Nov. 5, 1964
SULTONE STRUCTURE FROM P.M.R. SPECTRA
a mixture of the two dimers. Crystallization of the latter fraction from methanol, using a seed of dihydrovoacaniine, yielded a n additional quantity of the latter alkaloid. Thin layer chromatography indicated t h a t the resulting mother liquors contained essentially uncontaminated dihydrovoacamidine. This material was converted to its hydrobromide salt which was crystallized from acetone; m.p. >300°. Voacamidine (48) Hydrobromide.-The mother liquors from the voacamine synthesis were purified as described above. The hydrobromide salt was crystallized from acetone, m.p. >300". Its S u j o l mull infrared spectrum was identical with t h a t of the hydrobromide salt of natural material. Artifact Experiments.-.A mixture of dreganiinol (30 m g . ) and voacangine (30 mg.) in methanol ( 3 nil.) was heated under reflux for 3 hr. and then evaporated to dryness on a steam bath. T h e
[CONTRIBUTIONFROM
THE
DEPARTMEST OF
CHE!dISTRY,
4641
residue was dissolved in ethyl acetate and extracted twice with 5' C aqueous acetic acid. Both fractions were neutralized with potassium carbonate, the aqueous phase was extracted with methylene chloride, and both organic phases were taken to dryness. Recrystallization of the ethyl acetate residue from . methanol-ether yielded pure voacangine ( 1) (infrared). Chromatography of the acetic acid residue on alumina (activity 111) yielded pure voacangine ( 1 ) (infrared) and pure degraniinol ( 3 2 ) (infrared). Thin layer chroniatography of the crude mixtures showed only voacangine and dregaminol.
Acknowledgment.-\re wish to thank Drs. U. Renner Prins, J , R . Geigy XG, Basel, for samples of and D. various alkaloids. This research was financed by the National Institutes of Health (RG 9888).
YORTHLVESTERS USIVERSITY, EVASSTOS,ILLISOIS]
Structures of Sultones from Proton Magnetic Resonance Spectra BY R . It'.. OHLINE.]X.L. ALLRED,*.&ED F. G. BORDWELL RECEIVED JUSE 2'2, 1964 Proton magnetic resonance spectroscopy has beeii used to confirm structure assignments for eighteen sultones.
T h e degradative method for structure proof for a sultone is particularly difficult since it inevitably requires isolation of the salts of sulfonic acids, and these are hard to separate and to characterize. As a result. rigorous structure assignments for most sultones have not been made. T h e present study has shown t h a t proton magnetic resonance (p.m.r.) spectra can often be used to make unambiguous structure assignments to these compounds. Eighteen sultones have been characterized in this way. Many of these were prepared by the sulfonation of alkenes, a reaction involving skeletal r e a r r a ~ ~ g e m e n t I. n~ every instance the structure proved to be that expected on the basis of the reaction course assumed previously. Experimental Sources of Su1tones.-The method of preparation for most of these sultones has been given in previous p ~ b l i c a t i o n s . 3 ~Sol~ tones 9, 11, and 12 were obtained from the Shell Development Co., Emoryville, Calif., through the courtesy of Dr. Curtis Smith. The preparation of sultones 13-18 will be described in a separate publication. Spectra.-The p.m.r. spectra were taken at 25 i 2' with a Varian high-resolution spectrometer a t 40 and/or 60 Mc./sec. Pyrex tubes (5-mm. 0.d.) were filled to a height of approximately 8 cm. with chloroform solutions of the sultones. Two solutions of each sultone were prepared, one a t approximately 4 5 i concentration ( b y weight), and one a t approximately 25cV,. The chemical shift data given in Table I and in the following section are from the 4% solutions. The variation of chemical shift with concentration over the 4 to 2553 range was no more than 0.1 p . p . m . Chemical shifts were measured by the conventional side-band technique for all sultories relative to chloroform. The positive numbers in Table I refer t o resonance a t higher magnetic field, relative to the chloroform resonance; the line positions are accurate to within fl C.P.S.for 6's given t o three significant figures.
