Heterocycle synthesis based on allylic alcohol transposition using traceless trapping groups

Article abstract of DOI:10.1002/anie.201402010

Allylic alcohols undergo transposition reactions in the presence of Re ₂O₇ whereby the equilibrium can be dictated by trapping one isomer with a pendent electrophile. Additional ionization can occur when the trapping group is an aldehyde or ketone, thus leading to cyclic oxocarbenium ion formation. Terminating the process through bimolecular nucleophilic addition into the intermediate provides a versatile method for the synthesis of diverse oxygen-containing heterocycles. Understanding the relative rates of the steps in the sequence leads to the design of reactions which create multiple stereocenters with good to excellent levels of control.

Full text of DOI:10.1002/anie.201402010

.
Angewandte
Communications
DOI: 10.1002/anie.201402010
Synthetic Methods
Heterocycle Synthesis Based on Allylic Alcohol Transposition Using
Traceless Trapping Groups**
Youwei Xie and Paul E. Floreancig*
Abstract: Allylic alcohols undergo transposition reactions in
the presence of Re2O7 whereby the equilibrium can be dictated
by trapping one isomer with a pendent electrophile. Additional
ionization can occur when the trapping group is an aldehyde or
ketone, thus leading to cyclic oxocarbenium ion formation.
Terminating the process through bimolecular nucleophilic
addition into the intermediate provides a versatile method for
the synthesis of diverse oxygen-containing heterocycles. Under-
standing the relative rates of the steps in the sequence leads to
the design of reactions which create multiple stereocenters with
good to excellent levels of control.
Scheme 1. Transposition, trapping, ionization, and nucleophilic termi-
nation. R,R’=H or alkyl, Nu=nucleophile.
nation step. Exploiting rate differences between the individ-
ual steps in the sequence allows the design of unique
diastereoselective reactions.
The feasibility of this plan was demonstrated by using
Et3SiH as the terminating reagent. Several examples are
shown in Scheme 2.[6] The reactions were conducted with
3 mol% Re2O7 and two equivalents of Et3SiH at room
temperature. Many reactions proceed in excellent yield but
T
his manuscript describes the synthesis of oxygen-containing
heterocycles through a sequence of allylic alcohol trans-
position, intramolecular trapping, oxocarbenium ion forma-
tion, and intermolecular nucleophilic addition. We[1] and
others[2] have been exploring the use of allylic alcohol
transposition and trapping reactions for stereocontrolled
[3]
heterocycle syntheses. Our efforts have employed Re2O7
as a catalyst to initiate reversible allylic alcohol isomer-
ization[4] and promote ring formation through nucleophilic
addition of the hydroxy group of one isomeric alcohol to
a proximal electrophile. A range of electrophiles are suitable
for trapping the hydroxy group, and thermodynamically
controlled stereoselectivity can be achieved if the reactions
that lead to the final product are reversible. This stereochem-
ical editing strategy can facilitate synthesis by minimizing
reliance upon reagent-based stereoselective protocols.
The products in our initial studies generally contain
a vestige of the electrophile. Increased versatility for this
process could be achieved by employing a traceless trapping
group. This approach can be realized through adding the
allylic alcohol to a carbonyl group and ionizing the product to
form an oxocarbenium ion[5] which can be trapped by an
additional nucleophile to terminate the sequence (Scheme 1).
The sequence described herein significantly expands the
range of products that can be directly accessed through the
method by eliminating the remnant of the original trapping
group and by utilizing different nucleophiles in the termi-
Scheme 2. Reactivity in tetrahydrofuran and tetrahydropyran formation.
require somewhat prolonged exposure for completion
because the initial products arise from alcohol addition to
the intermediate oxocarbenium ion, and is consistent with
a recent report from the group of Dussault.[7] The final
products form through the ionization of the intermediate
acetal products with subsequent oxocarbenium ion reduction.
Many factors can influence the rates of these reactions.
