2084
COMMUNICATIONS TO THE EDITOR
the National Science Foundation (Grant GP-1941) for financial support. S B PESICK&COMPANY NEXVYORK8, SEW YORK DEPARTMENT OF CHEMISTRY THEPENSSYLVANIA STATE USIVERSITY UNIVERSITY PARK,PENNSYLVANIA RECEIVED MARCH26, 1964
ERIC SMITH ROBERTS JARET MAURICESHAMMA ROBERTJ SHINE
Nuclear Hydrogen-Deuterium Exchange in Resorci;.ols and Related Compounds in Weakly Alkaline Solution
Sir: Under forcing alkaline conditions all hydrogen atoms of the phenoxide anicn can be replaced by deuterium, l , * while under somewhat less stringent conditions only the ortho and para hydrogens exchange. In either case the reaction is slow. LVe have found that the presence of a second hydroxy group, when oriented meta, greatly enhances the rate of exchange; mild reaction conditions then suffice for complete ( i e . , equilibrium) exchange of the ortho and para hydrogens. The exchange was monitored by n.m.r. spectroscopy in deuterium oxide solutions of approximate p H 9. In some cases, after the exchange was complete, the sample was diluted with water and the reappearance of the peak(s) o b s e r ~ e d . ~The spectra of phenol, catechol, and hydroquinone in alkaline deuterium oxide did not change duIing several hours and were the same as their spectra in water. In contrast, the exchange of protons 2, 4, and 6 of resorcinol reached equilibrium in 1 hr. a t p H ca. 8 and in 10 min. a t p H ca. 11. The hydrogens in phloroglucinol also exchanged rapidly. Other compounds, listed below, showed the following behavior: (1) the combined activating effects of two or more metu-oriented OH groups sufficed for ready exchange of protons ortho and para to them (I-XI); (2) protons of aromatic rings containing only one OH group (I-VI) or two adjacent OH groups (VI1 and VIII) did not exchange; ( 3 ) the OR of the pyrone ring meta to an OH group did not serve in lieu of the second OH (none of the protons in XI1 exchanged) ; (4) the various protons CY to a ketone (1-117, IX-XI)exchanged, b u t much more slowly than the aromatic ones (see below, however). The exchangeable nuclear protons in compounds 111-17, VII, and VIII are nonequivalent and the proton absorbing a t higher field exchanged more r a p i d l ~ . ~ In 4-acetylresorcinol the exchangeable protons can be distinguished from each other since the signal due to H-G exhibits the usual ortho coupling of -9 C.P.S. The proton flanked by the two OH groups (H-2) exchanged readily, b u t H-G was replaced rapidly only on heating; the methyl ketone protons exchanged more rapidly ( 1 ) .2 I Shatenshtein a n d A . V l.edeneev, Z h Obshch K h i m . . 28, 2 6 1 1 ( l U . * A ; ( ‘ h r m . A b s l r . , 6 3 , 5836 ( l g 5 9 ) . ( 2 ) G . E. Hall, E , M L i b b y , a n d E.I-, J a m e s , J . O v g . C h e m . , 28, 311
A . P Best a n d C . 1- Wilson, J . Cham. Sot, 28 (1Y38), C . K . Ingold. C G Raisin. a n d C . I,, Wilson, ibid , 1637 ( 1 9 3 6 ) ; cf,, however, P. A . Small a n d J H Wolfenden, t b t d . 1811 (1936). ( 4 ) F o r c o m p a r i s o n . s p e c t r a of most c o m p o u n d s were determined in w a t e r a t t h e s a m e p H . T h e specti-a varied according t o t h e a m o u n t of anion p r e s e n t : signals d u e t o protons ortho t o a n OH group wei-e always shifted upfield a i more of t h e anion was f o r m e d , while signals d u e t o t h e m r l a p r o t o n s were either unaffected 01-qhifted downfield. ( 5 ) In nxringrnin ( I V r the chemical s h i f t s of these protons coincided so t h a t a difference i n t h e exchange could n o t be aqcertained Catechin ( V I I I ) decomtxjied a t higher p H , b u t could be recovered from solution a t p H 8 in w h i c h it- exchange w a s very f a s t
Vol. 86
RofloH How OR‘ 0
\
Ho#oH OH 0
1v
I, 11, I11
H
O
W
\
moH
:
\
R’
OH
I
OH
OH 0
v, VI, VI1
VI11 OCH3 “
\O
0 f
l
OH 0 IX,
x, X I
XI1
I , phloretin, R = R ’ = H. 11, naringin dihydrochalcone, R ’ = H. 111, phloridzin, R = 2-O-a-~-rhamnosyl-p-~-glucosyl; ~ . naringenili. V, apigenin, R = H ; K ’ = @ - D - ~ ~ u c o s YIV, R = It’ = K ” = H . V I , vitexin, R = C-p-D-glLltOSyl; R’ = R ” = H . V I I , quercitrin, R = H ; K’ = a-L-rhamnosyloxy; R ” = OH. V I I I , &catechin. I X , K = H. S ,K = 2-0-a-Lrhamnosyl-a-D-glucosyloxy. XI, K = OH. X I I , pratol.
