Unusual totally selective cyclodimerization of epoxides: Synthesis of a pair of diastereoisomers of enantiopure 2,5-disubstituted-1,4-dioxanes with C2 symmetry

Article abstract of DOI:10.1002/adsc.200700486

The synthesis of the C₂-symmetrical (2R,5R)- and (2S,5S)-2,5-bis-[(S)-1-(dibenzylaminoalkyl)]-1,4-dioxanes 1 or 2 in enantiopure form is reported. Compounds 1 and 2 were obtained by a completely selective and unusual cyclodimerization of chiral (2R,1′S)- or (2S,1′S)-2-(1- aminoalkyl)epoxides 3 or 4 promoted by a mixture of diisopropylamine and boron trifluoride-diethyl etherate complex. The structure of the obtained dioxane was established by single-crystal X-ray diffraction analysis. A mechanism has been proposed to explain this transformation.

Full text of DOI:10.1002/adsc.200700486

FULL PAPERS
DOI: 10.1002/adsc.200700486
Unusual Totally Selective Cyclodimerization of Epoxides:
Synthesis of a Pair of Diastereoisomers of Enantiopure
2,5-Disubstituted-1,4-Dioxanes with C2 Symmetry
JosØ M. Concellón,a,* Pablo L. Bernad,a Virginia del Solar,a
Santiago García-Granda,b and and M. Rosario Díazb
a
Departamento de Química Orgµnica e Inorgµnica, Facultad de Química, Universidad de Oviedo, 33071 Oviedo, Spain
Fax : (+34)-98-510-3446; e-mail: jmcg@uniovi.es
Departamento de Química Física y Analítica, Facultad de Química, Universidad de Oviedo, 33071 Oviedo, Spain
b
Received: October 4, 2007; Published online: February 11, 2008
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The synthesis of the C2-symmetrical plex. The structure of the obtained dioxane was es-
(2R,5R)- and (2S,5S)-2,5-bis-[(S)-1-(dibenzylami- tablished by single-crystal X-ray diffraction analysis.
noalkyl)]-1,4-dioxanes 1 or 2 in enantiopure form is A mechanism has been proposed to explain this
reported. Compounds 1 and 2 were obtained by a transformation.
completely selective and unusual cyclodimerization
of chiral (2R,1’S)- or (2S,1’S)-2-(1-aminoalkyl)epox-
ides 3 or 4 promoted by a mixture of diisopropyl- Keywords: amino epoxides; diastereoselectivity; di-
amine and boron trifluoride·diethyl etherate com- merization; dioxanes; enantiopurity
Introduction
in the literature.[10] Consequently, the methods to
obtain 1,4-dioxanes in enantiomerically enriched form
Epoxides are very important building blocks in organ- are even more scarce,[11] and no general preparation
ic synthesis and consequently an important number of exists, apart from isolated reports based on: a) oxyse-
reviews have been published on this topic.[1] The syn- lenylation of dienes with enantiopure diols;[11,12] b)
thetic applications of epoxides are based on their ver- Mitsunobu cyclization of diols;[13] c) photoinduction
satility. Thus, oxiranes can be opened by an important electron transfer (PET) cyclization of an appropriate
number of nucleophiles, can be transformed into alde- diene;[14] d) condensation of glyoxalic acid with chiral
hydes, or alkenes, as well as other important function- hydrobenzoin,[15] or e) intramolecular ring opening of
al groups. However, to the best of our knowledge, no oxirane by an alcohol moiety.[16] Therefore, a general
cyclodimerization of epoxides to produce 1,4-dioxanes method to obtain enantiopure 1,4-dioxanes with C2
has been reported to date. This unusual cyclodimeri- symmetry would be desirable, due to their synthetic
zation is very interesting, not only due to its novelty, utility,[17] especially when the two diastereoisomers
but also because of the important practical applica- could be accessed with complete selectivity.