However several trends become apparent upon analyzing the
results in Scheme 2. Secondary allylic alcohol substrates yield
products more rapidly than primary allylic alcohol substrates
(2 and 4 versus 1 and 3). We postulate that this results from
increased access to the oxocarbenium ion since steric
interactions in the intermediate mixed acetal should promote
ionization and, subsequently, reduction. Tetrahydrofurans
form more quickly than tetrahydropyrans with similar
[*] Y. Xie, Prof. P. E. Floreancig
Department of Chemistry, University of Pittsburgh
Pittsburgh, PA 15260 (USA)
E-mail: florean@pitt.edu
[**] This work was supported by a grant from the National Science
Foundation (CHE-1151979) and the University of Pittsburgh
through a Mellon Fellowship to Y.X. We thank Dr. Bhaskar Godugu
for conducting challenging mass spectrometric analyses.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201402010.
4926
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 4926 –4929

Angewandte
Chemie
Table 1: Scope expansion.[a]
substitution patterns (1, 2, 5, and 6 versus 3, 4, 7, and 8). Again
this arises from increased access to the oxocarbenium ion
intermediate since tetrahydrofuranyl ethers ionize more
rapidly than tetrahydropyranyl ethers.[8] Ketone trapping
groups promote faster reactions than the corresponding
aldehydes (5, 6, 7, and 8 versus 1, 2, 3, and 4). However,
dehydration becomes a competitive side reaction for ketone
substrates which contain secondary allylic alcohols. We
postulate that the relative stability of ketone-derived oxocar-
benium ions reduces the rate of trapping, and increased
substitution augments the potential for allylic cation forma-
tion, thereby enhancing opportunities for decomposition
through an E1 pathway. Acceptable yields can be achieved
for these reactions simply by lowering the temperature.
Stereocontrol was excellent for tetrahydropyran formation
from ketone-containing substrates, but was poor for tetrahy-
drofuran formation. This outcome corresponds to established
models for nucleophilic additions to cyclic oxocarbenium
ions.[9]
Entry Substrate
Nucleophile Product
Yield
[%][b]
1
2
3
4
Et3SiH
Et3SiH
Et3SiH
Et3SiH
81
92
90
92
5
6
7
Et3SiH
Et3SiH
85
Structural diversity in these reactions can be accessed by
variations in the substrate, through the use of different
ketones as trapping groups, or through the use of different
nucleophiles. Examples of these strategies are shown in
Table 1.
68
80[c]
The method is tolerant of numerous functional groups,
including moieties which stabilize the carbonyl trapping
group through conjugation. Utilizing an epoxide as the
trapping group provides opportunities for cascade cycliza-
tions[10] in which stereochemical information from the epox-
ide is transferred to distal ends of the product. A limited
number of nucleophiles was studied to determine the
suitability of the protocol for fragment coupling reactions.
Potassium alkynyltrifluoroborates,[11] allylsilanes,[7] and silyl
ketene acetals add into tetrahydrofuranyl oxocarbenium ions.
Stereocontrol was not observed in these reactions, and is
consistent with the reductive oxocarbenium ion quenching
experiments. Allyl trimethylsilane does not add into tetrahy-
dropyranyl oxocarbenium ions, presumably because these
reaction conditions do not induce a sufficient level of
ionization to promote reactions with weak p nucleophiles.
This problem was circumvented by adding the anion-binding
8
9
21
21
86[c]
46[c]
10
83[d]
[a] See the Supporting Information for detailed procedures, substrate
syntheses, and stereochemical determination. [b] Isolated as a single
stereoisomer unless otherwise noted. [c] Isolated as a nearly 1:1
diastereomeric mixture. [d] Run in the presence of [3,5-
(CF3)2C6H3NH]2SO2. See text for details. TBS=tert-butyldimethylsilyl.