than H-6, but much more slowly than H-2. The slowness of H-6 exchange becomes understandable if delocalization of thc negative charge in the intermediate XI11 is greatly enhanced by the acetyl group. Formation of an sp3center a t C-G precludes such interactions.
0 0-
XI11
In 4-carboxyresorcinol H-2 also exchanged faster than H-6, and in both compounds the peak due to H-2 was slightly upfield. However, in 4-chlororesorcinol, H-6 exchanged faster and its absorption was slightly upfield; and in 5carboxyresorcinol H-4 and H-G exchanged faster than H-2, although H-2 absorbed a t higher field. Thus, no general direct relationship exists between the effect of substituents on the chemical shift of a particular proton and their effect on the rate of exchange. Several naphthalenediols were examined to ascertain whether all protons that can, in principle, be activated by both OH groups do exchange. As expected, no exchange occurred in 1,?i-naphthalenediol. IYhile H-1 and H-S in 2,i-naphthalenediol were rapidly replaced b y deuterium, the other protons were unaffected even on heating.€ I n l,6-naphthalenediol H-2, H-4, and H-3 exchanged.’ A path for exchange is via ketone-enolate (6) Exchange occurred also in 2-naphthol. B y analogy with 2 . i - n a p h thalenediol, i t was assumed t h a t this t a k e s place a t H - 1 . R . F. W Cieciuch a n d P, C Morrison h a v e analyzed t h e splitting p a t t e r n of t h e 2-naphtholate ion a n d determined t h a t t h e exchange does occur a t H - I . W e t h a n k 111Cieciuch f o r permission t o cite these results in a d v a n c e of publication (7) T h e positions of exchange were deduced a s follows. after exchange t h e s p e c t r u m showed a tl-pica1 4 B system ( J 9 c.y s ) as well as a singlet Since t h e low-field p a r t of t h e A B system was also a p p a r e n t prior t o exchange ( t h e high-field p a r t ovei-lapped with t h e complex multiplet d u e t n t h e o t h e r protons except H-2 w.hich appeal-ed as a q u a r t e t at higher field) a n d since H-E is t h e only pl-oton in t h e nondeuterated compound t h a t would ordinarily be expected t o couple with only o n e p r o t o n , i e , , H - 7 , t h e A B s y s t e m i n t h e s p e c t r u m of t h e d e u t e r a t e d compound m u s t he d u e t o H - 7 a n d H - 8 The singlet is assigned t o t h e i n r t u hydrogen. H - 3
-
COMMUNICATIONS TO THE EDITOR
May 20, 1964
2085
The formolysis of cis-5,9-decadienyl p-nitrobenzenesulfonate (V) has now been examined. If this reaction, as well as that of the trans isomer, were t o proceed by a stepwise mechanism involving the intermediacy of a common cation like IV, both isomeric substrates would yield the same cyclic products. 5,9-Decadiyno12 was selectively hydrogenated over Lindlar catalyst t o yield cis-5,9-decadienol which was purified by preparative vapor phase chromatography, 00nZoD1.4637. Anal Found: C, 77.9; H, 11.8. The p-nitrobenzenesulfonate V, m.p. 2 6 . 5 2 7' (Anal. Found: C, 56.5; H 6.4; N, 4.4), was solvolyzed by heating a 0.02 *l4solution in anhydrous formic acid xv containing pyridine (0.04 M ) for 1 hr. a t 75'. The forXIV mate esters were reductively cleaved by treatment with Regardless of mechanism, the exchange described lithium aluminum hydride, and the resulting product above should be useful in structural elucidations of comwas analyzed by vapor phase chromatography.2 The plex phenols and is, of course, suitable for the preparaproducts were identified by peak enhancement experition of various deuterated aromatic compounds. ments and by infrared spectral comparison with (8) E. S. H a n d a n d R. M . Horowitz, unpublished results. I t is of inauthentic specimens.2 Since the solvolysis of the trans terest t h a t H-2 a n d H - 4 of 1,3-naphthalenediol exchange rapidly in deut e r i u m oxide solution even in t h e absence of base. sulfonate ester had not been carried out previously (9) T h e f a c t t h a t 5 - n i t r o - I - n a p h t h o l h a s a p K , n o t very different from under these conditions, it was re-examined for direct t h a t