tions of the 1,4-dioxanes generated. Hence, they are
Previously, we reported the efficient synthesis of
used in the manufacture of polycarbonates with pho- both enantiopure (2R,1’S)- or (2S,1’S)-2-(1-aminoal-
toreceptor properties,[2] cosmetic products,[3] flame re- kyl)epoxides, by total stereoselective reduction of
tardant additives for flammable plastics,[4] insecticides, enantiopure a-amino-a’-chloro ketones with LiAlH4
acaricides and ectoparasiticides,[5] or germicides and or by a highly stereoselective addition reaction of io-
fungicides.[6] In addition, some of these also offer bio- domethyllithium to chiral 2-amino aldehydes, respec-
medical applications[7] such as, for example, inhibitors tively.[18] More recently, we described various transfor-
of epidermal growth factor receptor tyrosine kinase,[8] mations of these amino epoxides into a variety of
or as antiobesity agents.[9]
enantiopure compounds, based on the highly regiose-
In spite of the significant applications of 1,4-diox- lective nucleophilic opening of the oxirane ring. Thus,
anes, it is surprising that very few methods for the we reported the transformation of the oxirane ring
synthesis of racemic 1,4-dioxanes have been reported into 1,3-dioxolanes,[19] 1,3-diaminoalkan-2-ols,[20] O1-
Adv. Synth. Catal. 2008, 350, 477 – 481
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
477

JosØ M. Concellón et al.
FULL PAPERS
acyl-3-aminoalkane-1,2-diols,[21] b-amino alcohols,[22]
and 1-alkylthio-3-aminoalkan-2-ols[23] by reaction with
ketones, nitriles, carboxylic acids, organolithium com-
pounds or thiols, respectively.
We now report the novel cyclodimerization reaction
of the oxirane ring of the chiral (2R,1’S)- or (2S,1’S)-
2-(1-aminoalkyl)epoxides 3 or 4, promoted by a mix-
ture of diisopropylamine and BF3·OEt2 in THF, to
afford 2,5-bis-[(S)-1-(dibenzylamino)alkyl)]-(2R,5R)-
1,4-dioxane 1 or its 2S,5S-diastereoisomer 2 in enan-
tiomerically pure form and with total selectivity. Inter-
estingly, from the synthetic applications viewpoint,
the 1,4-dioxanes obtained present a C2-axis of symme-
try. The proposed structures of the reported com-
pounds were established based on single-crystal X-ray
diffraction analysis of 1b. A mechanism has been pro-
posed to explain this unusual transformation.
Scheme 1. Synthesis of enantiopure 1,4-dioxanes 1 and 2.
Table 1. Synthesis of enantiopure 1,4-dioxanes 1 and 2.
Entry 1 or 2 R
t [h][a] dr[b]
Yield [%][c]
1
2
3
4
5
6
7
1a
1b
1c
1d
2a
2b
2c
Me
i-Bu
Bn
BnOCH2
Me
i-Bu
Bn
5
5
1
5
5
5
2
> 99/1
73
80
81
72
Results and Discussion
> 99/1
> 99/1
> 99/1
> 99/1 (99/1) 70
95/5 (95.5/4.5) 73
96.5/3.5 (96/4) 75
The first experiments were performed using the syn-
aminoepoxide 3b as model substrate. Thus, a solution
of 3b in toluene was treated with i-Pr2NEt and
BF3·OEt at room temperature for 12 h. The structure
of the obtained major product 1b (53% yield) could
[a]
[b]
Reaction time.
not be established from the 13C and H NMR spectro-
1
1
Diastereoisomeric ratios determined by H NMR analysis
of the crude products 1 or 2. The dr values of the starting
amino epoxides 4 are given in brackets.
