3710
C. H . DEPUYAND D. H. FROEMSDORF [CONTRIBUTION FROM T H E
DEPARTMEST OF
VOl. 79
CHEMISTRY, IOWA STATE COLLEGE]
Electronic Effects in Elimination Reactions. 11. The Ez Reaction'"''' BY C. H. DEPUYAND D. H. FROEMSDORF RECEIVED FEBRUARY 8, 1957 The base-promoted eliniinatioii reaction of substituted P-phenylethyl halides, tosylates and sulfonium salts has lxeii studied as a function of substituents. Good fits to a Hammett u-p plot were obtained, indicating that the acidity of the hydrogen being removed is controlling the direction of elimination. In contrast t o their reactivity in solvolysis and displacement reactions, tosylates are eliminated very slowly as compared to bromides and iodides.
In connection with a general study of electronic effects in elimination reactions,lb we were led to an investigation of the electronic effects in the Ez reaction, both because of the interest in the reaction for its own sake2 and because such a study would serve as a reference for our investigation in the similar 1,3-diphenylpropyl system. Surprisingly, no comprehensive investigation of substituent effects on E:,eliminations has been reported3 except for some polyhalo compounds4 and for a few P-phenylethyl chloride^.^ Quantitative data on E2 tosylate eliminations are likewise unavailable.6 Our data for the elimination of P-phenylethyldimet hylsulfoniuni iodide, 0-phenylet hyl bromide, iodide and tosylate are summarized in Tables I and 11. For the sulfonium salts p = 2.75 rt 0.21, the highest value of p for the series. For the bromides p = 2.14 i 0.15 and for the iodides p = 2.07 + 0.09. Omitting the point for the 9inethoxy group, p = 2.50 f 0.06 for the tosylates. -A high positive value for p implies that in the transition state of the reaction a large amount of negative charge has developed, and those groups which make a negative charge more stable speed up the reaction.' I t has been suggested that the Saytzeff rule is a consequence of a transition state in which the stability of the incipient double bond is the dominant characteristic.*"J Our results indicate that when the hydrogen being removed is on a carbon adjacent to a benzene ring, the stability of the product being formed is not of importance,9 for if it ( I ) ( a ) Supported b y a Grant-in-Aid from t h e hlonsanto Chemical Cu. Presented in p a r t a t t h e 130th Annual Meeting of t h e American Cheiuical Society, Atlantic City, N. J., September 16-21, I95G. (b) Paper 1 in this series, C. H . DePuy a n d R. E. Leary, THIS J O U R N A L , 79, 3705 (1957). ( 2 ) (a) C . K . Ingold, "Structure a n d Mechanism in Organic Chemistry," Curnell Univ. Press, I t h a c a , N. Y.,1053, p . 420; (b) W. H. Saunders a n d R. A. Williams, THISJ O U R N A L , 79, 3712 (19.57). (3) At t h e time of our preliminary announcement of some of these results, w e were informed b y Dr. W. H. Saunders of t h e results of some excellent work t h a t was being done i n his laboratory along similar lincs.2b T h e existence of these d a t a , more complete t h a n our own in terms of t h e temperature range studied b u t confined t o t w o leaving groups, has in0uenced t h e presentation of our d a t a . W e shall be cont e n t t o point o u t a few conclusions pertinent t o our main interests a n d t o refer t h e reader t o t h e report of Dr. Saunders for much of t h e background. W e are grateful t o Dr. Saunders for interesting discussions a n d for d a t a on t h e extinction coefficients of some of t h e olefins. (4) S. J. Cristol, N. L . Hause, A. J. Quant, H. W. Miller, IC. 11. Eilar a n d J. S. hfeek, THISJ O U R N A L , 72, 3333 (195%). ( 5 ) h1. Simonetta a n d G Favini, J . Chem. SOL.,1840 (1054). (6) C j . R. P. Linstead, L. h-,Owen a n d R. F. Webb, ibid., 1218 (1953). ( 7 ) See K. W. T a f t , J r . , in "Steric E f f e c t s in Organic Chemistry," Ti>hn Wilcy a n d Sons, Inc.. h e w York, N. Y . , 1956, p. disc,iis,ion o f t h e mcchanislic interpretation of p . i s ) if C .U i o w n a n d I. M o r i t x i i , THIS J O U R N A I7. ,8 , 2 2 0 3 ( I V > l , ) I!!) See, huncvcr. t h e discussiun of the devidtiun uf the P-meLhuxy1
10sylate.
