July 5, 1952
KINETICSOF DEHYDROGENATION OF 2,2-DIARYLCHLOROETHANES
117' ,I4 170-170.5' , l e 167.5-169', '6 163" , l o 164-165", '1 sinter 163', m.p. 171.5O.18 Anal. Calcd. for C ~ Z H ~ & & C, O ~52.17; : H , 4.38; N, 20.28. Found: C, 51.92; H , 4.35; N, 20.47. Ethyl A2-Cyclohexenone-3-acetate(V) .-A mixture of 11.7 g. (0.075 mole) of the monoketal (I), 13.4 g. (0.08 mole) of ethyl bromoacetate, 2 g. of cleaned granulated zinc, and it crystal of iodine was refluxed and stirred for 5.5 hours in 100 ml. of benzene. During this time fresh portions of zinc and iodine were added at half-hour intervals and after two hours another 13.4-g. portion of bromoester was added. Acidification of the reaction product with cold 10% sulfuric acid and extraction of the benzene solution with bicarbonate gave 2 g. of 1,3-cyclohexanedione. Fractionation of the neutral residue after removal of the benzene gave 5.0 g. (36%) of V; b.p. 108" (1 mm.); n y 31.4879; ~ Amax 233 mp (log e 4.14). Anal. Calcd. for C10H1403: C, 65.91; H , 7.74. Found: C, 65.70; H , 7.55. The dinitrophenylhydrazonc melts at 107-108' aftcr crystallization from alcohol. ilnal. Calcd. for C1&806N4: C, 53.03; H , 5.01; N, 15.46. Found: C, 53.04; H,4.94; N, 15.83. 3-Cyano-A2-cyclohexenone (IV).-A solution of 24.7 g. (0.13 mole) of sodium metabisulfite in 40 ml. of water was added dropwise with stirring and cooling to 40 g. (0.256 mole) of the monoketal (I). To the resulting homogeneous solution was added 12.7 g. (0.26 mole) of sodium cyanide in 25 ml. of water. The mixture was stirred at room temperature for 20 hours; it was then filtered, and the filtrate was extracted with ether. The ether was removed and to the residue was added 40 ml. of 3 N hydrochloric acid; the homogeneous solution was allowed to stand 24 hours during which time the product separated from solution. The aqueous solution was saturated with ammonium sulfate, extracted with ether, the ether was removed, and the residue (14) (15) (16) (17) (18) (19)
F. C. Whitmore and G . W, Pedlow, i b i d . , 63, 758 (1941). K. Dimroth and K. Resin, B e y . , 76B,322 (1942). A. J. Birch, J . Chcm. Soc., 430 (1944). A. J. Birch, i b i d . , 1270 (1947). M. Orchin, THIS JOURNAL, 66, 535 (1944). H. Adkins and G. Rossow,i b i d . , 71, 3836 (1049).
