c. BROWN . W D hf. A-AKAGAW.4
3614
Vol. ii
HERBERT [CONTRIBUTIOK FROM
THE
DEPARTMENT O F CHEMISTRY O F
PURDUE
UNIVERSITY]
Steric Effects in Elimination Reactions. 111. The Effect of the Steric Requirements of Alkyl Substituents upon the Extent and Direction of Unimolecular Elimination in the Solvolysis of Secondary Alkyl Brosylates BY HERBERT c. BROWNAND hf.
NAK4GAWX1 RECEIVEDOCTOBER 21, 1954
The solvolysis of the group of secondary alkyl brosylatcs, RCH2CH(OBs)CH3,with R = Ale, Et, i-I'r, t-Bu, mas studied in order t o obtain data t o compare the effect of the steric requirements of the group R in the secondary series with data previously obtained for the related tertiary series, RCH2CBr(CH:3 10. On the basis of the steric strain interpretation, it would be anticipated t h a t the increasing steric requirements of the group R should be much less in the secondary series than in the related tertiary in increasing (1) the rate of acetolysis of the brosylate and (2) the ratio of 1-t o 2-olefin in the product. The observed rate constants for the acetolysis a t 70" do indeed exhibit a relatively minor dependence upon the steric requirements of R : Me, 1.29 X lO-'sec.-'; E t , 1.23 X i-Pr, 1.13 X t-Bu, 2.64 X Moreover, the ratios of 1- to 2-olefin in the products also exhibit only a relatively minor dependence upon the steric requirements of R : Me, 0.11; Et, 0.19; i-Pr, 0.25; t-Bu, 0.32. These results are quite different from those observed in the related tertiary bromides and are in accord with the predicted magnitudes of steric strains in the two series. The yield of olefin does vary with the steric requirements of R : Et, 31.5%; GPr, 46.57,; t-Bu, 68%. This is in accord with the previous proposal that the yield of olefin is primarily dependent upon the degree of steric hindrance t o the substitution reaction of the carbonium ion and is not directly related to the degree of steric strain in the tetrahedral substitution product. Isolation and examination of the acetates formed in the solvolysis reveals that within the accuracy of the infrared analytical procedure, a maximum of 1-2'35 of rearrangement occurs in the solvolysis of 2-pentyl, 4-methyl-2-pentyI and 4,4-dimethyl-2-pentyI brosylates.
Since the acetolysis of arylsulionic acid esters has The marked changes observed in the rates of hydrolysis of the tertiary bromides, RCHICbr(CH3)2, been studied thoroughly as a means of producing with increasing steric requirements of R, were a t - secondary carbonium ions, 43 it was decided to tributed t o increasing steric strains in the parent utilize this reaction for the secondary series, Origcompound.* Likewise, the marked shift in the inally it was planned to work with the toluenesulratio of 1- to 2-olefin in the product, resulting in fonates. However, the observation that 2-butyl Hofmann-type elimination in the case of R = t-Bu, and 2-pentyl tosylates were liquids which could not was also attributed to steric These be crystallized led us to shift to the p-bromobenstrains arise from the steric interaction of cis R zenesulfonates (brosylates). Here the 2-butyl and and methyl groups in the 2-olefin, leading to an in- 4,4-dirnethyl-2-pentyl brosylates were solids which could be purified by crystallization, but the %pentyl creased relative stability of the 1-olefin. The study of the corresponding series of second- and 4-methyl-%pentyl brosylates proved to be ary derivatives, RCH2CHXCHR,offers a test of the liquids and were therefore utilized as such. proposed interpretation. If this interpretation is Results valid, the much lower steric crowding in the secondThe rates of solvolysis of the four brosylates, ary derivative should result in much smaller steric assistance by R. Consequently, the rate of