[12]
sulfamide cocatalyst [3,5-(CF3)2C6H3NH]2SO2 to the reac-
tion to provide 26. Literature precedent[13] suggests that the
cocatalyst enhances reactivity by binding to the perrhenate
anion, thereby enhancing the acidity of Re2O7 or HReO4. This
result is significant because it demonstrates that thrmody-
namically disfavored isomers can be accessed and that
reductive and alkylative oxocarbenium ion quenching pro-
vide complementary 2,6-sterereochemical relationships in the
products.
An interesting limitation arises from the incorporation of
p electrophiles, as shown in the conversion of 27 into 28 and
29 (Scheme 3). The intermediate oxocarbenium ion 30 exists
in equilibrium with the enol ether 31. Intramolecular addition
of the enol ether into the enone is faster than the reduction,
thus generating the oxocarbenium ion 32. Reduction of 32
yields 28 and 29. This pathway suggests new opportunities for
reaction development which utilize the trapping group first as
an electrophile and then as a nucleophile.
Scheme 3. Reaction in the presence of a p electrophile.
These results led us to study the impact of a pre-existing
chiral center on the stereochemical outcome of the reac-
tion.[2b] Success in this endeavor requires a proper balance
between the rates of equilibration with the rates of oxocar-
Angew. Chem. Int. Ed. 2014, 53, 4926 –4929
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4927

.
Angewandte
Communications
Table 2: Relative stereocontrol.[a]
Entry Substrate Nucleophile Product
Yield [%][b]
(d.r.)
1
Et3SiH
Ph3SiH
Et3SiH
96 (3:1)
82 (9:1)
77 (11:1)
2
3
33
34
Scheme 4. An analysis of competitive processes for relative diastereo-
control. a=isomerization, b=cyclization, c=ionization, d=termina-
tion.
4
5
6
Et3SiH
Et3SiH
Et3SiH
83 (3:1)
benium ion formation and trapping, as illustrated in
Scheme 4. High levels of stereocontrol should be achieved if
the stereochemical equilibration step is rapid, if the equilib-
rium between the hydroxy carbonyl compound and the cyclic
hemiacetal does not strongly favor the cyclic hemiacetal, if the
ionization step is readily reversible, and if the nucleophilic
trapping step is slow. This protocol also requires that the
energetic difference between the equilibrating lactols be
sufficient to favor one isomer and that the stereocontrol in the
trapping step be high.
87 (4.6:1)
88 (30:1)
Examples are shown in Table 2. These studies show the
relationship between allylic alcohol structure, trapping group,
product substitution pattern, and termination agent on
reaction diastereocontrol. All tetrahydropyran-forming reac-
tions provided a 2,6-cis-stereochemical relationship. A com-
parison of entries 1 and 2 shows that the termination agent
can influence the 2,3-stereochemical relationship, with
Ph3SiH providing a greater degree of stereocontrol, relative
to the more reactive Et3SiH.[14] This outcome is consistent
with ionization being reversible and the less reactive trapping
agent allowing greater opportunities for stereochemical
equilibration. Substrates with secondary allylic alcohols
show higher selectivity than similarly substituted primary
alcohol substrates. This difference results from the higher rate
of stereochemical equilibration through a cationic intermedi-
ate. The equilibrium of the cyclization step influences
stereocontrol. Aldehydes are more prone to exist in the
lactol form than ketones, thus limiting the concentration of
stereochemically labile allylic alcohol and leading to dimin-
ished levels of 2,3-stereocontrol (entry 4 versus entry 3).
Stereocontrol is enhanced for reactions that establish a 2,4-
stereochemical relationship in the product (entries 5, 6, and
7). A significant steric clash between an axially oriented
substituent and the silane is present in the termination step
that leads to the minor product in these reactions. This slows
reduction, allows equilibration to the more reactive precursor
which leads to the major product.
7
Et3SiH
Et3SiH
82 (1:0)
75[c]
(56:33:11)
8
9
95[c]
(65:22:13)[d]
[a] See the Supporting Information for detailed procedures, substrate
syntheses, and stereochemical determination. [b] Yield of the isolated
product. The d.r. values were determined by 1H NMR spectroscopy.