[ C O N rRlBUI ION FROM TIfE
3333
was fractionated to give 16.8 g. (54%) of the cyanoketone (IV); b.p. 105' (3.8 mm.); nzlD 1.5104; A,, 236 mp (log E 4.08). Anal. Calcd. for CTH~NO:C, 69.40; II,5.82; N, 11.57. Found: C, 69.11; H, 5.94; N, 11.97. Both the crude cyanohydrin and the unsaturated cyanoketone gave the same dinitrophenylhydrazone; m.p. 204204.5' after recrystallization from ethanol. Anal. Calcd. for C1OH11NSO4: C, 51.83; H , 3.68; N, 23.25. Found: C, 51.68; H , 3.89; N, 22.88. The cyanoketone readily formed an ethylene ketal; b.p. 96" (1 mm.); n% 1.4968. Anal. Calcd. for C9H1102N: C, 65.44; H , 6.71; r\i, 8.48. Found: C, 65.03; H , 6.92; N, 8.14. 3-Ethoxy-A2-cyclohexenone (VI).-A mixture of 1,3-cyclohexanedione (28 g., 0.25 mole), 50 ml. of absolute ethanol and a few crystals of*$-toluenesulfonic acid in 500 ml. of benzene was allowed to reflux under a modified Dean-Stark trap for 16 hours. Removal of the solvent and fractionation of the residue gave 24.8 g. (71%) of the enol ether (VI)2o; b.p. 76" (1.0 mm.); nZ3D1.5030. The enol ether gave no derivative with the D.N.P. solution. Diethyl 3-Carboxymethylene-A2-cyclohexenylacetate (VII).-Application of the Reformatsky reaction to 10 g. of the enol ether, using the same conditions as for the monoketal, gave 6.8 g. (37%) of a product, b.p. 110' (1 mm.), %17D 1.5311, Amax 298 mfi (log t 4.44). Anal. Calcd. for ClaH200c: C, 66.64; H , 7.99. Found: C, 66.56; H , 8.16. LVhen the reaction was carried out with less bromoester a mixture of mono- and disubstituted products was obtained. From 9.5 g. of enol ether (VI) and 15 g. of ethyl bromoacetate there was obtained 6.8 g. of a product with a b.p. of 108-109' (1.0 mm.); 7 2 2 4 ~1.5178; imax 233 mp (log e 3.86) and 298 mp (log e 4.32). (20) Woods and Tucker6 prepared the enol ether b y the action of ethyl iodide on the silver salt of 1,3-cyclohexanedione. The azeotropic method with p-toluenesulfonic acid as a catalyst has been used recently on substituted 1,3-cyclohexanediones by R. L. Frank and H. K. Hall, ibzd., 72, 1645 (1950).
BERKELEY 4, CALIFORNIA
DEPARTMENT OF CHEMISTRY, UNIVERSITY OF' COLORADO]
Mechanisms of Elimination Reactions. VI. The Kinetics of Dehydrochlorination of Various 2,2-Diarylchloroethanes BYSTANLEY J. CRISTOL, NORMAN L. HAUSE,ALFREDJ. QUANT, HAROLD W. MILLER, KENDRICK R . EILAR AND JOHN S. MEEK RECEIVED DECEMBER 21, 1951 A study of the kinetics of dehydrochlorination of fourteen para substituted 2,2-diaryl-monochloro-, -dichloro- and -trichloroethanes with cthanolic sodium hydroxide at various temperatures has been conducted, and the energies and entropies of activation have been calculated. A discussion of the effects of iiicrcasiiig alpha chlorine substitution upon dehydrochlorination reactivities and upon the energies of activation is included.
No systematic study of the effects of adding halogen atoms to the a-carbon atom of alkyl halides in their second-order elimination of the elements of hydrogen halide with alkali seems to have been made. In the course of our work on elimination reactions, it seemed worthwhile to initiate such a study. The compounds studied were of the type (1) This work was supported by the Office of Naval Research and was presented in part before the Division of Organic Chemistry at the San Francisco Meeting of the American Chemical Society, March, 1949.
(2) Previous paper in series: S. J. Cristol and N. L. Hause, THIS JOURNAL, 74, 2193 (1952).
1
where X and TT may be hydrogen or chlorine atoms, and the reactions studied were of the type HO-
I 1 + HC-CCl I
/
----f
Hz0
+ >C=C< + C1-
(1)
Some of the data desired have already been published,a but this paper completes the work on these (3) S . J. Cristol, i b i d . . 67, 1494 (1945).