solvoly- RCH2CH(OBs)CH3withR = Ale, Et, i-Pr and fsis should show a much smaller dependence upon E u , were run a t 70' in anhydrous acetic acid by the the steric requirements of R than in the case of the procedure previously described by Winstein and his c o - ~ o r k e r s . ~ The results are summarized in related tertiary derivatives. I n the cis-2-olefins the steric effect of R should be Table I. TABLE I the same as in the corresponding olefins from the tertiary bromides. FIRST-ORDER RATE COSST.4NTS FOR THE S O L V O L Y S I S OF ALKYLMETHYLCARBINYL BROSYLATES IS ASHYDROUS ACETIC ACID4r 70.0"
H, R
2-Butyl 2-Pen ty 1 4-Methyl-2-peiityl l,-l-Dimethyl-2-peiityI
XH3
)'="(,
Consequently, on the basis of the steric strain theory, there should be observed with increasing steric requirements of R only a relatively slight shift iii the ratio of 1- to ?--oleliii, with a much more iiiiportaiit shift in the ratio of cis to trtzns iii the 2-olefiii. (1) Post-doctorate assistant a t Purdue IJniversity. 1953-1!464 in part on a contract supported by the Otfice of Naval Research and iu part on a research grant supported b y the National Science Founda-
tion. ( 2 ) H. C . Brown and hl. Nakagawa. THIS Jorrnxhr,, 7 7 , 7tiIO (1955). (3) H. C. Brown and I. hforitani, i b X , 77, 3007 (l!)::).
k , X 10-4
Secondary brosylate, R C H i C H (OBs)CHI
However, the formation of the trans-2-olefin provides a mechanism for avoiding the strain while retaining the greater hyperconjugative stabilization of the 2-olefin. a
S. Winstein, B . R. Morse,
Course, THIS JOURSAL,
K
(sec. -1)
1 l C
1.29" I 23 113 2 64
Et i-Pr f-Bu I(.
C. Schreiber aud J .
74, 1113 (1952). report 1.23 X
set.-'.
The acetolysis esperinients for the examination o f the olefin products were carried out a t 118' in thc presenre of a slight excess of potassium acetate to neutralize the p-broinobrrizenesulfouic acid and to prevent isomerization uf the cMin. The yieltl of olefin was measured carefully :uid the product w:ts (4) S . Winstein, E . Grunmald and 1.. I,. Ingraham, ibid , 70, 821 (1918). (5) J I1 Robcrts, I\' B e n n e t t , R 1;. 1IclIaholl a n d IS. M' l l o l r u y i i , J r , ibi'i., 74, 1283 (1952).
July 5 , 1953
SOLVOLYSISOF Sec-ALKYL
BROSYLATES : UNIMOLECULAR ELIMINATION
3615
with 8% rearrangementa5 It therefore appeared of interest to isolate the acetates formed in the solvolysis of the brosylates of the present investigation in TABLE I1 order to ascertain the extent of rearrangement acYIELDAND COMPOSITION OF OLEFINS FORMED IN THE SOLVOLYSIS OF ALKYLMETHYLCARBINYL BROSYLATEIN AN- companying the acetolysis. The results are summarized in Table 111. From the physical properties HYDROUS ACETICACIDAT 70" and the infrared examination of the products (Figs. Composition Secondary brosylate, Yield, frans- Ratio 1 and 2) it appears that the acetolyses of the three RCHEH(0Bs)CHs R % 1cis-221-/2secondary brosylates proceed without significant 2-Butyl tosylate" Me 1 0 . 3 4 3 . 2 4 6 . 5 0 . 1 1 rearrangement.
analyzed by infrared. in Table 11.
The results are summarized
2-Pentyl tosylatea 2-Pentyl brosylate 4-Methyl-2-pentyl brosylate 4,4-Dimethyl-2pentyl brosylate a Ref. 3.
Et Et
25 31.5
16 35 49 15.7 3 5 . 5 4 8 . 8
.19 .19
i-Pr
46.5
1 9 . 7 2 7 . 3 53.0
.25
I00
280
$60
t-Bu
68
24.3
0.9 74.8
.32
5 40 e
20
2-Butyl brosylate was not run. However, data are available for 2-butyl tosylate.a The similarity in the results for the 2-pentyl brosylate and tosylate lends confidence in the use of the data for 2-butyl tosylate for the comparison of the effects of the structural changes on the extent and direction of olefin formation. It has been reported recently that the solvolysis of radiocarbon tagged 2-butyl tosylate proceeds I00 I-
I
c
I
100
Wavelenpth, mp.