[c] Isolated as a diastereomeric mixture. [d] 79:9:12 upon re-exposure to
Re2O7.
Stereocontrol in tetrahydrofuran formation was lower
than it was for tetrahydropyran formation, but interesting
trends were noted. The ketone 45 underwent reductive
cyclization to form three isomers in a 5:3:1 ratio, with the
isomer 46 being the major product. This reaction proceeds
through the oxocarbenium ions 49 and 50 (Scheme 5), with
the 2,3-trans-isomer 49 dominating, as expected, to avoid
eclipsing interactions in the ionization step.[15] Modest selec-
tivity was observed for the approach of the silane from the
Scheme 5. Stereocontrol in tetrahydofuran formation.
4928 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 4926 –4929

Angewandte
Chemie
concave face, opposite to the methyl group, and is in accord
with the model proposed by Woerpel and co-workers.[9c] The
aldehyde 47 also formed three stereoisomers upon allylative
cyclization, with 48 as the major product. Thus stereochemical
complementarity is observed for the major products of two
tetrahydrofuran-producing reactions. The stereochemical
preference for 53 over 54 is diminished relative to the
preference of 49 over 50, as was seen in the tetrahydropyran
series. However Woerpel selectivity in the termination step
was higher in this process, presumably because of the lower
reactivity of the nucleophile, thus leading to a 58% yield of
the major product. Resubjecting the product mixture to the
reaction conditions at reflux increases the selectivity for the
formation of 48, with the ratio of 48/55/56 changing to 79:9:12
by ionization of 55 to form the allylic cation intermediate with
subsequent closure to form the more stable 2,3-trans isomer.
We have demonstrated that heterocycles can be prepared
through a sequence of allylic alcohol transposition, intra-
molecular trapping, ionization, and intermolecular termina-
tion. This protocol renders the initial trapping group traceless,
thereby conferring significant scope to the process. Primary
and secondary allylic alcohols can be used as substrates,
ketones and aldehydes are suitable trapping groups, and
silanes, p nucleophiles, and alcohols serve as terminating
agents in the formation of tetrahydrofurans and tetrahydro-
pyrans. Reactivity is controlled by the concentration of the
intermediate oxocarbenium ion and the nucleophilicity of the
trapping agent. Therefore the substitution pattern of the
allylic alcohol, the trapping group, and the nucleophile
influence the rate of the process. Relative stereocontrol can
be achieved by rational manipulations of the transposition
and termination rates, and complementary stereoisomers can
be accessed through reductive and alkylative quenching
processes. High stereocontrol can be achieved when the rate
of transposition is high and the rate of termination is low. The
capacity to couple ring formation with equilibrating stereo-
center generation provides a useful alternative to reagent-
based stereocontrol for the synthesis of stereochemically rich
cyclic structures.
[1] a) H. H. Jung, J. R. Seiders II, P. E. Floreancig, Angew. Chem.
2007, 119, 8616; Angew. Chem. Int. Ed. 2007, 46, 8464; Y. Xie,
P. E. Floreancig, Chem. Sci. 2011, 2, 2423; Y. Xie, P. E.
Floreancig, Angew. Chem. 2013, 125, 653; Angew. Chem. Int.
Ed. 2013, 52, 625.
[2] a) E. C. Hansen, D. Lee, J. Am. Chem. Soc. 2006, 128, 8142;
b) A. T. Herrmann, T. Saito, C. E. Stivala, J. Tom, A. Zakarian, J.
Am. Chem. Soc. 2010, 132, 5962; c) S. Y. Yun, E. C. Hansen, I.
Volchkov, W. Y. Lo, D. Lee, Angew. Chem. 2010, 122, 4357;
Angew. Chem. Int. Ed. 2010, 49, 4261.
[3] S. Bellemin-Laponnaz, ChemCatChem 2009, 1, 357.
