Trapping of Payne rearrangement intermediates with arylselenide anions

Article abstract of DOI:10.1016/j.tetlet.2014.11.103

The intermediate epoxy alcohols prepared via a Payne rearrangement can be trapped with arylselenide anions, giving mixtures of ring-opened products. The 1-arylseleno-2,3-diols are generally favored over the 3-arylseleno-1,2-diols in this process although

Full text of DOI:10.1016/j.tetlet.2014.11.103

Tetrahedron Letters 56 (2015) 3082–3085
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Trapping of Payne rearrangement intermediates with arylselenide
anions
⇑
Michael E. Jung , Daniel L. Sun
Department of Chemistry and Biochemistry, University of California, Los Angeles, CA 90095-1569, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The intermediate epoxy alcohols prepared via a Payne rearrangement can be trapped with arylselenide
anions, giving mixtures of ring-opened products. The 1-arylseleno-2,3-diols are generally favored over
the 3-arylseleno-1,2-diols in this process although the reaction of trisubstituted epoxyalcohols, for
example, 17, differs from those of disubstituted epoxyalcohols, for example, 21.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 26 October 2014
Revised 18 November 2014
Accepted 22 November 2014
Available online 27 November 2014
Keywords:
Epoxy alcohol opening
Aryl selenide anions
Payne rearrangement
Arylseleno diols
Several years ago we reported the rearrangement of epoxy alco-
hols and their silyl ethers using silyl triflates to produce silyl-
protected aldol products in a process we termed the ‘non-aldol
aldol’.1 Although the epoxy alcohols 2 prepared using a Sharpless
asymmetric epoxidation of the E 2-methyl allylic alcohols 1 gave
very good yields of the aldol products 3, the reaction of the
epoxides 5 derived from the corresponding Z allylic alcohols 4
could be problematic due presumably to steric hindrance to the
desired rearrangement, although conditions were found to produce
the anti aldol products 6 (Scheme 1).2 We wondered whether a
longer sequence involving a double inversion process might allow
us to obtain the anti aldol products 10 from the E epoxy alcohols 2
(Scheme 2). This would involve the trapping of the Payne rear-
rangement3 equilibrium of 2 and 7 with a strong nucleophile to
give 8, followed by selective protection of the secondary alcohol
and cyclization to the terminal epoxide 9 and then final Yamamoto
rearrangement4 to give the anti aldol product 10. We report here
our results of the trapping of the Payne rearrangement intermedi-
ates using aryl selenide anions.
predominately the product 13. Later Boeckman showed that this
same process occurred with the methyl substituted epoxy alcohol
15 to give the diol 16 in an elegant synthesis of (À)-kromycin.6 We
decided to study the analogous rearrangement using arylselenide
anions instead of the phenylthiolates. Herein we report the results
of that investigation.
The epoxy alcohols were all prepared by an application of
Sharpless asymmetric epoxidation7 of the readily available E- and
Z-allylic alcohols. We decided to test one substrate in order to find
the best conditions for the rearrangement-trapping process
(Scheme 4) and therefore prepared the known epoxy alcohol 17.1a
Since the preference for opening the terminal epoxide 18 over
the internal epoxide 17 is due to steric hindrance, we decided to
use a very sterically hindered arylselenide. Thus the known
In a beautiful approach to the synthesis of the alditols,5
Sharpless and Masamune reported the interception of one of the
intermediates of the Payne rearrangement with thiolates to give
selectively the 1-thiophenyl-2,3-diols (Scheme 3). Thus treatment
of the epoxy alcohol 11 with thiophenol and aq sodium hydroxide
in dioxane afforded the product of selective opening of the inter-
mediate rearranged epoxy alcohol 12 at the primary center to give
⇑
Corresponding author.
E-mail address: jung@chem.ucla.edu (M.E. Jung).
Scheme 1. Non-aldol aldol reaction to give 3 and 6.
http://dx.doi.org/10.1016/j.tetlet.2014.11.103
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