Fig. 2.-Infrared spectra: A, acetolysis product of 4,4dimethyl-2-pentyl brosylate; E, acetolysis product of 4-methyl-2-pentyl brosylate; C, acetolysis product of 2-pentyl brosylate; D, synthetic mixture of 98,3y0 2pentyl and 1.7yc3-pentyl acetate.
The presence of 1-27, of 3-pentyl acetate in the 2-pentyl derivative can be detected readily by the characteristic absorption peak a t 11.3 p (Fig. 2, D). From the infrared spectra of the recovered acetate (Fig. 2, C) it appears t h a t any rearrangement under the reaction conditions used for the acetolysis must be relatively small, certainly no greater than 1-2y0 as a maximum.
Discussion Wavelength. m u
Fig. 1.-Infrared spectra: A, 4,4-dirnethyl-Z-pentyl acetate; B, 4-methyl-2-pentyl acetate; C, 2-pentyl acetate; D, 3-pentyl acetate.
The rate of solvolysis decreases slightly from R = Me to E t to i-Pr with an increase a t t-Bu. The total increase in rate from 2-butyl brosylate to 4,4dimethyl-2-pentyl brosylate (from R = Me to t -
HERBEKT C. BROWNAND AI. NAKAGAWA
3ti16
Vol. 77
TABLE I11 DATA O S TIIE Y I E L D AND PHYSICAL PROPERTIES O F ACETATES FORMED I N THE SOLVOLYSIS O R TIIE
SECONDARY ALKYLBROSY-
LATES Secondary alkyl brosylate
RCH&H(OBs)CIIn F. mole
2-Pentyl brosylate 4-Methyl-2-pentyl brosylate 4,4-Dimethyl-2-pentyl brosylate See Table 11. for physical
92.1 06 2
Main fractiona
Forerun p. nZb
0 3
0 7
,8
. 3
F.
1.3964 1.4012
19.3 15.4
Purity, b
nZ0D
B.p., OC. (mm.)
%
%
1.3963 1.4005
131-131.5(741) 141-144(738)
50 86
98.7 100
12
103
61 . 0 .18 .3 1.4062 3.44 1.4064 76.3-77 (50) properties of synthetic samples. * Infrared analysis. See Table VII.
Bu) is only a factor of 2 . This compares with the corresponding factor of 12 for the same structural change in the tertiary bromide series2 The much greater effect of the steric requirements of the group R in the tertiary as compared to the secondary series is made evident in Fig. 3. The results are evidently in accord with the predicted effect of the increasing steric requirements of R on the relative steric strains in the secondary and tertiary systems and the postulated effect of these strains in facilitating ionization.6
Yield,
It is significant that the ratios of cis- t o Lrurcs-2olefin in the secondary series exhibit a regular decrease with the increasing steric requirements of li. Indeed, in the case of -l,-l-dimethyl-2-pentyl brosylate (R = t-Bu) the analysis indicates the presence of but 0.9% of cis-2-olefin, with 75% of the trans 2derivative present. The cis-2-olefin, as has beeii pointed out, would have a t-butyl and methyl group in a highly crowded steric relationship of the sdme type previously postulated to account for the existence of Hofmann-type elimination in dimethylneoperi tylcarbinyl bromide.
P
I
0 .e
R- CH?