Results and Discussion It was anticipated that sultones 1, 2, and 3 would have simple spectra, since in the structures assigned no two adjacent carbon atoms hold hydrogen atoms, and (1) Abstracted in p a r t f r o m t h e X S . Thesis of R. W. Ohline, X o r t h western University, August, 1958. (2) Alfred P. Sloan Research Fellow. (3) F. G . Bordwell, R. D . C h a p m a n , a n d C. E. Osborne. .J A m . Chem. S O L . , 81, 2002 (19.59) (4) F. G . Bordwell, C. E. Osborne, a n d K . D. C h a p m a n , ibid., 81, 2698 (195'3).
the possibility of proton spin coupling is thereby precluded. The appearance of three sharp peaks in the spectrum of 1 and of 3 , and of four sharp peaks in the spectrum of 2, bears out this expectation. Judging from bond distances, one would expect five-membered sultone rings to be somewhat larger than cyclopentane rings. In sultones 1 and 3, as well as in 2 , the hydrogen atoms, methyl groups, and oxygen atoms attached to adjacent ring atoms (carbon or sulfur) must then be staggered with respect to one another. T h e single sharp peaks in the spectra of 1, 2 , and 3 are evidently average signals resulting from molecules undergoing rapid chair-chair interconversions. T h e signals for the three types of methylene groups possible for five-membered ring sultones appear in distinctly different regions of the spectrum and are useful for structure assignment. Thus, the average methylene signal for protons on the carbon atom a to oxygen (-CH,0S02-) appears a t 2 . i 9 p.p.m., whereas the average signal for the protons on the carbon atom cy to sulfur (-CH2S020-) is a t 4.05 p , p . m . ,and that for the protons on the carbon atom p to oxygen (or sulfur; -CH2-C-OS02--C-) is a t 4.8 p.p.m. T h e average signals for the corresponding methine protons for these three positions are 2.54, 3.98, and 4.96, respectively. For methyl groups the average values are 5.80, 5.85, and 6.12, respectively. T h e assignments are summarized in Fig. 1. (Compounds containing halogen or phenyl groups are not included in arriving at these averages.) Reference to Table I will show that a relatively large number of values are available for methylene groups cy to sulfur, and for methyl groups on carbon atoms cy or fi to oxygen. T h e other values are less reliable. T h e chemical shifts for the corresponding methylene and methyl groups in the two six-membered ring sultones 2 and 12 are of comparable magnitude. \Then one of the hydrogen atoms in one methyl of a gem-dimethyl group is replaced by a bromine or chlorine atom, as in 4, 14, 15, 16, or 18, the C H 2 X signal is shifted downfield from that of the parent sultone (3, 13, or 17) to an extent anticipated on the basis of
4642
VOl. 86
R. %'. OHLINE,A. L. ALLRED,A N D F. G. BORDWELL TABLE I .4SSIGNMENTS O F PROTON M A G N E T I C
RESONANCE PEAKS
TO SULTONES ( P . P . M .
RELATIVETO
THE SOLVENT CHLOROFORM)'
HICC
\ HCSOzO
HCOSOz
1
HCCOSOz CHaCOSOz CHCSOzO (including b o t h m e t h y l e n e and m e t h i n e protons)
5.53
2
4 34
3
3 97
4
3 91
5
-3
9 (t)
3 , 6 3 (-CHZBr)
-4.8
(t)
4 25 (quin)
a
3 07(d) 3 84(d)
-4
O(m)
9
2 5 (m)
10
2 58(m)
11
2 . 4 8 (t)
4 2 (m)
12
2.48 (t)
4 2(m)
13
14
15
5.68
5.98
5.78
5.99
5.62
5.89
5.71
6 10(d)
5 74 5 85
6
7
1
5.93
4 12
3.10
c-0
4 . 9 6 (quin)
5 78 5 88
-5
87 ( d )
-6
18(d)
6.04
550 5.62
-4.8 (m)
5.78 (d)
4.88 5,22
5 81 ( d )
5.76
5 74 5 88
5 93 ( d )
6 18 6 25
-4
8(m)
3 81 ( 4 )
3 . 8 0 (9)
3 . 5 5 (CHzC1)
5,59
5.90 ( d )
6.06
3.81 (9)
3 72(CHzBr)
5.58
5 91 ( d )
6.06
4643
SULTONE STRUCTURE FROM P.M.R. SPECTRA
Nov. 5, 1964
TABLE I (Continued)
HCOSO?