Isolated yields after column chromatography based on
the starting amino epoxide 3 or 4.
scopic data. As it was expected, the spectra of 1b
were different to those of the starting amino epoxide
3b, but the number of signals of the starting and final
compound was the same. The NMR spectra of 1b
showed the presence of one dibenzylamino group and
two signals of a CH and CH2 groups (DEPT experi-
ment) at d=75.7 and 65.8 ppm, which could be as-
signed to CHO and CH2O. In addition the analysis of
1b by mass spectrometry showed a peak for M+ near
to twice that of the starting amino epoxide 3b. These
data could suggest that the NMR spectra of 1b
showed only half of their signals, which in turn, could
indicate that 1b was a meso or a C2-symmetrical dia-
stereoisomer.
[c]
To improve the synthesis of 1b, various different re-
action conditions were tested. The best result was ob-
tained by treatment of a solution of 3b in THF (in-
stead of toluene) with diisopropylamine (instead di-
A
ture for 5 h. Hence, compound 1b was obtained in
80% yield as single isomer, after column chromato-
graphic purification (Scheme 1 and Table 1).
Figure 1. ORTEP drawing of structure 1b.[25]
Product 1b could be crystallized and its correspond-
ing X-ray analysis (Figure 1) allowed the unambigu- absolute configurations of the stereogenic centers
ous assignment of the structure of 1b, as depicted in were the same as those in its precursor 3b.
Scheme 1.[24] This structure comprises a C2 axis of
To test the generality of this unusual cyclodimeriza-
symmetry, explaining that only half of their carbons tion reaction, syn amino epoxides 3a, 3c–d and anti-
1
and protons were seen in its H and 13C NMR spectra. amino epoxides 4 were allowed to react under the
In addition the structure of 1b demonstrated that the same reaction conditions, obtaining the corresponding
dimers 1 and 2 in high yield (Scheme 1 and Table 1).
478
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 477 – 481

Unusual Totally Selective Cyclodimerization of Epoxides
FULL PAPERS
The spectroscopic data (NMR and IR) of com-
pounds 1a, 1c–d and 2 were closely related to those of
1b and the analyses of compounds 1a, c, d and 2a–c
by high- and low-resolution mass spectrometry were
also in accord with the corresponding dioxanes 1 and
2. Therefore, the structures of 1 and 2 (Scheme 1)
were assigned by anology to 1b. An analysis of
Table 1 reveals that no important differences in this
cyclodimerization reaction of epoxides were observed
as a consequence of structural differences on the
starting amino epoxides 3 or to the change of the ste-
reochemistry in the starting amino epoxides when
amino epoxides 4 were used as starting materials in-
stead of 3, under the same reaction conditions.
It was noteworthy that no dry solvents were used
and no inert atmosphere was necessary to perform
the reaction.
Determinations of the diastereoisomeric purity of
1
products 1 and 2 were performed by GC/MS and H
and 13C NMR analysis on the crude reaction products.
Hence, it could be established that, from diastereo-
pure amino epoxides 3,[18] all dioxanes 1 were ob-
tained as single diastereoisomers, whereas compounds
2 were obtained as a mixture of diastereoisomers in
the same relationship to that of the starting amino ep-
oxides 4.[18] This fact could be an indirect support of
the total selectivity of the reaction. The major diaste-
reoisomer 2 was isolated as a single diastereoisomer,
after easy purification of the crude reaction products
by column chromatography.
Scheme 2. Proposed mechanism.
carbon configuration, affording (after BF3 dissocia-
To explain the transformations of amino epoxides 3 tion) the intermediate 8. The final 1,4-dioxane 1 could
and 4 into products 1 and 2, respectively, we tenta- be obtained by a cyclization of 8. A similar mecha-
tively propose a mechanism based on the following nism could explain the transformation of 4 into 2.