M. E. Jung, D.L. Sun / Tetrahedron Letters 56 (2015) 3082–3085
3083
Table 1
Treatment of 17 with NaSeAr to give 19 and 20
Entry
Solvent
Temp (°C)
19 (%)
20 (%)
Overall yield (%)
a
b
c
d
EtOH
EtOH
EtOH
EtOH
EtOH
iPrOH
aq tBuOH
0
22
45
85
22
13
31
40
48
44
30
10
45
53
23
34
35
13
10
58
84
63
82
79
43
20
Scheme 2. Alternative route for anti aldol products 10.
ea
fb
gb
100
100
a
Added all at once.
0.5 M NaOH was used.
b
Table 2
Reaction of 17 under microwave conditions to give compounds 19 and 20
Entry
Time (h)
Temp (°C)
19 (%)
20 (%)
Overall yield (%)
a
b
c
48
12
48
12
60
100
100
85
85
80
51
46
41
60
56
10
23
0
17
0
61
69
41
77
56
d
ea
Scheme 3. Selective trapping of the Payne intermediate 12.
a
Not microwave; sealed tube.
bis(2,4,6-trimethylphenyl)diselenide (dimesityl diselenide)8 was
reduced with sodium borohydride to give the mesitylselenide
anion. To a solution of the epoxy alcohol 17 in ethanol was added
over 2 h a solution of the sodium mesitylselenide and 1 M (or
0.5 M) NaOH. All of the reactions were stirred for 18 h. The results
(Table 1) showed that the temperature had a significant effect,
namely higher temperatures generally gave more of the desired
1-seleno-2,3-diol 19 than the product of direct opening, the
3-seleno-1,2-diol 20 (entries a–d).9 Heating 17 with the selenide
in ethanol at 85 °C (reflux) afforded the desired product in 48%
yield along with 34% yield of the undesired product (entry d).
Interestingly, adding all of the selenide and base at once seemed
to improve the yield of the desired product (entry e vs entry b).
However, higher temperatures in other solvents, for example,
100 °C in isopropanol or aq tert-butanol, gave much poorer results
(entries f and g). The ratio of the desired to undesired product was
best at 45 °C (1.7:1) but the overall yield was lower than that at
85 °C where the ratio was 1.4:1 (entries c and d).
We next investigated the use of microwave heating for this pro-
cess since some non-thermal effects have been observed especially
in reactions of polar substrates.10 Thus the epoxy alcohol 17 was
treated with the sodium mesitylselenide and 1 M NaOH in ethanol
at 22 °C. The mixture was then placed in an industrial microwave
oven and heated to the indicated temperatures for the time shown.
The results (Table 2) indicated that these microwave conditions
were favorable for the formation of the desired product 19 in pref-
erence to the undesired product 20. At all temperatures, shorter
reaction times were better, presumably due to decomposition of
the products on prolonged heating with base. Thus at 100 °C, the
shorter time of 12 h was somewhat better than 48 h, although
the ratio was better at longer times (entries a and b). And at
85 °C, heating for 12 h afforded 60% of the desired product 19 along
Scheme 5. Rearrangement-trapping of the disubstituted epoxy alcohols 21 to give
the three products 23–25.
Scheme 4. Formation of the arylseleno diols.

3084
M. E. Jung, D.L. Sun / Tetrahedron Letters 56 (2015) 3082–3085
with 17% of the undesired 20, while heating for 48 h gave only the
desired product but in the lower yield of 41% (entries c and d).
Finally we also investigated the use of heating in a sealed tube
for an extended period. Thus heating an ethanolic solution of 17
at 80 °C in a sealed tube for 60 h gave only the desired product
19 in an isolated yield of 56%. The undesired product 20 seems
to be selectively destroyed on prolonged heating. Therefore the
best method for preparing the desired products was heating at
80–85 °C in ethanol, either via microwave or thermal heating.
Using a series of disubstituted epoxides, we then looked at the
substrate scope of the rearrangement-trapping process. In these
substrates, three different products can be obtained, from opening
at either end of the original epoxide 21, giving 24 and 25, or the
terminal epoxide 22 giving 23 (Scheme 5). All of the epoxy alcohol
substrates, 21a–g, were prepared by known routes5 using the
Sharpless asymmetric epoxidation. Each was mixed with the
sodium mesitylselenide and 1 M NaOH in ethanol and heated at
85 °C for 18 h. The results (Table 3) indicate firstly that the overall
yields for these substrates were excellent, generally greater than
90%. Secondly, for these substrates not having the 2-methyl
substituent, significant opening at C2 occurred to generate the
2-seleno-1,3-diol 24.11 Thus the relatively unhindered substrates
gave significant amounts of the 2-seleno products (entries a–d)
with relatively similar amounts of the 1-seleno and 3-seleno
products, 23 and 25, respectively. When the attack of the hindered
arylselenide anion at C-3 is made more difficult, due to the steric
interaction of the acetonides, then no opening at C-3 is observed
but only a mixture of the C-1 and C-2 products, 23 and 24. In these
latter three cases (entries e–g), reasonable yields of the 1-seleno
product 23 could be isolated.
Scheme 6. Reversible formation of the selenodiols 19 and 20.
Since heating the mixture of the epoxy alcohols and the arylsel-
enide anion in ethanol for longer periods of time seemed to give
more of the desired 1-seleno product with respect to the 3-seleno
product (Table 2), we wondered if the formation of the selenodiols
might well be reversible under these conditions. The product 20
formed by opening the starting epoxyalcohol 17 at C-3 might well
be revert back to 17 via loss of the mesitylselenide anion before
being converted via 18 into the desired product 19 (Scheme 6).
Therefore we treated the isolated, pure 3-seleno-1,2-diol 20 with
sodium mesitylselenide and 1 M sodium hydroxide in ethanol at
85 °C for 18 h and isolated a mixture of the two products, the
starting material in 74% yield and the rearranged product 19 in
23% isolated yield (Scheme 7). Thus the seleno diol 20 must lose
the mesitylselenide anion and revert back to the epoxide 17, rear-
range via Payne rearrangement to the isomeric epoxide 18, and
Table 3
Reactions of disubstituted epoxy alcohols
Entry
a
Substrate
Yield of 1-seleno-2,3-diol 23 (%)
Yield of 2-seleno-1,3-diol 24 (%)
Yield of 3-seleno-1,2-diol 25 (%)
Overall yield (%)
85
39
0
46
b
24
51
24
99
c
42
28
27
29
30
36
99
93
d
e
f
65
53
29
40
0
0
94
93
g
52
24
0
76