Sultone
HCSOtO
HCCOSO?
y)n-sr
HcCC\ CH~COSOI
1.3: (CHBr2)
CHCSOnO
5.33
5.88 (d)
c-0
5.88 5.99
3.87
5.78 5.93
6.20
5.55
5.95
Me Me
3.6
18
ML
3.75 ( CHZBr)
CH,B~
Peaks designated without parenthetical descriptions are singlets (d, doublet; t, triplet; q, quartet; quin, quintet; m, multiplet).
Numbers refer to the center when more than one peak is indicate d
Shoolery's constant^.^ T h e pattern observed is usually that of a double doublet, which is to be expected since the carbon atom a t which the C H l X group is attached is asymmetric.6 T h e signal for the methyl group attached to this carbon atom is also shifted downfield slightly. T h e signal for the gem-dimethyl group in 4 might also be resolvable, in view of the adjacent asymmetric carbon atom, but this was not observed.
and 5.85 peaks for 6). Peaks G and H (367.3 and 374.1 c.p.s. a t 60 Mc./sec. ; J = 6.8 c.P.s.) are assigned t o the (spin-coupled) methyl group on the 0-carbon atom (compare with 363 and 369 C.P.S. and J = 6.3 C.P.S. for 6). This leaves peaks D and F for the remaining methyl group, which is also spin coupled (349.1 and 355.5 c.P.s.; J = 6.4 c.P.s.). T h e methine hydrogen atoms absorb in the regions of 4.25 and 4.96 p.p.m. and are spin coupled to each other and t o a contiguous methyl group. All spin-coupling constants are 6.4 + 0.4 C.P.S. This analysis was confirmed by examination of a spectrum taken a t 40 Mc./sec. Sultone 8 corresponds in structure to 3, except that a phenyl group has replaced a methyl group on the pcarbon atom. By analogy to open-chain systems, the phenyl group should cause deshielding of the methyl and methylene protons attached to the p-carbon atoms of about 0.35 p.p.m.' Actually, one of the methylene protons is shifted only slightly downfield (by about 0.15 p.p.m.), whereas the other is shifted downfield by almost 1 p.p.m. A molecular model indicates that one of the methylene protons is held near the plane of the phenyl ring, which accounts for its unexpectedly strong deshielding. A characteristic AB pattern is observed for the methylene protons (peaks B-E in Fig. 3)
a
so
(4.05) H&' ' 0 I I (4.8) H2C-CH2
(2.97)
(methyleneprotons) (methyl and methine protons) Fig. 1.-Average chemical shifts for methylene and for methyl and methine p.m.r. signals (in p . p , m . relative t o chloroform).
T h e relative intensities of the peaks in the two triplets for 5 are not those expected b y simple firstorder theory. Analysis of this A& system a t 40 Mc./ sec. gave J A B = 7.2 c.p.s. and Y = 35.2 C.P.S. In 6 the single methyl group attached to the 6ring carbon is spin coupled to the proton on this carbon atom ( J = 6.3 c.P.s.). The geminal methyl groups are rendered nonequivalent by the adjacent asymmetric carbon atom,6 and give rise to individual signals (at 5.75 and 5.85, 6 = 0.13). T h e methylene protons in 6 are also nonequivalent and the -CH2CHCH3 grouping constitutes an ABCX3 system, which accounts for the complex multiplet pattern observed. T h e complex absorption in the methyl region for 7 (Fig. 2) can be analyzed satisfactorily by analogy to 6.
I
,jl
n
..---2,-.A-_
J JJ,__
Fig. 2.-P.m.r. spectrum a t 60 Mc./sec. of 3,4-dimethyl-4-hydroxy-2-pentanesulfonic acid sultone ( 7 ) .