6- C H ~ 0BS
R=
Et i-Pr L-Bu Fig. 3 - - h cotnparizoii of the cffcct of Ii 011 the rellttivc solvolysiz rates in the sccondary and tertiary alkyl serici, RCHzCH(0Bs)CHa and RCHzCBr(CH3)g. Me
In the tertiary series there was observed an increase in the ratio of 1- to 2-olefin with increasing steric requirements of R. The ratio changed from 0.27 for R = Me t u 4.26 for R = t-Bu. This corresponds to Ilofmarin elimination for R = t-Bu. In the secondary system the shift in the ratio of 1to 2-olefin is much smaller, from 0.11 for R = Me to 0.32 for K = t-Ru (Fig. 4). Thus all eliminations in the secondary series proceed in accordance with the Saytzcff rule. , :is.> (1941;). (Ci) 11. C. I i i o n i i , . > c ~ c , > t r103,
1;ig. 4.----A coiiipzri\oii of thc e f f e c t of Li 011 the ratio o1 1t o 2-olcfili in the solvolysis of the sccoldary and tcrt itiry alkyl scriei, RCH*CH(OBs)CH3 and RCH2CBr(CEIs)..
-1gain these results are in complete agreement with the predicted effect of the steric requirements of the group R on the relative stabilities of the 1and %olefins, as well as upon the effect of the group on the stabilities of the cis- and trczns-2olefins. Thus both the rate of solvolysis data and the results on the direction of elimination are entirely consistent with the theory of steric strain. In the tertiary series the rate of hvdrolysis dat:i show a very small decrease from K = Me to I< = Et, followed by an increase with R = LPr, and a very largc iiicrease with R = t-Bu. Ignoring for the imseiit the sniull decrease ol,servccl with I< =
July 5, 1955
SOLVOLYSIS OF
Sec-ALKYL BROSYLATESUNIMOLECULAR ELIMINATION
3617
Et, we can say that the changes are in accord with are unable to account for the difference in results in the predicted increases in steric strain in the ter- the 2-butyl and 2-pentyl systems and are continutiary alkyl bromide. Only small increases should ing our studies in an attempt to resolve the differresult from the replacement of one methyl group by ence in results.' an ethyl or isopropyl. The geometries of these Experimental Part groups are such that they can minimize the steric Materials.-2-PentanoI was prepared in 82% yield by the strain by rotation to a position of minimum comreaction of 4 moles of methylmagnesium iodide on 3.6 moles pression. The t-butyl group, however, possesses of n-butyraldehyde (b.p. 73-73.5" (735 mm.), nZoD 1.3792). spherical symmetry and rotation cannot serve to 3-Pentanol was prepared in 55y0 yield by the reaction of alter the strain significantly. Hence, the t-butyl 1.5 moles of ethylmagnesium bromide with 0.75 niole of group causes a sudden large increase in the rate of ethyl formate (b.p. 53-54' (746), XZOD 1.3541). 4-Methyl-2-pentanol was synthesized in 84% ?.icld by tlic solvolysis. sodium borohydride reduction in methanol of 3 moles of In the secondary series the same behavior is evi- methyl isobutyl ketone (b.p. 114-115' (745 mm.), nZoD dent. There are small decreases in the rate of sol- 1.3954). 4,4-Dimethyl-2-pentanone ( b .p . 123-125' (743 m m .), volysis in going from Me to E t and i-Pr, with an in%*OD 1.4038) was prepared in 41 To yield by the oxidation of crease by a factor of 2 observed with R = t-Bu. 3.0 moles of diisobutylene (b.p. 99.5-102" (74.5 mm.), Here also it is believed that the t-Bu group results in X ~ O D 1.4090) with potassium dichromate (4 moles) and sulthe presence of strain which facilitates the ioniza- furic acid (16 moles).8 4,4-Dimethyl-2-pentanol was synthesized by the reduction, However, because of the less crowded condiof the ketone with sodium borohydride. A mixture of tion of the secondary derivative, the effect of the t- tion 100 ml. of ethanol and 4,4-dimethyl-2-pentanone, 144 g., butyl group should be considerably smaller than in 1.26 moles, was added dropwise to a stirred solution of 14.2 the tertiary derivative, in agreement with observa- g. (0.375 mole) in 140 ml. of ethanol. After addition was complete, the reaction mixture was heated under reflux for tion. hr. A solution of 30 g. of sodium hydroxide in 100 ml. of The t-butyl group does not result in a sudden 2water then \vas added to hydrolyze the intermediate and large increase in the 1- to 2-olefin ratio in the sec- heating under reflux continued for one hour. The cold reacondary series. However, in these compounds there tion mixture was poured into ice-water and the organic fracis available an alternative route for the relief of tions were washed with cold water and dried over anhydrous potassium carbonate. The product was distilled through a strain-the formation of the trans 2-olefin-and modified Widmer column and material of constant refracthe carbonium ion elects this route for