New Structures from Multiple Rearrangements of Propargylic Dialkoxy Disulfides

Article abstract of DOI:10.1021/ja036962y

Sundry dipropargyloxy disulfides have been successfully prepared, and their rearrangement paths and products have been found to be dependent upon their substitution pattern. Inter alia compounds of two heretofore unknown structure types - 10 and 11 on one

Full text of DOI:10.1021/ja036962y

Published on Web 11/04/2003
New Structures from Multiple Rearrangements of Propargylic Dialkoxy
Disulfides
Samuel Braverman,* Tatiana Pechenick, Hugo E. Gottlieb, and Milon Sprecher
Department of Chemistry, Bar-Ilan UniVersity, Ramat-Gan 52900, Israel
Received June 29, 2003; E-mail: bravers@mail.biu.ac.il
Scheme 1
The field of sigmatropic rearrangements of allylic and propargylic
esters of sulfur acids at various oxidation states has proven to be a
rich source of synthetically valuable and mechanistically intriguing
reactions, often yielding novel and surprising products.1 Herein we
report on some additional examples of this type, namely, the
formation and characterization of the heretofore unknown structures
of types 10, 11, 14, and 15.
The reported failure of Thompson2 notwithstanding, we have
recently succeeded in developing the necessary methodology to
prepare allylic3 and propargylic4 dialkoxy disulfides in high yield
and have found them to be stable in CHCl3 solution at -18 °C for
extended periods. In refluxing acetonitrile, diallyloxy disulfides
undergo double [2,3]-sigmatropic rearrangement to Vic-disulfoxides,
Scheme 2
which spontaneously rearrange to the appropriate bis(allyl) thio-
sulfonates, as expected (Scheme 1).3,5,6
By parallel series of reactions, dipropargyloxy disulfides would
have been expected to yield bis(allenyl) thiosulfonates (2). In fact,
however, the first isolated products were found to have a new and
unusual type of structure.4 They were 6,7-dithiabicyclo[3.1.1]-
heptane-2-one-6-oxides (7; Scheme 2) incorporating the 1,3-
dithiacyclobutane-1-oxide moiety recently identified by Block7 in
the zwiebelanes isolated from freshly cut onion.
Apparently, the presence of the additional double bonds in the
allenyl groups diverts the reaction from the path of Scheme 1 at
the disulfoxide stage. A careful examination and exacting workup
of the chloroform solution of 1a (1, RdR′ ) H; Scheme 2)
following 7 h reflux showed that 7a was accompanied by two
isomeric products. These were individually isolated and on the basis
of extensive spectroscopic determinations, which included full
1
analysis of H and 13C NMR data with the aid of 2D techniques
such as COSY, NOESY, HMQC, and HMBC, as well as IR and
HRMS, were assigned the surprising structures 10a and 11a (10,
11, RdR′ ) H; Scheme 2).8 The two regioisomers can be
distinguished by comparing the chemical shifts of the olefinic
carbons (C-1 and C-2; these are identified by their long-range C-H
interactions), since the main effect of the thiosulfonate function is
deshielding of the carbon â to the sulfonyl group and shielding of
the carbon â to the sulfide atom. The ratio of 7a/10a/11a isolated
was 32:3:14, respectively (combined yield, 49%). Substitution at
C-γ of the propargyl moieties significantly improved the isolated
yields and affected the ratio of products. Thus, the rearrangement
of 1b (1, R ) H, R′) CH3CH2-) led to 7b, 10b, and 11b,
individually isolated, in yields of 10, 41, and 34%, respectively.