experimental results: a) when 2-cyclohexyloxirane
Taking into account that no dimerization took
was treated under the same reaction conditions, the place from the oxide of cyclohexene, alternative
corresponding amino alcohol, derived from the open- mechanisms without involvement of the dibenzylami-
ing of the oxirane ring by the amine at the C-3 posi- no group should be rejected. Thus, the reaction of the
tion, instead of the dimer, was obtained. This result epoxide oxygen as a nucleophile with the terminal
could suggest that the dibenzylamino group of the carbon of a second equivalent of an epoxide activated
starting epoxides 3 or 4 could have an essential role by BF3 and followed by cyclization of the obtained
in the transformation. b) The presence of both re- alkoxide from epoxide opening onto the terminal
agents, diisopropylamine and BF3·Et2O, was essential carbon of the O-alkylated epoxonium ion is not sup-
to perform the transformations of 3 into 1 or 4 into 2. ported by our experimental results.
In the absence of amine or BF3·Et2O, unchanged
amino epoxides 3 or 4 were fully recovered.
So, initially the oxygen of the amino epoxide 3 Conclusions
could selectively react with the Lewis acid, BF3,[26] to
give the activated oxirane 5 (Scheme 2). Then, the di- We have reported the unusual cyclodimerization of
benzylamino group could open the activated epoxide chiral (2R,1’S)- or (2S,1’S)-2-(1-aminoalkyl)epoxides 3
at C-2 with inversion of its configuration, generating or 4 to afford 2,5-bis-[(S)-1-(dibenzylaminoalkyl)]-
an aziridinium salt 6. The intermediate 6, in which the (2R,5R)-1,4-dioxanes 1 or the 2S,5S-diastereoisomers
oxygen is coordinated to the Lewis acid, and the base 2, both in enantiopure form and with C2 symmetry.
could be in equilibrium with an aziridinium inter- The transformation took place with complete selectiv-
mediate 7 (without the Lewis acid) and the BF3·amine ity, by treatment of 3 or 4 with i-Pr2NH and BF3·Et2O
salt.[27] Then, the alcoholate group of 7 could open the in THF at room temperature. The structure of com-
aziridinium ring of a second equivalent of the inter- pound 1b was established by single-crystal X-ray dif-
mediate 6 at C-2,[28] with a second inversion of the fraction analysis and the corresponding mechanism
Adv. Synth. Catal. 2008, 350, 477 – 481
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
479

JosØ M. Concellón et al.
FULL PAPERS
[4] W. Himmele, H. Mueller-Tamm, D. Wolff, A. Hof-
mann, German Patent 1,961,365, 1969; Chem. Abstr.
1971, 75, 77766.
[5] F. Karrer, European Patent 88–810011, 1988; Chem.
Abstr. 1978, 109, 190402.
has been proposed to explain this transformation.
Further studies directed toward the development of
the synthetic applications of the enantiopure substi-
tuted 1,4-dioxanes prepared are currently under in-
vestigation within in our laboratory.
[6] J. D. McClure, US Patent 19,630,313, 1964; Chem.
Abstr. 1964, 61, 47918.
[7] a) H. Y. Kim, R. J. Kuhn, C. Patkar, R. Warrier, M.
Cushman, Bioorg. Med. Chem. 2007, 15, 2667–2679;
b) J. A. L. Simond, A. J. C. Monteil, P. Carlier, French
Patent 2,547,582, 1984; Chem. Abstr. 1985, 103, 54086.
[8] J. Y. Lee, Y. K. Park, S. H. Seo, B. S. Yang, H. Park,
Y. S. Lee, Arch. Pharm. 2002, 335, 487–494.
Experimental Section
General Procedure for the Synthesis of 1 or 2
To a stirred solution of the corresponding amino epoxide 3
or 4 (0.3 mmol), in THF (2 mL), BF3·OEt2 (1.1 equiv,
0.33 mmol, 0.041 mL) and diisopropylamine (2 equiv.