of I - n a p h t h o l itself bas been rationalized in similar t e r m s : K . C. comparison with the cis isomer. The relative proporSchreiber a n d A l . C K e n n e d y , J . A m . C h e m . Soc., 7 8 , I 5 3 (1956). (IO) D e p a r t m e n t o f C h e m i s t r y , University of P i t t s b u r g h , P i t t s b u r g h , tions of alcohols from the trans sulfonate ester were as Pennsylvania 15213. follows: 3YG of 1-(A3-butenyl)cyclohexanol, 10% of A3FRUITASD VEGETABLECHEMISTRY ELLISMAKULA HAND~O butenylcyclopentylcarbinol, 57yo of trans-2-(A3-buLABORATORY ROBERT M. HOROWITZ teny1)cyclohexanol (II), 8% of trans-5,9-decadienol, WESTERSUTILIZATION RESEARCH 5% of trans-anti-2-decalol, 14% of trans-syn-2-decalo1, A N D DEVELOPMEST DIVISION and 3% total of several unidentified components. XCRICULTUR.4L RESEARCH SERVICE U. S. DEPARTMENT OF AGRICULTURE There was no detectable amount of the cis monocyclic PASADESA, CALIFORSIA alcohol VI or the cis decalols VII among the products. RECEIVED MARCH 9, 1964 These results are quite comparable to those obtained with 80% formic acid.* Formolysis of the cis sulfonate ester V gave the following relative proportions of alcoCationic Cyclizations Involving Olefinic Bonds. hols: 3% of 1-(A3-butenyl)cyclohexanol,8% of A3V. Solvolysis of cis-5,9-Decadienyl butenylcyclopentylcarbinol, 56% of cis-2-(A3-butenyl)$-Nitrobenzenesulfonate cyclohexanol (VI), 16% of cis-5,9-decadienolj 13% of an Sir: inseparable mixture of cis-syn- and cis-anti-2-decalo1, In the formolysis of trans-5,g-decadienyl p-nitrobenand 4a/c total of several unidentified components. zenesulfonate ( I ) , 2 the processes resulting in the forThere was no detectable amount of the trans monocyclic mation of cyclic products were highly stereoselective, alcohol 11 or of the trans decalols 111 among the products. possibly stereospecific. Thus (after cleavage of forcis-2-(A3-Butenyl)cyclohexanol (VI) was identified by mate esters) trans-2-(A3-buteny1)cyclohexanol(11) was comparison with an authentic specimen, n Z o D 1.4770 formed to the exclusion of the cis isomer, while the de(Anal. Found: C, 77.8; H, 11.75)) which was precalols were exclusively those with trans-fused rings pared by reduction of the corresponding ketone with (111). I t was not possible to decide whether this lithium tri-t-butoxyaluminum hydride followed by prestereochemical consequence was the result of a synparative vapor phase chromatographic separation from chronous process or of a stereoselective equatorial atthe predominant trans isomer 11. tack of nucleophile on a cationic intermediate like IV.3 \Ye now wish to present evidence which supports the former mechanism.
intermediates. That such intermediates are feasible is shown by the fact that the anion of 1,3-naphthalenediol exists entirely as a ketone-enolate.* In intermediates postulated for the exchangeable protons, the aromaticity of one ring is retained in a t least one of the resonance hybrids (e.g., X I V ) ; the failure of the other ortho protons t o exchange can be attributed to loss of the benzenoid character of both rings (e.g., XV).9
p?T OSOzCsH4N02 I
--
_-
&-OH H I11
IV
(1) P a p e r I\' of t h i s series: W. S. J o h n s o n , W. H. L u n n , a n d K . Fitzi, J A m . C h e m . Soc , 8 6 , 1972 (1964). (2) W. S . Johnson, D . M. Bailey, R. Owyang, R . A. Bell, B. Jacques, and 5. K . C r a n d a l l , ibid.,86, 1959 (1964). (3) W. S. Johnson, S . L. G r a y , J. K . C r a n d a l l , a n d D. M . Bailey, ibid.,
86, 1966 (1964)
Since the cyclizations of I and V proceed stereochemically in exactly the opposite sense, a common cationic intermediate IV cannot possibly be involved. If the solvolyses of both I and V proceed by the same basic mechanism-an assumption which is reasonable, particularly in view of the strikingly similar type of product distribution-then it follows that the cyclizations must either be concerted processes or involve cationic intermediates (e.g., bridged carbonium ions) which retain the stereochemical integrity of the respective substrates. A decision between these latter two