temperature upon removal of the ether with which i t had been extracted. Before the preparation could be repeated, it was learned that the compound had been synthesized in another laboratory and its rate of acetolysis measuredS8 Accordingly no attempt was made to repeat this preparation. The properties of the cyclanols and the aryland Prelog. Moreover, it should be of Interest to compare the agreement realized in a closely similar study in two different laboratories. Consequently, we are reporting ail of our results even where these duplicate those obtained by Heck and Prelog. (9) R. S. Tipson, J. Org. Chcm., 9, 235 (1944).
HERBERT C.BROWNAND GEORGEHAM
2736
Vol. 78
TABLE I
PHYSICAL PROPERTIES OF C Y C L A N OAL N~ D CYCLOALKYL ARYLSULFOXATES
-
n
Observed M.P., " C .
B.P., OC. (mm.)
Literature M.P., oc.
B.p., "C.(mm.)
n2@D
Ref.
nzoD
m Cyclanols, (CH2)n-1 CHOH 4 5 6 7 8 9 11 12 13 14 15 17
123.0-124.0 138.2 (754) 160.5-161.0 8 6 . 0 (20)
125.0 139 158-159 95 (21)
1.4529 1.4647 1.4770 1.4836 1.4901
115.0-116.0 (16) 118.0-119.0 (7)
g
1
L
d
i
f J d
115 (17) 128-131 (20)
d
75.0-76 0 59.5-60.3 79.0-80.0 80.5-81.2 80.0-81.0 7
h
1,4530 1,4642 1.477Zt6 1.48W5
80 60-60.5 79.0-80.0 81 81
d d d d d
24-25 27.0-28.0 43.5-44.0 Liq. Liq. Liq. 43-44 88-89.5
e e
Cycloalkyl tosylates, (CH2)n-1 CHOTs 4 5 6 7
24.0-25 0 28.5-29.0 44.5-45.0 19.0-19.6
1.5263 1.5276
8 9 11 12 13 14 15 17
43.7-44.5 44.2-45.4 87 0-87.6 36.3-37.2 77.5-78.8 47.2-48.0
C
k
k k k
k a a
a a
1.5092"
Cycloalkyl brosylates, (6Ht)n-1 CHOBS 5 6 7
45.8-46.6 48.5-49.0 32.0-32.5
45.5-46.0 48.1-48.6
j b
a
New compound. S. Winstein, E. Grunwald and L. Ingraham, THISJOURNAL, 70, 821 (1948). C S. Winstein, E. M. Kobelt, P. Barman, V. Prelog and L. Ruzicka, Helv. Grunwald, R. E. Buckles and C. Hanson, i b i d , 70, 816 (1948). J. D . Roberts and Chim. Acta, 32,256 (1949). e Ref. 4. H . H . Zeiss and M. Tsutsui, THISJOURNAL, 75, 897 (1953). H . E. Ungnade and D. C. W. Sauer, ibid., 71, 3925 (1949). C. R. Noller and R . Adams. ibid., 48, 1080 (1926). NightingaIe, ibid., 66, 1219 (1944). i R . B. Loftfield, ibid., 73, 4707 (1951). Ref. 8. 10
TABLE I1 RATECONSTANTS A X D DERIVED DATAFOR OF THE
5
6
7
8
9
ACETOLYSIS
11
CYCLOALKYL TOSYLATES ( C H n ) n - 1 CHOTS L
Temp., n 4
THE
OC.