coiiipounds and reports work for a total of fourteen compounds. Keactioii-rate co iistants were obtained for deh!.drochloriiiatioii with sodium hydroside in 92.6 weight per cent. ethanol a t three teiriperatures for six trichloro coriipouiids of type 1 (X = Y = Cl), hereinafter called "DD'T" corripounds, with K = CH3, CH&, H, F, C1 or ISr, six siiiiilarly coiistituted ciichloro compounds ( I , X = C1, 'I' = H), hereinafter called "1)DD" C O l I l ~ J ~ ~ l l i t k rtiicl two nionochloro ~ o i i i ~ ~ ~!I,~ X i i= i d1~- = Ii), hereinafter called ' ' I ~ n' ' ~~Oltl~~(JUlldS, I with IC = Cl or Br. Solutions of known halide and hydroxide conceriI1 trations were prepared, and the rates were followed by titration for chloride ion of dicpots removed a t various times during the course ~f a run. The niethod used was similar t h a t prcviously c l e scribed.3 I Satisfactory values were obtained with tlic scvond-order rate expression (first order in halide coilcentration and first order in hydroxide ion ~ o i i c e ~ i tration) for all of the compounds studied. 111 each Cl case aliiiost exactly oiie iiiole of chloride ion was ~)roducedper mole of halide after long intervals of tinic. In the case not previously studied, it was sliown that the corresponding olefin was produced by action of sodiuiii hydroxide on the halide. Ta131 ble I presents the (lata arid results of these espcriinents. The experiiiierital energies ~i activation, E,,., were calculated from the data of Table I in the usual fashion. These values, combined with values of the reaction-rate constants k corresponding to :t teniperature of 30' (obtained from the plots of log k w. i 1') wcrc' iiscd i i i tlic iippropriatc ~ q ~ i a t i o i ti) i! C1 'l'.\lAl.l -1\ Lr
t.4C I I~,l.Y-R:Yl
F
I- 0
( c ) LID11 Compoullds ( I ,
5240
776 2380
7.570
x
f = 1-1)
S'OI< a , ,If ?(I 11 :i0 3T I l l I:! 3 J ill;
2!J.(i!J
1'Al-d
10 I 2 a
3.33 3.38 12.6 12.5 42.8 42.5 4 :$.I a.32 15.7
0 03349 0.5349 05361 ,05361 or,3 53 . 05.353
0.7 2.1
0 00-$970 0 O X 2 1 .00504>1 . 0.-)621 .0.56?1 ,009931 ,008667 .03621 ,01001 .0>821 , 0100!l . 05621 ,01022 ,05621 ,005249 .Oh307
ljcitLlobtaiiicti ftoiii pre\ iou\
0 6
834 Sa1 126 127 40.; 406
11iU
11i0 1170 17 01
h
.I1
J
c,ilculate values for the eiitropies of activation AS*. The values for energies and entropies of activation are given in Table 11. The estimated uncertainties in Eactare 0.3 kcal./mole and in AS*, about 0.S cal. per degree per mole.
Discussion of Results \\'c wish first to discuss the effects 01 ackliiig lialogeii atoiiis to the systeni with fixed pdra substitucnts. I t will be noted froiii the data of Table I that the DDT (trichloro) compounds have specific reaction-rate constants (at 30') 2.4 to 4.3 times as great as the DDD (dichloro) compounds, and the dichloro compounds, 5 to 6 times those of the DDM (monochloro) compounds, that is, the addition of halogen atoms to the a-carbon atoms of the alkyl halide results in an increased reactivity toward dehydrochlorination with ethanolic sodium hydroxide. It is to be noted particularly, however, that the increases in reactivity are due to different eifects as iiidicatcd by the cjuaiitities of activation in l'ablc 11.