were the p-methoxystyrene should be formed most rapidly.lbP1O The fact that the sulfonium salts, bromides and iodides, and to a major extent the tosylates,$ all show a good correspondence of rate with a measure of hydrogen acidity is to be contrasted with our results with acetate pyrolysis reactions in which a good correlation with product stability was observedlb in an analogous system. In these two cases we have demonstrated the extremes of transition state structural variation, from one in which double bond development is too small to be significant (E*) to one in which the stability of the double bond being formed determines the rate of elimination. RATESOF X
Tim
TABLE I ELIMINATION REACTION OF XCsHdCHzCHsY a t 30.10'
Y
H
Br p-C1 Br p-OCH, Br m-Br Br lr I
Olefin,
k
%
4.10 x 1.88x 1.73 x :3.71 x 2.66X
10-4 10-3 lo-" 10-3 10-2
kE2
100 100 100 100 100 100 100 100 33 47 70
lit-Br H p-Cl P-OCI-II m-Br
I OToi OTOS OTos +
1.05x 9.56 X 2.03 x 1.18 x 2.93 x s.01 x 3.53 X
11
S(CH3):I
:j.iCJ X 1 0 - J
100
x o . 13 x
IO0 100
p-c1 I pOCH; I OTos
+
/J-c1 s( C ~ I ; ) L I />-ClIj
S(CH3)zI -
2.18
10-2 10-4 10-4 10-& l0F4
Ci 7
10-~
10~-4
1 io x 1.88 x 1.73x 3.71 x 2.66 X 1.05 X 9.56 x 2.03 X 3.92 X 1.38 x 1.69 X 3 60 x
x 2 1s x 9.13 x 3 79
10-4 10-3
10-4 10-3
10-4
10-4 10-4 10 - 3 10 2 10"
TAULE 11 I
(1neLhud ,,i least squares!
Tosylates (omitting p-iuethoxy) Tosylates (including p-mcthoxy) Bromides Iodides Sulfonium salts
3.50 i 0 . 0 6 2.04 +Z .47 2.14 rt .15 2.07 d= .09 2.75 d= .21
luc h" (calcd ) Intercept
- 4.427 -4.280
-3.2GO -2.497 -2.330
The fact that the p-value for the sulfonium salts is larger than that for the bromides and iodides implies that the transition state for Hofmann eliminations has significantly more carbanionic character than that for the elimination of iodides or bromides. 'CTihether this difference would be enough to account for some of the differences in direction of lialide and 'onium salt elitninatio~lsin strictly d i phatic systems? caiitiot yet be said with ccrtainty. ~ 1 U iC . I; Ingold d n d C . \V S h u y p e e , J . L / I C W .S u c . , 417 (1922,
July 20, 1957
ELECTRONIC EFFECTS IN Ez ELIMINATION REACTIONS
Perhaps the most surprising result of this investigation is the very slow rate of elimination of the tosylates. For the unsubstituted cases, kI/ kTos = 68 and k B r / k T o s = 10. In the case of direct displacement reactions, sulfonic acid esters appear to be more reactive than either bromides or iodides," although admittedly the data are meager. This slowness in the elimination reaction has a synthetic as well as a theoretical value. From Table I it can be seen that the percentage olefin in the product varies from 20 to 67% for the tosylate eliminations. I n all other cases of leaving groups studied, the olefin yield was close to 1007,. Hence, while the E P reaction is slow, apparently the S N reaction remains fast. For this reason the tosylate can be useful in preparing the corresponding ethers or amines12 when the iodides and bromides lead only to olefin. The meagerness of available