elimination. tive index collected. The yield was 123 g., 85%. The total yield of olefin in the solvolysis of the The physical properties of the alcohols are listed in Tablc compounds RCEr(CH& increases fairly regularly IV. The acetates of these carbinols were prepared to supply with the increasing bulk of the group R (R = standards for the infrared analysis of the acetates formed in Me < E t < i-Pr < t-Bu). The failure to observe the solvolmis. The acetates were prepared by treating the any sudden large increase in yield with R = t-Bu sug- alcohol with acetic anhydride in pyridine solution and the gests that an increase in the steric requirements of acetates were fractionated until b.p. and n*OD exhibited the group R is of much smaller consequence in the satisfactory constancy. carbonium ion RC+(CH& than it is in the tertiary IV.The physical constants of the acetates are listed in Table halide itself. The conclusion that the steric rePreparation of the p-Bromosulfonates .-Commercial pquirements of the group R exert a relatively small bromobenzenesulfonyl chloride was dissolved in benzcne, effect in the carbonium ion itself, suggests that the the solution Trashed with aqueous sodium carbonate, water dried. The chloride, precipitated with ligroin at OD, large effect of R on the olefin yield cannot be pri- and exhibited 1n.p. 77-78'. The equivalent amount of purified marilv an effect of R on the rate of the elimination p-broinobenzenesulfonyl chloride was added in one portion reaction. Instead, it would appear that the in- to a solution of 0.1 mole of carbinol in 100 ml. of anhydrous crease in olefin yield in this series must be due pri- purified pyridine cooled in an ice-salt-bath. After standing overnight in a refrigerator, the reaction mixture was added marily to a decreased rate of substitution, ks, arising dropwise with stirring to 200 ml. of strongly cooled 6 A: from resistance of the planar carbonium ion to the hydrochloric acid. The acidic aqueous layer was separated addition of a solvent molecule with the consequent from the heavy oily layer or crystalline mass and n:as extracted with chloroform. The chloroform eytract was comformation of a strained tetrahedral arrangement. bined with the crude brosylate and the solution waqhed sucThis conclusion is supported by the data in the cessively with water, sodium bicarbonate solution and water. secondary series. Here also the t-butyl group fails After drying, the solvent was removed a t 15 mm. The liquid t o exhibit any large effect. Instead, just as in the brosylates were evacuated to 0.01 mm. for 8 to 9 hr. a t tertiary series, the olefin yield increases fairly regu- room temperature to rernqve volatile impurities. The solid brosylates were recrystallized from petroleum ether (b.p. 35larly in the series R = Me < E t < i-Pr < t-Bu. 65'). I n conclusion, the results of these studies are conThe yields were 85-89y0 except in the case of the 4,4sistent with the interpretation that large bulky dimethyl-2-pentanol, where the yield waq only 29%. This groups increase the rate of solvolysis through re- brosylate showed evidence of instability, especially in the lease of steric strain (steric assistance) and they (7) It should be mentioned t h a t our experiments were carried out favor a shift from Saytzefl to Hofmann elimination with anhydrous acetic acid which was 0.7 M in potassium acetate, the solvolysis of t h e 2-butyl-l-C*{ p-toluenesulfonate (ref. 5 ) because of the steric strains in the 2-olefin. Large whereas was carried out in acetic acid which was 0.46 A4 in potassium acetate bulky groups also favor an increase in the k E l k s ra- and 0.45 AJ in potassium p-toluenesulfonate. I t is therefore possible tio, primarily through a decrease in ks because of t h a t the difference in results may have its origin in either t h e difference in reaction conditions o r in t h e use of t h e brosylate instead of the steric hindrance to the substitution reaction. tosylate. We are currently examining t h e solvolysis of 2-pentyl t o s y Our failure to observe more than 1-2% rearrange- ate under the identical conditions of Roberts a n d co-workers (ref. 5 ) ment in the acetolvsis of the secondary alkyl brosyl- in order t o ascertain whether these factors are responsible for t h e of rearrangement. ates is puzzling in view of the Syo rearrangement extent ( 8 ) W. A. lhlosher and J. C. COY, Jr., 1'11rsJ L I I T R N ~ I .72, , 3702 reported for the acetolvsi.; of 2-butvl tosylate.c LiTe (1!450), T h e w a i t h o r s rrport h . p . 124-125' i i l i 0 rnm ) , n i p i I I-4018
3618
Vol. 77
HERBERT C. BROWNAND M. NAKAGAWA TABLE IV PHYSICAL
PROPERTIES O F M.4TERIALS
Observed constants B.p., O C . (mm).