70.0 80.0 90.0
R a t e constant ki, set.-' 2 67 X IO-@ 7 . 3 8 X 10-4 1 . 6 5 X 10-8
X X X X X
90.0
3 23 3.82 3.32 1.82 2.37 2.22
50.0 70.0 90.0 25.0 50.0 70.0 25.0 35.0 70.0
6.45 6.00 3.95 2.82 5.68 4.52 2.43 8.72 4.08
X 10-6 X IO-'
30.0 50.0 70.0 50.0 70.0
x x
X X X X X X
10-6 10-5 10-4 lo-@ 10-6 10-4
10-8 10-6 10-4 10-8 10-6
10-5 10-1
A
AH
*,
12
Rel. rate a t 70' 11.3
kcal./ mole
14.0
24.1
-4.2
14
27.3
-0.5
15
AS*, e.u. 13
25.0 50.0 35.0 50.0 70.0 50.0 70.0 90.0 50.0 70.0 90.0 50.0 70.0 90.0
1.00
25.3
191
23.3 -5 7 (23.5Ib (-5.4bb
17
22.3
20
(22.5)'
172
-4.5 ( - 4 . 1)'
22.5 -4.2 (23.9)b ( + O . l ) b
(n-CaH7)r CHOTs
50.0 70.0
(4.69 X 10-6)b (1.03 x
2.02 1.29 1.Ih 5.97 7.70 7.63 7.27 8.30 7.25 2.47 3.12 3.28 4.30 5.18 4.53
X 10-6 10-4
x
48 9
( 2 3 . 0 ) b (-1.1)' 23.7 (24.7)'
-3.2 (-0.2)b
X 10-8
X lo-@ X 10-6
3.25
27.6 (27.9)'
2.8
(3.8)b
X 10-4 X 10-6 X 10-6 X 10-4
3.50
2fi.2
X 10-6 X 10-5 x 10-4 X 10-6
1.32
27.8
2 19
26.5
-1.3
10-6 10-6 10-4
2.17
26.5
-1.4
10-8)b 10-5)b
1.80'
-1.3
1.7
X 10-5 10-4
90.0 x 50.0 4.38 X 70.0 5.15 x 90.0 4.60 X 50.0 (3.57 X 75.0 (7.52 X 70.0
380'
3 . 1 5 X 10-5
(26.6)* (-1.3)''
1.33
Initial rate constant, uncorrected for rearrangement. *Ref. 8. e Calcd. from extrapolated value of rate constant.
ACETOLYSIS OF CYCLOALKYL ARYLSULFONATES
June 20 1956
sulfonates used in this investigation are summarized in Table I. The rates of acetolysis of the cyclic arysulfonates were measured in acetic acid containing 0.046 mole per liter of acetic anhydride essentially according to the procedure described by Winstein and co-workers.10 I n order to have the value for a typical open-chain compound for comparison, 4-heptyl tosylate was synthesized and its rate of acetolysis determined. The results are summarized in Tables I1 and 111. TABLE I11 RATECONSTAXTS A N D DERIVEDDATAFOR OF THE
THE
ACETOLYSIS
CYCLOALKYL BROSYLATES (kH2)n-1 CHOBs L
A
*,
n
O C .
Rate constant ki, sec. - 1
AH* kcal./mhe
AS e.u.
5
25.0 50.0 70.0 50.0 70.0 90.0 25.0 50.0 60.0 70.0
6.28 X 1.29 x 10-4 1.10 x 10-8 6.60 X 10-8 8.02 x 10-5 7 . 2 8 x 10-4 1 . 0 4 x 10-5 2 . 2 3 x 10-4 6.70 x 10-4 1.98 x 10-3
22.7
-6.3
Temp.,
6
7
2737
-I
2Log k I.
-3-
-4 26.8
23.0
0.4
-4.2
The enthalpies and entropies of activation for the acetolysis of cyclopentyl and cyclohexyl p toluenesulfonates have been reported previously both by Roberts and Chambers4 and by Winstein and his co-workers.l 1 Unfortunately, their results are not in agreement. Therefore we undertook to measure the rate constants for these two compounds in order to compare our values with those previously reported. The available data are summarized in Table IV.
-5
c
1
I
I
I
I
I
I
I
27
28
29
30
31
32
33
I / T x io4 Fig, 1,-Activation energies for the acetolysis of cyclohexyl and cyclopentyl p-toluenesulfonates: 0, data reported by Winstein and co-workersI1; 0 , data reported by Roberts and co-workers4; (3, data obtained in this study.