Sterically demanding substituents at these positions slowed the
reaction but had an even more marked effect on yield and product
composition. After 25 h reflux in chloroform, 1c (1, R ) H, R′ )
(CH3)3C-; Scheme 2) yielded only 8% of 7c and 92% of 10c, and
1d (1, R ) H, R′ ) (CH3)3Si-; Scheme 2) after 20 h yielded 57%
10d and 19% 11d, but no 7d. The R,â-unsaturated four-membered
cyclic thiosulfonate grouping present in 10 and 11 appears to have
been unknown to date, though compounds having a saturated four-
membered ring thiosulfonate function have been reported
previously7a,9 as the products of dimerization of sulfines which is
9
14290
J. AM. CHEM. SOC. 2003, 125, 14290-14291
10.1021/ja036962y CCC: $25.00 © 2003 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3
The stereochemistry of the two double bonds in structure 14/15
was established by NOESY experiments, which showed a strong
interaction of the respective bridgehead hydrogens with the nearby
tert-butyl or adamantyl hydrogens but no interaction with the vinyl
hydrogens.
In these last cases, the reaction path of Scheme 2 is abandoned
at the point of the [3,3]-sigmatropic rearrangement of 5, possibly
because it would involve incipient allylic steric interactions of the
bulky substituents. Instead, it seems, the allenic carbon of 12 bonds
to the sulfinyl sulfur with concomitant cleavage of the other S-O
bond (12 f 13). Though this proposed bond rearrangement may
be rationalized in various ways, we suggest that it be viewed as a
homologous hetero-Cope rearrangement in which the delocalized
cross-conjugated 2-oxyallyl-1,3-dipole produced in 13 replaces a
simple CdC π-bond. Structure 14/15 is accessible from 13 by an
internal 1,3-dipolar addition to the thione double bond.
The 3,6-dialkylidene-2-oxa-5,7-dithiabicyclo[2.2.1]heptane 5-ox-
ide structure 14/15 appears to have no precedence in the literature,
though Baudin11 has reported the isolation of a 3,6-dialkylidene-
2-oxa-5,7-dithiabicyclo[2.2.1]heptane 5,5-dioxide from the acid
treatment of a N-morpholino 3,3-disubstituted propa-1,2-dienesulfi-
namide.
followed by intramolecular disproportionation. A number of a priori
reasonable mechanisms may be proposed for the formation of 7,
10, and 11. However, applying Occam’s razor, we tentatively
suggest the reaction path of Scheme 2. As in Scheme 1, a double
[2,3]-sigmatropic rearrangement converts 1 to an R-disulfoxide, 3,
which dissociates to two allenyl sulfinyl radicals, 4.5a Sulfinyl
radicals are known to be capable of reacting either at the oxygen
or the sulfur atom.5 Recombination of two such radicals via the
sulfinyl oxygen of one and C-2 of the other gives 5, which converts
to 7 by tandem [3,3]-sigmatropic rearrangement (5 f 6) and [2 +
2] cycloaddition (6 f 7). Alternatively, two radicals 4 recombine
via the sulfinyl sulfur of one and C-2 of the other to yield 8 (Scheme
2). It appears that the stable conformation of 4, permitting
conjugative stabilization of the free radical and distancing the
sulfinyl oxygen from the allenyl π-electrons, is one in which
approach to oxygen is hindered when R′ is bulky. This may be the
rationale for the preferential reaction of the sulfur in such cases.
The [3,3]-sigmatropic rearrangement of 8 produces the disulfine
9, intramolecular disproportionation of which in a manner analogous
to that found in the intermolecular dimerization of sulfines7a,9 and
in the conversion of R-disulfoxides to thiosulfonates5,6 leads to 10
and 11.
We have previously reported that R-substituted dipropargyloxy
disulfides such as 1e (1, R ) CH3-, R′ ) H; Scheme 2) rearranged
relatively rapidly (2 h, refluxing chloroform) to a mixture of the Z
and E isomers of 7e. No evidence for accompanying 10e or 11e
was found. Investigating the possible effect of bulky R-substituents,
we were astounded at the vagary of this system. The rearrange-
ments of 1f (1, R ) (CH3)3C-, R′ ) H; Scheme 2) and 1g (1, R
) adamantyl-, R′ ) H; Scheme 2) were rapid, as was 1e, but the
two isomeric products obtained and chromatographically separated
in each case, were not derivatives of structures 7, 10, or 11.