Compound
1,iterature c o n c t x n t s n . p . , " C . (mm.j iafb
nzb
2-Butaliol 2-Pentanol 3-Pentanol
98-99 (747'1 118-118.5 (74% 113 5-133.7 (7381
I . 3967 1.4060 1,4097
4-Methyl-2-pen tan01 l,l-DirnethyI-%-pentanol
129-129.5 (737) 137-137.5(736)
1 4110 1 4180
2--Pentylacetate 3-Pentyl acetate
131-131 ,5 (741 I 1.'10-130.5 (74;;)
1 39W 1 ,3980
99 (758) 120. 1 (74.3 116 1 1 5 . 4 (7.54) 131.4(745) 1 3 5 . 4 (7421 137-137 3 (746 i 1:31 8 - 1 3 (7-lii I 132.5-133 (7-48) 13 1 147- 1-18
1.3954 1.4068 ( 1 ,4037'1 (1.4077 ) 1 ,4132 I 4241 1 ,4188 1 3RijO 1 4OO.F 1 . :300(;
Ref D
ii
L h
i'
I Q h
l-Methy1-2-pentyl acetate 143-143.5(740) 1 ,400,j 4,4-Dimethy1-2-pentyl acetate 76-76.5(47) 1.4066 2-Butyl brosylate 1I.p. 30-32 xp. :3i-z 1.5313 2-Pentyl brosylate 4-Methyl-2-pentyl brosylate 1.5260 4,4-Dimethyl-Z-pentyl brosylate M.P. 52-53 R. H. Pickard and J. Keiiyon, J . Chem. SOC.,103, 1935 (1913). * R. C. Huston and H. E. Tiefenthal, .l. O r g . Chem., 16, R. F. Brunnel, ;bid., 673 (1951). F. H . Norton and H. B. H a s , THIS JOURNAL, 5 8 , 2147 (1936); refractive index n2%. 45, 1334 (1923); refractive index ~ 2 5 ~ e .F. C. U-hitinore and A. H. Homeyer, ibid., 5 5 , 4194 (1933). f II. E. French and M. Guerbet, Conzpt. r o z d . , 149, G . G . lvrightsman, ibid., 60, 50 (1938). Q J . Kenyon, J . Chem. Soc., 105, 2226 (1914). 1.29 (1909). S. LVinstein, et ai., THISJOURNAL, 74, 1119 (1952). 2-Butyl and 4,4-dimetli~l-2-perityl brosylates, both solids, vielded 99% of the calculated acid. The remaining tn-o RATE DATA F O R THE O F ~ , ~ - I ~ I W E l F I Y L - 2 brosylates, liquids, ?-ielded 95% of the calculated acid. PENTYL B R O S Y L A T E ~ AT 70.0' Physical properties for the bras>-lates are wn-narized in Table IV. Time, Sodium ki X &y4, sec. acetate, ml. x (1 - x sec. Rates of Acetolysis of the Brosy1ates.-The technique used was similar to that previously described.Q Rate data 0 0 590 for a reprecentative kinetic run on 4,4-diinethyl-Z-pentyl 1200 1 ,240 1 .;ti0 2 6% 0 650 hrosylate are renorted in Table V. The mean value of 2400 1 . 720 1 . 2so 2 64 1 130 k l is (2.65 0.02) X 10-4 see.-'. A duplicate run yielded :?I600 2 080 0.960 '7 67 1 490 kl = (2.63 =k 0.02) X 10-4 see.-'. The data are representative of the precision realized in all of the deterniinationr. 4800 2.320 1.730 680 2 64 Elimination Experiments.-The following general proce6000 7 5 10 1 920 490 2 66 dure was utilized. One-tenth mole of the brosylate was 7200 '7 640 '7 030 360 2 64 placed in a r?und-battomed flask with 0.11 mole of anhy'l Anh?-drous acetic acid, 0.02 S in acetic mlivdride. drous potassiu'n acetate and 150 ml. of anhydrous acetic 0.5071 g. of hrosylate in 50.0 ml. of solvent: 0.08027 -U; acid (0.02 S in acetic ati?lycIride). The mixture \?-as a t tached to a Todd micro colu.nt1 and the olrfin >vas distilled calcd. infinit:- titer was 3.033 nil. as compared to 3.000 ml. out of the reaction mixture into a coc;lett receiver. The diqo b 4 . (99%). V ACETOLYSIS" TABLE