T h e 5.78 and 5.88 p.p.m. peaks (C and E) are assigned to the gem-dimethyl group (compare the 5.75 ( 5 ) See L. M. J a c k m a n , "Applications of Xuclear h l a g n e t i c Resonance Spectroscopy i n Organic C h e m i s t r y , " P e r g a m o n Press, S e w Y o r k , N . Y . , 1959, C h a p t e r 4. (6) See r e f . 5 , p p . 99-103.
B
C
Fig. 3.-P.m.r.
spectrum a t 40 Mc./sec. of 2,3-dimethyl-3hydroxy-2-phenylbutanesulfonic acid 1,3-sultone (8).
with 6 = 0.77 p.p.m. and JAB = 14 C.P.S. T h e methyl group attached t o the carbon atom holding the phenyl appears to be in a relatively unaffected region with respect to phenyl, and is assigned peak H. T h e geminal methyls are slightly deshielded (peaks F and G ) . In the spectrum of 9 the methyl doublet (J = 6.3 c.P.s.) appears in the expected region and the remainder of the spectrum consists of an unresolved series of multiplets. ( 7 ) See ref. 5 , p . 53.
R IV OHLINE,A L XLLRED.A N D F G BORDWELL
4664
CH \a
cH;c-c-cn.
Vol hti C
H
'
1 EF
€0Mc/mc
li
c
b
so,
cl\o/
'I
81
.
D
I
/
I
i
Fig. 5.--T>.ni.r. spectrum a t A 0 Mc. 'sec of 3,3,1-tritnctl1).1-.11iydroxq.-%-pentanesulfonicacid sultone ( 13).
eight carbon atoms must be present as five methyl groups. T h e remaining three carbon atoms must then be contained in a five-membered sultone ring. The single remaining hydrogen atom gives a quartet centered at 3.81 p . p , m . (peaks a t A). The quartet is consistent with the expected spin coupling of this hyB 'I drogen atom with the methyl group attached to the ii __-_____sdhJLJ vL same carbon atom (1= i . 0 c.P.s.). The corresponding Fig. 4.-P.rn.r. spectra a t 40 and 60 Mc,/sec. of l-methyl-3methyl doublet is not observable in Fig. 3 sirice one-half hydroxypentanesulfonic acid 1,3-sultone (10). of the doublet is merged with another methyl signal in peak C . (Peak C was resolved in a 4 - l l c . sec. In the spectrum of 10 (Fig. 4) peaks C and E ( J = 7.8 spectrum, and also in a ( X - M c . sec. spectrum taken i n c.P.s.) are assigned t o the doublet for the single methyl hexadeuteriobenzene'"; J = 7.0 c.p.s.) group by analogy with the assignment in 9. One of the components of the doublet overlaps with the sharp peak The position of the signal for the single hydrogen atom at 3.81 p , p . m . corresponds closely to that of a (D) assigned to the gem-dimethyl groups. One would expect the geminal methyl groups to be nonequivalent hydrogen atom LY t o sulfur (see Table I and Fig, 1 ) . The two alternative positions are ruled out by this owing to the asymmetric carbon atom across the ring, observation, as well as b y the positions of the methyl but apparently the closeness of their chemical shifts precludes resolution of the peaks. The absence of signals (see below). This leads to the miambiguous assignment of structure 13 for the sultone.'! peaks near 4.0 p.p.ni. is consistent with the assigned Consistent assignments of the triethyl peaks in 13 structure,$ since there are no protons a t o sulfur (in can be made on the basis of the structure arrived a t contrast t o all the other sultones in Table I ) . The above. Thus peak B and part of C ( F i g . .ii a t 3 . 