data makes a firm theoretical conclusion hazardous, and we shall discuss in subsequent papers the results of a more detailed study of tosylate elimination. The 9-methoxy tosylate eliminates significantly faster than would be predicted on the basis of the p-value of the reaction. This may imply that, in this relatively unactivated case, soiiie effect of the olefin stability is being felt in the transition state. Experimental Preparation of 2-Arylethanols. 2-(m-Bromophenyl)-etanO~.-m-Bromodcetophenone was prepared by bromination of acetophenone in the presence of excess anhydrous aluminum ~ h l o r i d e . ' ~This was transformed into m-bromophenylacetic acid by heating under reflux for 16 hr. 103 g. of the ketone with 68 g. of morpholine and 25 g. of sulfur. The resulting thiomorpholide was hydrolyzed by boiling with 10% potassium hydroxide for 18 hr. and then acidifying. The precipitated m-bromophenylacetic acid weighed 78.7 g. (70.770 yield) and had m.p. 99-100" (lit." m.p. 100"). Lithium aluminum hydride reduction of the ethyl ester gave thealcohol, b.p. 107-110" (1 mm.). 2-(p-Chlorophenyl)-ethanol was prepared by refluxing pchlorobenzyl cyanide with 9570 ethanol and concentrated sulfuric acidlj to yield the ethyl ester of p-chlorophenylacetic acid. Reduction with lithium aluminum hydride gave the alcohol (75.570 yield), b.p. 98-100" (1 mm.) (lit.I6 b.p. 110' (0.5 mni.)). 2-(~-1LIetlioxyplic~iyl)-ctha1i~l \vas preparcd fro111 p ~iictlioxyacetophenonc by the sanie niethod17 as the 2-(mbroinopheny1)-ethaiiol, b.p. 110-113° (1 mm.) (lit.'* 148' (19 nim.)). 2-Phenylethaiiol was Eastman Ludak Co. LVhite Labcl. Preparation of Tosy1ates.-The tosylates were prepared from the corresponding alcohol by Tipson's proccdure.10 (11) A . Streitwieser, Chenz. Reus.. 66, G O 2 (1933). (12) See D. 3.C r a m , F. D. Greene and C. H. DeI'uy, THISJ O U K '78, 790 (1936), f o r a synthetic example. (13) D. E. Pearson a n d A . TV Pope, J . Org. C h i n . , 21, 381 (19jtij. (14) H . Berger, J . p r a k l . Chem., 133, 315 (1932). (15) H. Gilman and 4.H . B l a t t , "Organic Syntheses,'' Coil. Vol. I , John Wiley & Sons, I n c . , New York, iY.Y., 1941, p . 270. (16) G . Baddeley and G. &I. Bennett, J . C h e m . Soc., IS19 (1935). (17) U. V. Solmssen and E. Wenis, THISJ O U R N A L , 70, 4200 (1948). (18) G. 31. Bennett and bl. Hafaz, J . Chem. Soc., 652 (1941). (19) It. S . Tipson, J . Org. Chem., 9 , 233 (1944). NAL,
3711
2-(m-Bromophenyl)-ethyl-p-toluenesulfonate,m.p. 4344". Anal. Calcd. C, 50.75; H, 4.26; S, 9.02. Found: C, 51.05; H, 4.54; S, 9.44. 2-(p-Chlorophenyl)-ethyl9-toluenesulfonate, m.p. 79.380.3'. Anel. Calcd. C, 58.00; H, 4.87; S, 10.32. Found: C, 58.00; H, 4.76; S, 10.69. 2-(p-Methoxyphenyl)-ethyl p-toluenesulfonate, m.p. 58.6-59.6'. Anal. Calcd. C, 62.72; H, 5.92; S, 10.46. Found: C, 62.86; H, 5.82; S, 10.60. 2-Phenylethyl p-toluenesulfonate, m.p. 38.5-39". Anal. Calcd. C, 65.20; H, 5.84; S, 11.60. Found: C, 65.45; H, 5.77; S, 11.25. Preparation of 2-Arylethyl Bromide.