+
,
July 5 , 1955
OF UNIMOLECULAR ELIMINATION EXTENTAND DIRECTION
3619
The residual crude acetate was distilled through a short distillation head. The small amount of low boiling fraction was placed in a small flask, cooled to Oo, and evacuated to remove any residual ether. The data on the yield, physical properties and composition of the acetates are summarized in Table 111. The identity of the acetates isolated from the acetolysis of the brosylates v, ith those synthesized from the carbinols is evident from the infrared spectra (Figs. 1 and 2). Comparison of the spectra of the acetates isolated from the solvolysis of 4,4-dirnethyl-2-pentyl brosylate (Fig. 2,A) and 4methyl-2-pentyl brosylate (Fig. 2,B) with the spectra of synthetic samples of the acetates (Figs. l , A and 1,B, respectively) indicated that the products were essentially pure, containing no more than very minor amounts of the rearranged acetate (Table VII). The availability of both 2- and 3-pentyl acetate permitted a more detailed analysis for the rearranged product. Blank experiments revealed that it was possible to determine small amounts of the 3-pentyl acetate in the 2-derivative to within *0.5%. Typical analyses of this kind are reported in Table VII. Application of this procedure to the acetate obtained in the solvolysis of 2-pentyl acetate indicated the presence of a maximum of 1-2% of 3-pentyl acetate in the product (Table V I I ) .
TABLE VI1 INFRARED AXALYSIS OF ACETATESPRODUCED IN ACETOLYSISOF SECONDARY BROSYLATE
Acknowledgment.-The generous assistance of Dr. Kenneth LV. Greenlee and the American Institute Project No. 45 a t The Ohio State University in supplying samples of the pure olefins is
gratefully acknowledged. We also wish to express our appreciation of the assistance rendered by Mr. Y . Okamoto with several of the infrared analyses.
[CONTRIBUTION FROM THE
Acetate unknown
4,4-Dimethyl2-pentyl“ 4-Methyl-2pentyl”
Solution Acetate, g. CS2, g.
0.1051
Wave length,
THE
Analysis,
%
p
1.4050
8 . 9 8 105 10.63 101 ,1111 1.3152 8 . 9 0 101 9 . 5 0 101 10.67 98 2-Pentyl“ ,5005 1.2423 12.27 lOl(2-) 11.3 1.4(3-) 2-pentylb .4942(2-) 1.2465 12.27 98.7(2-) .0088(3-) 11.3 1.2(3-) 2-Pen tylC .2425(2-) 0.6469 12,27 98.2(2-) .0056(3-) 11.3 1.8(3-) Synthetic mixa Product from acetolysis of brosylate. ture: 98.3% 2- and 1.7% %pentyl acetate. Synthetic mixture: 9 i . i % 2- and 2.3% 3-pentyl acetate.