July 5, 1952
KINETICSO F
DEHYDRUGENA‘TION UP 2,S-131ARYLCHLoRo~TIANES
33%
ClJ-
TABLE I1
c z c
ENERGIESAND EXTROPIES OF ACTIVATION FOR THE DETYPEI COMPOUNDS WITH SODIUM HYDROXIDE IN 92.6% ETHAXOL
HYDROCHLORINATION O F Para substituent
Compound type
Eact
AEacta
kcal./mole
AS8$
H’ H08-
AASZ~~
cal./deg./mole
then the energy of activation will be a function of (a) the ease of heterolytic fission of the carbon-hydrogen bond and (b) the ease of heterolytic fission of the carbon-chlorine bond assuming that the carbon-carbon bonds are similar in each case. (The validity of this assumption will be discussed below. It is likewise assumed that the solvation characteristics of the initial us. the transition state are independent of the amount of chlorine substitution.) Unfortunately, direct measurements of either (a) or (b) are not available, but the argument from analogy proceeds as follows : Taking the monochloro compounds as the basis for discussion, addition of more chlorine atoms to give the DDD and D D T compounds should lower the energy necessary to bring about the fission of the C-H bond into C- and Hi, due to the negative inductive effect of chlorine (compare the acid strengths of mono-, di- and tri-chloroacetic acids).”,” This, then, should tend to deThus, the increase in reactivity of the D D T com- crease the energy of activation of the elimination pounds over the corresponding DDD compounds is process.13 Arguments regarding the effect of addition of a reflection of an increase in activation energy of approximately 1.6 kcal./mole for the dichloro coni- chlorine atoms on the strength of the carbon-chlorpounds (equivalent to a factor of 14 in rate) com- ine bond undergoing fission are not so straightforpared with the trichloro compounds, this factor be- ward, but here again certain analogies are available. ing partially compensated for by a corresponding Brockway14has measured the carbon-fluorine bond increase in the molar entropy of activation of 3.1 lengths in the fluoromethanes, and has obseryed a entropy units (equivalent to a factor of approxi- bond shortening in difluoromethane (l.a6 A,) as mately 5 in rate). On the other hand, the activa- compared with methyl fluoride (1.42 A.), This tion energies of the inonochloro and dichloro com- shortened bond length has been assumed to indipounds are subs tantially identical, and the decrease cate a stronger carbon-fluorine bond. -1ddition of in reactivities of the DDM compounds compared a third or fourth fluorine atom does not further with the DDD compounds is due entirely to a de- shorten the bond. Although bond strengths (homolytic fission) and “ionic” bond strengths (heterocrease in the entropy of activation. Wre should like to consider the following tentative lytic fission) are not necessarily parallel, it seems likely that there might be such a parallelism here explanation of the activation energy data. The removal of the elements of hydrogen halides and that one may conclude that addition of a secfrom an alkyl halide by base in a second-order (E2) ond fluorine atom to fluoromethane stabilizes the elimination has generally been r e p r e ~ e n t e d ~ +to~ carbon-fluorine bond markedly, whereas addition of the third and fourth atoms do not influence the proceed as bond strengths to as great a degree. Similar results are apparent from silicon-chlorine bond distances in the c h l o r ~ s i l a n e s ~and ~-~~ the methylchlorosilanes15,18J9and have been discussed by Walsh.20 >C=C< + x- (2) Although any bond shortenings in the chloroiiictlianes are too small to be significant experiaieiitally, as a one-stage concerted process involving simul(11) G. N. Lewis, THISJ O U R N A L38, , 762 (1916). taneous loss of the proton, formation of the carbon(12) H. B. Watson, “Modern Theories of Organic Chemistry,” Scccarbon double bond and departure of halide ion.1° ond Edition, Oxford Press, 1941, p. 27. (13) A. E. Remick, “Electronic Interpretations of Organic ChcmIf the transition state for concerted E2 elimination istry,” John Wiley a n d Sons, I n c . , New York, N. Y.,1943, pp is represented as cI-130 UDT 21.4 1.6 -6.1 3.6 DDU 23.0 -2.5 CH3 DDT 21.0 1.3 -7.1 2.7 DDD 22.3 -4.4 H DDT 20.5 1.4 -6.5 2.9 DDD 21.9 -3.4 F DDT 19.8 1.8 -4.6 3.3 DDD 21.6 -1.3 c1 DDT 18.3 -5.5 DDD 20.1 1.8 -2.3 3.2 DDM 20.2 0.lC -5.8 -3.5d Ur DDT 18.8 -3.0 DDD 20.6 1.8 0.0 3.0 DDM 20.2 -0.4“ -5.0 -5.0d a Difference in energy of activation of DDD and of D D T compound. Difference in entropy of activation of DDD and of DDT compound. Differenceof energy of activation of DDM and of DDD compound. Difference of entropy of activation of DDM and of DDD compound.
2 19-222.