-The bromides were prepared by addition of the appropriate tosylate to a solution of anhydrous lithium bromide in acetonem and careful fractionation of the resultant bromides: 2-(m-bromophenyl)bromide, b.p. 98-100" (1 mm.); 2-(p-chlorophenyl)~ethyl ethyl bromide, b.p. 79' (0.5 mm.); 2-(p-methoxyphenyl)ethyl bromide, b.p. 79" (0.5 mm.); 2-phenylethyl bromide was Eastman Kodak Co. White Label which was redistilled, b.p. 65' (2 mm.). Preparation of 2-Arvlethvl Iodides.-The iodides were prepafed by addition df theappropriate tosylate t o a solution of anhydrous sodium iodide in acetone2O and fractionation: 2-(m-bromophenyl)-ethyl iodide, b.p. 105-110° (1 mm.); 2-(p-chlorophenyl)-ethyl iodide, b.p. 115-117' ( 1 mm .) ; 2-( p-methoxyphenyl) -et hyl iodide, b .p. 117-1 19 (1 mm.); 2-phenylethpl iodide was Eastman Kodak Co. White Label which was redistilled, b.p. 95' (3 mm.). Preparation of Sulfonium Salts.-The sulfonium salts were prepared by treating the appropriate iodide with sodium methylmercaptidea; this gave the sulfide. The sulfide was then treated with methyl iodides in acetonitrile which gave the dimethylsulfonium iodide: 2-(p-~hlorophenyl)-ethyldimethylsulfonium iodide, m.p. 127-128'; 2-(p-niethoxyphenyl)-ethyldimethylsulfonium iodide, m.p. 118-1i9'; 2-phenylethyldimethylsulfonium iodide, m.p. 124-125 Procedure for Kinetics.-The reactions were carried out in 100-ml. volumetric flasks. About 0.06 mole of the desired compound was placed in the volumetric flask and weighed accurately. Then 0.12 N sodium ethoxide in absolute ethanol, previously equilibrated thermally with the bath, was added t o the calibration mark of the volumetric flask. After adequate agitation, the flask was placed in the bath and allowed to come to thermal equilibrium. Aliquots were then withdrawn periodically with a calibrated automatic pipet. A modification of this method was necessary for the sulfonium salts because of their slow rate of dissolution. T h e weighed sulfonium salt was dissolved in 50 ml. of absolute ethanol and 0.24 N sodium ethoxide in absolute ethanol added so the total volume was 100 nil. Olefin Determinations.-The reactions were allowed to go t o completion, and 25-ml. aliquots were poured into water and extracted with carbon tetrachloride (3 X 10 ml.), the extract washed with water and the amount of olefin determined by bromination with standard broinide-brornate solution.21 Duplicate runs agreed to within 106. In the bromination of the p-methoxystyrene it was necessary t o modify the above method by adding the bromidebromate solution in 1-id. aliquots until a color remained t o the solution. This was necessary t o minimize substitution by avoiding a n excess of bromine. In the case of sulfonium salts it was necessary to wash the carbon tetrachloride layer twice with saturated mercuric chloride solution t o extract dimethyl sulfide. AMES,IOWA
.
( 2 0 ) H. C. Brown and 0. H. Wheeler, THISJ O U R N A L , 78, 2109 (1956).
(21) Sidney Siggia, "Quantitative Organic Analysis via Functional Groups," 2nd E d . , John Wiley & Sons, Inc., Kew York, 9. Y., 1054, p . tis.