*
LAFAYETTE, INDIANA
DEPARTMENT OF
CHEMISTRY OF PURDUE UNIVERSITY]
Steric Effects in Elimination Reactions. IV. The Question of Rearrangements as a Factor in the Extent and Direction of Unimolecular Elimination BY HERBERTC. BROWNAND Y. OKAMOTO’ RECEIVED OCTOBER21, 1954 The solvolysis of 2-chloro-2,3,3-trimethylpentane in 80% aqueous ethanol forms pure 2,3,3-trimethyl-l-pentenein 58% yield. The mixture of alcohol and ethyl ether in the solvolysis is converted by hydrogen chloride into a tertiary chloride with identical properties with the original material. In 80% acetone the solvolysis proceeds to give the pure olefin (50% yield) and pure 2,3,3-trimethyl-2-pentanol (20% yield). The solvolysis of 3-chloro-2,2,3-triinethylpentane in 80% aqueous acetone results in the formatibn of pure 2,3,3-trimethyl-3-pentanol(18% yield) and olefin (58% yield). The olefin fraction contains none of the rearranged structure, 2,3,3-trimethyl-l-pentene.Treatment of the olefin fraction with hydrogen chloride converts it into 3-chloro-2,2,3-trimethylpentane.Careful examination of the products by infrared analysis reveals the absence of any measurable rearrangement within the limits of the analytical method, 1-295. It appears, therefore, that the ionization, substitution and elimination reactions of these highly branched derivatives proceed without significant rearrangement. It is concluded that there is presently no evidence that bridging and rearrangements play any significant role in affecting either the rate of ionization or the extent and direction of elimination in these and related highly branched alkyl halides.
In earlier publications it has been proposed that stable bridged intermediate in the ionization stage. the enhanced rate of ionization2 and the increased tendency toward elimination3 of highly branched . .. . .. tertiary halides resulted from the operation of steric [(CH3)aC]aCX + (CH~)~C-C--C(CH~)Z Xstrain. Evidence has been presented in the present C(C% CH3 3 group of papers in support of this i n t e r p r e t a t i ~ n . ~ . ~ However, a number of alternative proposals Similarly Hughes, Ingold and Shiner8 would prehave appeared recently, based upon the possible fer not to attribute the large quantities of olefins formation of relatively stable bridged carbonium formed in the solvolysis of highly branched tertiary ions6 Thus, i t has been suggested that the en- halides to the operation of steric strain. Instead, hanced rate of solvolysis of tri-t-butylcarbinyl de- they propose that “synartetic” effects (ie., bridgrivatives may be due to the formation of a relatively ing) are responsible for the high proportion of olefin (1) Research assistant a t Purdue University, 1952-1965, on a conobtained in the solvolysis of the tertiary chloride, t tract supported by t h e Office of Naval Research for the s t u d y of “Steric BuCClMeZ. Strains in Chemical Reactions,”
[ ‘
(2) (3) (4) (5)
H. C. Brown a n d R . S. Fletcher, THISJOURNAL, 71,1845 (1949). H. C. Brown a n d R. S. Fletcher, ibid., 7 8 , 1223 (1950). H.C.Brown a n d I. Moritani, ibid.,77, 3607 (1955). H.C.Brown and M. Nakagawa, ibid., 77,3610, 3614 (1955).
fC,) A I J S. Dewor. “The fllectronic Theory of Organic Chemiqtry,” O s i c ~ r dCniverqitr Pre ,i,I.ondotl, 1949, pp. 211-213.
H.CH2.CMe(Me).6Met
I+
+
+H+ + CHZ:CMe.CMez(Me)
(7) P. D. Bartlett. J . Chem. Education, 30, 22 (1953); Bull. sac. chim., 18, 100 (1951). ( 8 ) E. D.Hughes, C. K. Ingold and V. J. Shiner, J. C h e w . Soc., 3827 (1 Q,-,R).