0.6 mmol, 0.085 mL) were successively added at room tem-
perature. After stirring at this temperature for the time indi-
cated in Table 1, the mixture was neutralized with an aque-
ous saturated solution of NaHCO3 (5 mL). The aqueous
phase was extracted with diethyl ether (3”5 mL), and the
combined organic layers were dried over anhydrous Na2SO4,
filtered and concentrated under vacuum. Flash column chro-
matography on silica gel (hexane/EtOAc 10:1) provided
pure compounds 1 or 2.
[9] A. M. Verkatesan, World Patent 9,829,405, 1998;
Chem. Abstr. 1998, 129, 109091.
[10] a) K. Baum, J. Am. Chem. Soc. 1968, 90, 7083–7089;
b) H. Sugimoto, D. T. Sawyer, J. Org. Chem. 1985, 50,
1784–1786; c) A. Braun, L. Toupet, J.-P. Lellouche, J.
Org. Chem. 1996, 61, 1914–1915; d) R. Martínez, M.
Velasco, I. Martínez, I. Menconi, A. Ramírez, E. An-
geles, I. Regla, R. López, J. Heterocyclic Chem. 1997,
34, 1865–1866; e) L. Weber, M. Barlmeyer, J.-M. Quas-
dorff, H. L. Sievers, H.-G. Stammler, B. Neumann, Or-
ganometallics 1999, 18, 2497–2504.
Analytical and spectroscopic data for products 1 and 2
and the general remarks about the equipment used made
are available in the Supporting Information.
[11] M. Tiecco, L. Testaferri, F. Marini, S. Sternativo, C.
Santi, L. Bagnoli, A. Temperini, Tetrahedron: Asymme-
try 2003, 14, 1095–1102.
[12] K. S. Kim, J. II Park, P. Ding, Tetrahedron Lett. 1998,
39, 6471–6474.
[13] M. C. Wilkinson, R. Bell, R. Landon, P. O. Nikiforov,
A. J. Walker, Synlett 2006, 2151–2153.
[14] G. Pandey, A. L. Gaikwad, S. R. Gadre, Tetrahedron
Acknowledgements
Lett. 2006, 47, 701–703.
We thank Ministerio de Ciencia y Tecnología (CTQ2004–
1191/BQU and MAT2006–01997) for financial support and
Euan Goddard for his revision of the English. J. M. C.
thanks Carmen Fernµndez-Flórez for her time. V del S. is
grateful to Principado de Asturias for a research contract.
[15] H. Fujioka, N. Matsunaga, H. Kitagawa, Y. Nagatomi,
M. Kondo, Y. Kita, Tetrahedron: Asymmetry 1995, 6,
2117–2120; H. Fujioka, N. Matsunaga, H. Kitagawa, Y.
Nagatomi, M. Kondo, Y. Kita, Tetrahedron: Asymmetry
1995, 6, 2113–2116.
[16] J. AubØ, C. J. Mossman, S. Dickey, Tetrahedron 1992,
48, 9819–9826.
[17] For synthetic applications of compounds with C2 sym-
metry: a) M. Vrettou, A. A. Gray, A. R. E. Brewer,
A. G. M. Barrett, Tetrahedron 2007, 63, 1487–1536;
b) K. C. Bhowmick, N. N. Joshi, Tetrahedron: Asymme-
try 2006, 17, 1901–1929; c) G. Desimoni, G. Faita,
K. A. Jorgensen, Chem. Rev. 2006, 106, 3561–3651;
d) S. Castrillón, C. Claver, Y. Diaz, Chem. Soc. Rev.
2005, 34, 702–713; e) B. Delouvrie, L. Fensterbank, F.
Najera, M. Malacria, Eur. J. Org. Chem. 2002, 67,
3507–3525; f) A. K. Ghosh, P. Mathivanan, J.; Cappiel-
lo, Tetrahedron: Asymmetry 1998, 9, 1–45.
References
[1] Reviews on the synthesis and synthetic applications of
epoxides: a) D. M. Hodgson, P. G. Humphreys, S. P.