Extensive spectroscopic determinations as detailed above for the
latter led to the identification of the two pairs of isomers as 14f/
15f (14/15, R ) (CH3)3C-; Scheme 3; yield 57%) and 14g/15g
(14/15, R ) adamantyl-; Scheme 3; yield 62%), the components
of each pair differing from each other in the stereochemistry of
the sulfoxide group, exo or endo.10 Heating overnight in chloroform
solution led to interconversion of the isomers of each pair,
presumably by pyramidal inversion of the sulfoxide function. Each
isomer separately led to the same equilibrium mixture (e.g., the
equilibrium ratio of 14f to 15f was 2:5).
A more detailed discussion of mechanistic and steric consider-
ations, as well as the results of ongoing investigations, will be
presented in the full article.
Acknowledgment. This research was supported by the Israel
Science Foundation (Grant No. 280/01-1).
Supporting Information Available: Experimental procedures and
full spectroscopic data for all new compounds (10a/11a, 10b/11b, 10c,
10d/11d, 14f/15f, and 14g/15g) (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Braverman, S. In Chemistry of Sulfones and Sulfoxides; Patai, S.,
Rappoport, Z., Stirling, C. J. M., Eds.; Wiley: New York, 1988; Chapter
13. (b) Braverman, S. In Chemistry of Sulfones and Sulfoxides; Patai, S.,
Rappoport, Z., Stirling, C. J. M., Eds.; Wiley: New York, 1988; Chapter
14. (c) Braverman, S. In The Chemistry of Sulfinic Acids, Esters and Their
DeriVatiVes; Patai, S. Ed.; Wiley: New York, 1990; Chapter 11. (d)
Braverman, S. in The Chemistry of Sulfenic Acids and Their DeriVatiVes,
Patai, S. Ed.; Wiley: New York, 1990; Chapter 8.
(2) Thompson, Q. E.; Crutchfield, M. M.; Dietrich, M. W.; Pierron, E. J.
Org. Chem. 1965, 30, 2692-2696.
(3) Braverman, S.; Pechenick, T. Tetrahedron Lett. 2002, 43, 499-502.
(4) Braverman, S.; Pechenick, T.; Gottlieb, E. H. Tetrahedron Lett. 2003,
44, 777-780.
(5) (a) Freeman, F. Chem. ReV. 1984, 84, 117-135. (b) Lacombe, S. M. ReV.
Heteroat. Chem. 1999, 21, 1-41.
(6) A recent paper, Gregory, D. D.; Jenks, W. S. J. Phys. Chem. A 2003,
107, 3414-3423, disputes the previous proposal that the conversion of
the intermediate sulfenyl sulfinate to the thiosulfonate proceeds by a free
radical mechanism.
(7) (a) Block, E. Angew. Chem., Int. Ed. Engl. 1992, 31, 1135-1178. (b)
Block, E.; Thiruvazhi, M.; Toscano, P. J.; Bayer, T.; Grisoni, S.; Zhao,
S. H. J. Am. Chem. Soc. 1996, 118, 2790-2798. (c) Block, E.; Bayer, T.;
Naganathan, S.; Zhao, S. H. J. Am. Chem. Soc. 1996, 118, 2799-2810.
(8) For spectral data for 10a/11a, 10b/11b, 10c, and 10d/11d, see Supporting
Information.
(9) (a) Block, E.; Bazzi, A. A.; Revelle, L. K. J. Am. Chem. Soc. 1980, 102,
2490-2491. (b) Baudin, J.-B.; Commenil, M.-G.; Julia, S. A.; Wang, Y.
Bull. Soc. Chim. Fr. 1996, 133, 515-529. (c) For a general review on
sulfines, see: Zwanenburg, B. Recl. TraV. Chim. Pays-Bas 1982, 101,
1-27.
(10) For spectral data for 14f/15f and 14g/15g, see Supporting Information.
(11) Baudin, J.-B.; Julia, S. A.; Ruel, O.; Toupet, L.; Wang, Y. Bull. Soc. Chim.
Fr. 1995, 132, 79-85.
JA036962Y
9
J. AM. CHEM. SOC. VOL. 125, NO. 47, 2003 14291