2 -CH2-CHCH3 grouping constitutes an ABM3X system. and 5.SS p . p , m . , respectively. are assigtiecl to the Analysis of both the 40-and BO-LIc. spectra were carried gpm-dimethyl group a to oxygen (Table I and Fig. 1 ) . out to arrive a t the parameters shown. In assigning The appearance of individual peaks is uxiil~rst~txitlable, since the carbon atom LY t o sulfur is :asyiiilnetric.i' O\\,/P As mentioned above, part of peak C and pc'ak 11 represent the methyl doublet. Peaks E arid I; are assigned to the gem-dimethyl group 13 to oxygm !compare Table I and Fig. 1 ) . The assignments for 13 were strengthened b y examination of the spectra of its chloro. bromo, and dibromo derivatives (14, 15, and 16, respectively). In 14 and 15 part of peak C disappears, as it should (CHrj +. C H 2 X ) , and a new peak for C H 2 X appears downfield (at 3.55 p.p.m. for 14 and 3.7% p . p , n i . for 15). In the labels X and B it is assumed that the molecule will be primarily in the conformation shown, and t h a t the each instance the quartet for the single hydrogen larger coupling constant is ascribable to the "diaxial" atom continues t o b e centered near 3 . S , and peaks I3 hydrogen pair (AX), rather than the "equatorialshift t o 5.59 and 5.58 p.p.m.. respectively. Peaks E axial" pair (BX). and F are shifted slightly downfield and are merged The regions of absorption and complexity of the spec(at 6.06 p,p.m. for both 14 and 15). The positions of peak D and the remaining part (one-third) of C are tra for sultones 11 and 12 are consistent with expectaessentially unchanged in each instance [doublet centered tions based on the other spectra. T h e -CH20group is spin coupled to the adjacent methylene group a t 5.90 for 14 ( J = 7.2 c . p . s . ) ; doublet centered a t 5.91 for 15 ( J = 7.0 c . P . s . ) ] . The downfield shifts by f i . i c.p.s. for 11 and 5.3 C . P . S . for 12. Sultone 13 was first encountered as a by-product of for B. E, and F are expected by analogy with the a commercial hydration of a butene mixture with sulshifts observed in the positions of comparable methyl furic acid.g I t s p.m.r. spectrum is shown in Fig. 5 . groups in 3 when i t is converted b y comparable substiI t will be observed on examination of Fig. 3 t h a t tution t o 4. The failure of C and D t o move supports most of the absorption occurs in the methyl region. the conclusion t h a t this methyl group is in a position (IO) W e wish t o t h a n k Dr K W . B a r t z . H u m b l e Oil a n d Refining C o From the areas under the peaks it is clear t h a t B a y t o w n . T e x a s , for t h i s s p e c t r u m . fifteen of the sixteen hydrogen atoms and five of the ( i l , I t would be a n t i c i p a t e d o n t h e basis of plevious w o r k 3 t h a t a sultijne ~
1'1
A
(81 J. \Villems, Buii. .uc c h i m . B r i g r s . 64, 409 ( l Y 5 5 ) . i 0 i A small s a m p l e of t h i s s u l t o n e was o b t a i n e d t h r o u g h t h e c o u r t e s y H . 0 > l o t t e r n . Esso Laboi-atoriez. L i n d e n , S J , sevei-al y e a r s ago. F r o m i t i chemical pi-opei-ties a n d c a r b o n a n d hydrogen analysis i t was k n o w n to be a n Octane s u i t o n e , CiHlrO3S h u t n o t e n o u g h m a t e r i a l was a v a i l a b l e t o w a r r a n t f u r t h e r woi-k toN-ard s t r u c t u r e a s s i g n m e n t u n t i l t h e a d v e n t of p m I- spectroscopy
of
of t h i s s t r u c t u r e would he f o r m e d b y sulfonation o i . ' I , , l . - l ~ t l . i m e t h y l ~ % p e n t e n e . D r . J Schaefiei- f u n p u b l i s h e d ,results) has indeed n b t a i n e d 13 in good yield b y this r e a c t i o n . ( 1 1 ) I t is of i n t e r e s t t o n o t e t h a t t h e diffet-ence in chemical s h i f t s i i large enough here (0.14 p p m . ) t o give rise t o s e p a r a t e p e a k s w h r r e a s in 10, which also has a n asymmeti-ic c a r b o n a t o m acrocs t h e ring ir.S e n i o r . R K H e t r i c k , a n d J. G. LIiIler, J . A m Chrrn . j o f , 66, 1987 ( 1 9 4 4 ~