( 5 ) E. D. IIughes a n d C. R . Ingold, Trans. Faraday Soc., 3 7 , 657 (1941). (6) C. R. Hauser, THISJOURNAL, 62, 933 (1940). (7) P. S. Skell a n d C. R. Hauser, ibid., 67, 1661 (1945). (8) S. J. Cristol, ibid., 69, 338 (1947). (9) A I . L. D h a r , E. D. Hughes, C. K. Ingold, A. M. M. Illandour, G. A. M a w a n d L. I. Woolf, J . C h e m Soc., 2093 (1948). (IO) This mechanism has been suggested a s t h e usual one lor “lunns” elimination,zre a n d t h u s presumably is involved for systems without restricted rotation around t h e carbon-carbon bond.
(14) I*. 0. Brockway, J . P b y s . Che?n., 41, 193 (1837). (15) L. 0. Brockway a n d F. T. Wall, T m s J O U R N A L , 66, 2373
(1934). (16) L. 0. Brockway a n d J. Y . Beach, ibid., 6 0 , 1836 (1938). (17) L. 0. Brockway a n d I. E. Coop, Trans. Faraday SOC.,8 4 , 1429 (1938). (18) R. L. Livingston a n d L. 0. Brockway, THISJ O U R N A L , 66, 94 (1944). (lr3) R. I,. Liviiigston and L. 0.Brockway, ibid., 6 8 , 720 (1046). s SOC.,2, GI(1847). (20) A. D. Walsh, D i s c u s s i o ~ ~Faraday
3336 S. J. CRISTOL,N. L. HAUSE,A. J. QUANT,H. W. MILLER,K. R. EILARAND J. S. MEEK Vol. 74 LValsh and Skinner and Sutton?’ give strong arguineiits for a strcngtlieniiig oi tlie carbon-chlorinc bond in dichloroniethanc by tlic substitution oi the secand chlorine atorri iiito uicthyl chloride, whcrcas Lhc addition of a third atom docs iiot bring about it corrcsporiding illcrease iii bond strength. ‘This lessening of effect inny ~ ~ bc1 c!uc1 to tlie strong rc(IC $ clJ--*:>C--C< -1 ci(311) 1)ulsive interactions betwecn the non-bonded chlorine atoms, which arc suggestec! to be rcsponsible ior If step (:Sa) is rate-determining, then the effect of tlie dipole irioiiicrits observed for the clilorornetli- adding chlorine atoms would very likely bc cumula:uies.22 These reglulsions (similar to J3-strai1i23) ti1.c. A mechanism involving step (3a) as reverscauscd by over1o;ttling of the carbo11atoiii I+* bulky ible and (3b) as slow might conceivably be satisiiegative groups, iiioy operate to make the -CC13 factory, but until recently no evidence has been system less stable t h a n the -CHClp systciii. obtained to support such a mecl-ianism in a dehyAisthe ;ttiditioii oi c x i i c.hloriiic atoiu causes a drolialogenation p r o c ~ s s . ’ ~ ~ ~ ~ clecrease in the carboti--h~drogciibond strength, it .Is rioted above, in addition to the rcgulsr differ’ thus possible to cq)laiIi the identical values for ciices i i i activation energies between series, there are for tlic I)I.)lI’s :~iid I>Dl)’s, addition of tlic also regular differences in entropy of activation, the second chlorine atoiii causiiig two opposite and dichloro series having approximately 4.2 e.u./mole q u a l effects. The lowcr value for Lact for the more iavorable activation entropy than the nionoUDT’s as coiiiparcd with the DDD’s may then be cliloro series, and the trichloro compounds being apcxplainetl by thc assumption that addition of the proximately 3.1 units less favorable than the dithird chloritic atorii tlccrcases the c:irbon-hydrogen cliloro coinpounds. These alternating terms, sul i ( > I i d stretigth without a corresponding increase, perimposed upon the values for the activation enaiitl probably with ;I decrease, in the carbon-chlo- ergies, make the reactivities of the series increase rine bond strength isce belowj. regularly from irioriochloro to dichloro to trichloro, A test of this tentative explanation is available a t least a t the temperatures studied. in thc work of vaii l