Hughes, Pure Appl. Chem. 2007, 79, 269–279; b) E. M.
Mcgarrigle, D. G. Gilheany, Chem. Rev. 2005, 105,
1563–1602; c) V. K. Aggarwal, C. L. Winn, Acc. Chem.
Res. 2004, 37, 611–620; d) R. A. Sheldon, Applied Ho-
mogeneous Catalysis with Organometallic Compounds,
Wiley-VCH, Weinheim, 2002; e) L. A. Campbell, T.
Kodadek, J. Mol. Catal. A: Chemical 1996, 113, 293–
310; f) D. Schinzer, Organic Synthesis Highlights II,
Wiley-VCH, Weinheim, 1996; g) G. A. Barf, R. A.
Sheldon, J. Mol. Catal. A: Chemical 1995, 102, 23–39;
h) J. P. Collman, X. Zhang, V. J. Lee, E. S. Uffelman,
J. I. Brauman, Science 1993, 261, 1404–1411.
[18] a) J. Barluenga, B. BaragaÇa, A. Alonso, J. M. Concel-
lón, J. Chem. Soc., Chem. Commun. 1994, 969–970;
b) J. Barluenga, B. BaragaÇa, J. M. Concellón, J. Org.
Chem. 1995, 60, 6696–6699.
[19] J. M. Concellón, J. R. Suµrez, S. García-Granda, M. R.
Díaz, Org. Lett. 2005, 7, 247–250.
[20] J. M. Concellón, J. R. Suµrez, V. del Solar, J. Org.
Chem. 2005, 70, 7447–7450.
[2] I. Takekawa, H. Nukata, H. Nakamura, M. Miyamoto,
Japanese Patent 10,265,564, 1998; Chem. Abstr. 1998,
129, 337617.
[3] J. Conrad, U. A. Schaper, K. Bruns, German Patent
2,629,000, 1978; Chem. Abstr. 1978, 89, 80129.
[21] J. M. Concellón, J. R. Suµrez, V. del Solar, R. Llavona,
J. Org. Chem. 2005, 70, 10348–10353.
480
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 477 – 481

Unusual Totally Selective Cyclodimerization of Epoxides
FULL PAPERS
[22] J. M. Concellón, P. L. Bernad, V. del Solar, J. R. Suµrez,
S. García-Granda, M. R. Díaz, J Org Chem. 2006, 71,
6420–6426.
[23] J. M. Concellón, V. del Solar, J. R. Suµrez, E. G.
Blanco, Tetrahedron 2007, 63, 2805–2810.
[24] CCDC 654081 (1b) contains the supplementary crystal-
lographic data for this paper. They can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif
(or the Cambridge Data Centre, 12, Union Road.
Cambridge CB21EZ, UK: fax: (+44)-1223-336-033; or
deposit@ccdc.cam.ac.uk.
[25] For clarity, only some relevant atoms are labelled.
[26] The coordination of the Lewis acid with the oxygen of
the epoxide is favoured in comparison with the nitro-
gen of the dibenzylamino group due to steric hindrance
of the dibenzylamino group, and it has previously been
observed in other ring openings of amino epoxides 3 or
4: see ref.[18].
[27] The synthesis of a mixture of products when the reac-
tion was performed under the same reaction conditions,
but using a tertiary amine (i-Pr2NEt) instead of i-
Pr2NH, could support the existence of this equilibrium,
due to the coordination of a tertiary amine with a
Lewis acid is most difficult than a secondary amine.
[28] A ring-opening of aziridinium 6 at C-3·could be also
possible. However the obtained intermediate would
allow access to a 7-membered cycle, which it is disfav-
oured with respect to the formation of the 6-membered
cycle such as dioxane. So, the hypothetical formation of
this intermediate would be reversible and it would be
newly transformed into intermediates 6 and 7.
Adv. Synth. Catal. 2008, 350, 477 – 481
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
481