[4] a) K. Narasaka, H. Kusama, Y. Hiyashi, Tetrahedron 1992, 48,
2059; b) S. Bellemin-Laponnaz, H. Gisie, J. P. LeNy, J. A.
Osborn, Angew. Chem. 1997, 109, 1011; Angew. Chem. Int. Ed.
Engl. 1997, 36, 976; c) C. Morrill, R. H. Grubbs, J. Am. Chem.
Soc. 2006, 128, 8142; d) B. M. Trost, F. D. Toste, J. Am. Chem.
Soc. 2000, 122, 11262; e) J. M. Hutchison, H. A. Lindsay, S. S.
Dormi, G. D. Jones, D. A. Vivic, M. C. McIntosh, Org. Lett. 2006,
8, 3663; f) B. M. Trost, D. Amans, W. M. Seganish, C. K. Chung,
Chem. Eur. J. 2012, 18, 2961.
[5] P. Ghorai, P. H. Dussault, Org. Lett. 2008, 10, 4577.
[6] For processes that are related to the termination step, see:
a) P. A. Evans, J. Cui, S. J. Gharpure, R. J. Hinkle, J. Am. Chem.
Soc. 2003, 125, 11456; b) P. A. Evans, J. Cui, S. J. Gharpure, Org.
Lett. 2003, 5, 3883; c) R. J. Hinkle, Y. Lian, N. D. Litvinas, A. T.
Jenkins, D. C. Burnette, Tetrahedron 2005, 61, 11679.
[7] W. Sittiwong, M. W. Richardson, C. E. Schiaffo, T. J. Fisher, P. H.
Dussault, Beilstein J. Org. Chem. 2013, 9, 1526.
[8] a) C. G. Kruse, E. K. Poels, F. L. Jonkers, A. Vandergen, J. Org.
Chem. 1978, 43, 3548; b) M. E. Green, J. C. Rech, P. E.
Floreancig, Angew. Chem. 2008, 120, 7427; Angew. Chem. Int.
Ed. 2008, 47, 7317.
[9] a) M. D. Lewis, J. K. Cha, Y. Kishi, J. Am. Chem. Soc. 1982, 104,
4976; b) C. H. Larsen, B. H. Ridgway, J. T. Shaw, K. A. Woerpel,
J. Am. Chem. Soc. 1999, 121, 12208; c) C. H. Larsen, B. H.
Ridgway, J. T. Shaw, D. M. Smith, K. A. Woerpel, J. Am. Chem.
Soc. 2005, 127, 10879.
[10] I. Vilotijevic, T. F. Jamison, Angew. Chem. 2009, 121, 5352;
Angew. Chem. Int. Ed. 2009, 48, 5250.
[11] S. Darses, J.-P. Genet, Chem. Rev. 2008, 108, 288.
[12] a) A. A. Rodriguez, H. Yoo, J. W. Ziller, K. J. Shea, Tetrahedron
Lett. 2009, 50, 6830; b) P. B. Cranwell, J. R. Hiscock, C. J. E.
Haynes, M. E. Light, N. J. Wells, P. A. Gale, Chem. Commun.
2013, 49, 874.
[13] a) H. Xu, J. Zuend, M. G. Wall, Y. Tao, E. N. Jacobsen, Science
2010, 327, 986; b) A. Borovika, P.-I. Tang, S. Klapman, P.
Nagorny, Angew. Chem. 2013, 125, 13666; Angew. Chem. Int. Ed.
2013, 52, 13424.
Received: February 1, 2014
Published online: April 7, 2014
[14] M. Horn, L. H. Schapple, G. Lang-Wittkowski, H. Mayr, A. R.
Ofial, Chem. Eur. J. 2013, 19, 249.
[15] D. M. Smith, M. B. Tran, K. A. Woerpel, J. Am. Chem. Soc. 2003,
125, 14149.
Keywords: allylic compounds · carbocations ·
diastereoselectivity · isomerization · oxygen heterocycles
.
Angew. Chem. Int. Ed. 2014, 53, 4926 –4929
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4929