M. E. Jung, D.L. Sun / Tetrahedron Letters 56 (2015) 3082–3085
3085
References and notes
1. (a) Jung, M. E.; D’Amico, D. C. J. Am. Chem. Soc. 1993, 115, 12208–12209; (b)
Jung, M. E.; D’Amico, D. C. J. Am. Chem. Soc. 1997, 119, 12150–12158; (c) Jung,
M. E.; Marquez, R. Tetrahedron Lett. 1999, 40, 3129–3132; (d) Jung, M. E.;
Hoffmann, B.; Rausch, B.; Contreras, J.-M. Org. Lett. 2003, 5, 3159–3161.
2. Jung, M. E.; Lee, W. S.; Sun, D. Org. Lett. 1999, 1, 307–309. The yields of the anti
aldol products in these hindered cases were lower due to competing
elimination to give the allylic bis(silyl ethers).
3. (a) Payne, G. B. J. Org. Chem. 1962, 27, 3819–3822; (b) Hanson, R. M. In Organic
Reactions; Overman, L. E. et al., Eds.; Wiley, 2002; Vol. 60, pp 1–156; (c) Wrobel,
J. E.; Ganem, B. J. Org. Chem. 1983, 48, 3761–3769; (d) Jung, M. E.; van den
Heuvel, A. Tetrahedron Lett. 2002, 43, 8169–8172.
Scheme 7. Rearrangement-trapping of the disubstituted epoxy alcohols 21 to give
the three products 23–25.
4. (a) Maruoka, K.; Sato, J.; Yamamoto, H. J. Am. Chem. Soc. 1991, 113, 5449–5450;
(b) Maruoka, K.; Sato, J.; Yamamoto, H. Tetrahedron 1992, 48, 3749–3762.
5. (a) Katsuki, T.; Lee, A. W. M.; Ma, P.; Martin, S. V.; Masamune, S.; Sharpless, K.
B.; Tuddenham, D.; Walker, F. J. J. Org. Chem. 1982, 47, 1373–1378; (b) Behrens,
C. H.; Ko, S. Y.; Sharpless, K. B.; Walker, F. J. J. Org. Chem. 1985, 50, 5687–5696.
6. Boeckman, R. K., Jr.; Pruitt, J. R. J. Am. Chem. Soc. 1989, 111, 8286–8288.
7. (a) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974–5976; (b)
Johnson, R. A.; Sharpless, K. B. Comp. Org. Synth. 1991, 7, 389–436; (c) Katsuki,
T.; Martin, V. S. Org. React. 1996, 48, 1–300.
then be opened at the terminal end of the epoxide 18 with the aryl-
selenide anion to give 19. This implies that the formation of the
products may be partially controlled by thermodynamic effects
and not merely kinetic effects.
In conclusion, we have developed a useful method for the
conversion of 1,2-epoxy alcohols to the corresponding 1-seleno-
2,3-diols using the hindered mesitylselenide anion under basic
conditions in refluxing ethanol. The use of these compounds is
under investigation and will be reported in due course.
8. Hiroi, K.; Sato, S. Synthesis 1985, 635–638.
9. The structures of the 1-seleno-2,3-diol 19 and the 3-seleno-1,2-diol 20 were
determined by high field proton NMR experiments, in which the chemical shift
of the easily identifiable methylene group was at either lower field (
oxygen) or at higher field ( to selenium).
a to
a
Acknowledgments
10. (a) de la Hoz, A.; Diaz-Ortiz, A.; Moreno, A. Chem. Soc. Rev. 2005, 34, 164–178;
(b) Galema, S. A. Chem. Soc. Rev. 1997, 26, 233–238; (c) Mingos, M. P.; Baghurst,
D. R. Chem. Soc. Rev. 1991, 20, 1–47; (d) Rosana, M. R.; Tao, Y.; Stiegman, A. E.;
Dudley, G. B. Chem. Sci. 2012, 3, 1240–1244; (e) Kappe, C. O.; Pieber, B.;
Dallinger, D. Angew. Chem., Int. Ed. 2013, 52, 1088–1094.
11. The structures of the 2-seleno-1,3-diols 24 and the 3-seleno-1,2-diols, 25 were
determined by high field proton NMR experiments, especially of the acetylated
products, by careful matching of coupling constants and chemical shifts of the
lower field protons, along with NOESY and COSY spectra.
This material is based upon work supported by the United
States National Science Foundation under equipment Grant no.
CHE-1048804. We dedicate this publication to Harry Wasserman,
a man of rare talent and grace.
Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.
11.103.