Discussion
At the present time, three sources of steric strains are recognized: (1) the compression of van der Waals radii, ( 2 ) the distortion of bond angles, and (3) bond opposition forces. I n the case of small rings, the distortion of bond angles appears to be the major source of train.^^^ I n cycloTABLE IV pentane and cycloheptane it is the bond opposition ENTHALPIES A N D ENTROPIESOF ACTIVATIONFOR THE forces which are primarily responsible for the ACETOLYSISOF CYCLOPENTYL AND CYCLOHEXYL TOSYLATES strains revealed in combustion studies, l 2 while the AH* AS*, available evidence suggests that both bond opposiTosylate kcal./mole e.u. Ref tions and compressions of van der Waals radii are Cyclopentyl 27.6 6.7 Roberts' probably responsible for the strains exhibited by 23.7 -6 . 4 Winstein" the medium rings3 24.1 -4 . 2 Present study It was suggested that reactions involving an inCyclohexyl 27.4 0.6 Roberts4 crease i n these internal strains will be hindered, Winstein" 27.0 -1 1 whereas those proceeding with a decrease in internal 27.3 -0 . 5 Present study strains will be a ~ s i s t e d . ~Ideally one should discuss the effect of making or breaking a bond to one Roberts and Chambers found no significant dif- of the ring atoms in terms of each of these possible ference in the enthalpies of activation for the cyclo- strain factors. Unfortunately, this cannot be pentyl and cyclohexyl tosylates and theref ore done a t the present time. These different strains attributed the difference in reactivity to a large are not mutually independent nor can they be estientropy factor. On the other hand, Winstein and mated individually with any precision a t the his co-workers attributed the difference in reac- present time. tivity of the two compounds primarily to the difThe net change in internal strain accompanying ference of 3.3 kcal./mole in the enthalpy of ac- the formation or the breaking of a bond to a ring tivation. Our present results indicate a differ- atom can be estimated with considerable precision ence in the enthalpy of activation of 3.2 kcal./mole from rate, equilibrium and thermochemical data. (10) S. Winstein, C. Hanson and E . Grunwald, THISJOURNAL, '70, In view of this circumstance i t appears more de812 (1948). sirable a t the present time to discuss the reactions (11) S. Winstein, E Grunwald and L. Inxraham, ibid., 7 0 , 821 (1948): S Winstein, B. K. Morse, E. Grunwald, H. W. Jones, J. Corse, D . Trifan and H. Marshall, ibid., 74, 1127 (1952).
(12) K . S . Pitzer, Scrence, 101, 672 (1945); J. E. Kilpatrick, K. S. Pitzer and R . Spitzer. THIS J O U R N A L , 69, 2483 (1017).
2738
HERBERTC. BROWX APiD GEORGE H.khf
Yol. 75
of ring compounds in terms of the net change in internal strain accompanying the reaction (Istrain). It should be recognized that I-strain will be made up of individual contributions from angle deformations, bond oppositions and atomic compressions, the relative magnitudes of which will vary from system to system. The solvolysis of secondary alkyl p-toluenesulfonates in dry acetic acid is considered to proceed by a mechanism involving a slow rate-determining ionization of the p-toluenesulfonate group. l 3
introduced two partial oppositions, l 4 so that this molecule should tend to resist the ionization reaction in comparison to an open-chain derivative. Prelog has presented convincing evidence that the medium rings are strongly ~ t r a i n e d . ~The strain is presumably the effect of bond opposition forces, with some contribution possible from atom compression. Here also the internal strain should be partially relieved by a decrease in the number of bonds. Finally, the large rings should be essentially strain free and the reactivities should approach those of the open-chain compounds. RI R? Ri The experimental results (Fig. 2 ) are in complete ki \ /' accord with the predictions of the I-strain interpreH-C-OTS + c+ o-rstation. / R; E1 According to the I-strain hypothesis the opposite effects should be observed in rate or equilibrium The I-strain theory predicts that such ionization data involving a reaction in which there is an inreactions should proceed relatively slowly in the 3- crease in the number of bonds attached t o a ring and 4-membered rings primarily because of an in- atom. The similarity in the effects of ring size crease in angle strain accompanying i ~ n i z a t i o n . ~on the rates of acetolysis of the cycloalkyl tosylates Such a slow reaction has been observed in the cyclo- (Fig. 2 ) and on the equilibrium constants for the propyl tosylate and in l-chloro-l-methylcyclo- formation of the cyclic cyanohydrin^^^ is noteworthy. butane.6 However, the rate of solvolysis of cyclor--1 r -- ---1 butyl tosylate is relatively fast.4 The fast rate of ( C H r ) n - l C=O + C S (CH2)n-1 C
