Mechanistic insights into the rearrangement of azetidine N-oxides to isoxazolidines

Article abstract of DOI:10.1016/j.tetlet.2012.06.092

Various functionalized azetidines were oxidized with mCPBA or hydrogen peroxide, to produce the corresponding N-oxide and study its fate in the [1,2] Meisenheimer rearrangement. This ring expansion leading to isoxazolidines occurs readily, without trapping of the transient N-oxide. Starting with azetidines bearing a nitrile or an ester group at C-2, the rearrangement is regioselective. However, a varying amount of epimerization on the migrating radical is observed, which can also be observed with the related [1,2] Stevens rearrangement.

Full text of DOI:10.1016/j.tetlet.2012.06.092

Tetrahedron Letters 53 (2012) 4697–4699
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Mechanistic insights into the rearrangement of azetidine N-oxides to
isoxazolidines
⇑
Laurence Menguy, Bruno Drouillat, Jérome Marrot, François Couty
Institut Lavoisier, Université de Versailles St. Quentin-en-Yvelines, UMR 8180, 45 Avenue des Etats-Unis, 78035 Versailles, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Various functionalized azetidines were oxidized with mCPBA or hydrogen peroxide, to produce the cor-
responding N-oxide and study its fate in the [1,2] Meisenheimer rearrangement. This ring expansion
leading to isoxazolidines occurs readily, without trapping of the transient N-oxide. Starting with azeti-
dines bearing a nitrile or an ester group at C-2, the rearrangement is regioselective. However, a varying
amount of epimerization on the migrating radical is observed, which can also be observed with the
related [1,2] Stevens rearrangement.
Received 11 May 2012
Revised 14 June 2012
Accepted 18 June 2012
Available online 28 June 2012
Keywords:
Azetidines
Ó 2012 Elsevier Ltd. All rights reserved.
[1,2] Meisenheimer rearrangement
Isoxazolidines
N-oxides
The [1,2] Meisenheimer rearrangement of amine N-oxides is a
reaction that has found some applications in the synthesis of alka-
loids.1 When carried out with cyclic amines, this thermally induced
reaction produces the corresponding ring expanded product
including an oxygen atom. Together with its vinyl homologue that
is, the [2,3] Meisenheimer rearrangement, for which asymmetric
versions have appeared,2 its mechanism has been thoroughly
examined both experimentally3 and, more recently, in silico.4 It
is now well-established that this reaction goes through a diradical
species, produced via the homolytic cleavage of the C–N bond, that
further recombines to form the C–O bond (Fig. 1).
carbon radical recombines in the solvent cage mainly with reten-
tion of configuration5 (Fig. 2).
However, this issue of stereoselectivity is different from the
Meisenheimer rearrangement, for which Schöllkopf3a earlier dem-
onstrated that, when starting with optically enriched
a-deutero-
N,N-dimethylbenzylamine oxide, the deuterated benzylic radical
migrates with ca. 70% of racemization, and the ‘cage effect’ which
accounts for the stereospecificity of the recombination was later on
estimated to be 40%.3c This is indeed a major parameter for this
reaction, but which has been very rarely studied. Isolated examples
in cyclic systems allowing a quantitative measurement of the
amount of retention of configuration for the recombination step
are very scarce.1
This rearrangement is quite similar to the Stevens rearrange-
ment, for which it has been demonstrated that the migrating
Azetidines, that is, four-membered ring nitrogen heterocycles
are ideal substrates for studies directed to this rearrangement be-
cause the ring strain present in these heterocycles facilitates the
rate limiting step, that is, the homolytic bond cleavage. Further-
more, the produced isoxazolidines, usually prepared by cycloaddi-
tions,6 are of high synthetic interest. The first report of the
Diradical intermediate
Homolytic
cleavage
N
N
N
R
O
R
O
R
O
Diradical intermediate
Homolytic
[2,3]
cleavage
Recombination
[1,2]
R"
N
N
R"
R"
N
O
R
R
O
R
N
R
N
R
R'
R'
R'
R"
Recombination
Retention of configuration
Figure 1. [1,2] and [2,3] Meisenheimer rearrangement of amine N-oxides.
R'
N
R
⇑
Corresponding author. Fax: +33 (0) 1 39 25 44 52.
E-mail address: couty@chimie.uvsq.fr (F. Couty).
Figure 2. Stevens rearrangement occurs with retention of configuration.
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.06.092

4698
L. Menguy et al. / Tetrahedron Letters 53 (2012) 4697–4699
Screening of the conditions for the oxidation of the nitrogen in-
R
cluded mCPBA with Na2CO3,8 hydrogen peroxide with sodium
hydroxide,17 hydrogen peroxide with NaHCO3 in the presence of
Na2WO4, 2H2O catalyst,18 or TFAA, K2CO3, UHP.19 It appeared that
in no case, starting with these substrates and under these condi-
tions, we have been able to isolate or characterize by NMR the
intermediate azetidine N-oxide. As a matter of fact, azetidine
N-oxides have been isolated in previous work in isolated cases7c
and proved stable enough only if stabilization by intramolecular
H-bonding7a,20 is possible. Here, the produced N-oxide undergoes
a rapid [1,2] Meisenheimer rearrangement to give the correspond-
ing isoxazolidine, but isolated yields are usually low due to incom-
plete oxidation of the starting material and competing oxidation of
the produced isoxazolidine leading to polar nitrones. The latter
could be identified by a signal around 6.9 ppm in 1H NMR of the
crude reaction mixture.8 For azetidine 1, the best conditions were
identified as hydrogen peroxide in the presence of Na2WO4, 2H2O
catalyst and gave N-benzhydryl isoxazolidine 9 in 80% isolated
yield. We next realized that presence of a nitrile or an ester in
the azetidine substrate precluded the use of hydrogen peroxide
in the basic medium as the oxidant, since, for example, isoxazolid-
inyl amide 10, in which the nitrile had been hydrolyzed, was
[2,3]
[1,2]
O
R
N
Cy
O
O
N
Cy
R
R
Cy
N
O
N
Cy
Figure 3. Importance of the configuration of the N-oxide, when [1,2] and [2,3]
Meisenheimer rearrangements can compete.
Meisenheimer rearrangement in azetidine N-oxides came from
Kurihara et al.,7 who noticed their ease and applied this reaction
to the synthesis of alkaloids. Later on, they engaged in a study8
on the [2,3] Meisenheimer rearrangement of 2-alkenyl azetidine
N-oxides and pointed out the importance of the configuration of
the N-oxide on the outcome of the reaction. Thus, when the oxygen
is cis to the adjacent alkenyl moiety, then [2,3] Meisenheimer rear-
rangement occurs, but when it is trans, then exclusive [1,2] rear-
rangement is observed (Fig. 3). We have noticed
a similar
behavior in the Stevens rearrangement of such systems, which af-
ford unsaturated azocanes through [2,3] rearrangement.9
In this Letter, in continuation with our interest in the ring
expansion and ring cleavage of azetidines,10 we describe further
studies on [1,2] Meisenheimer rearrangement of azetidines,
addressing the issues of regioselectivity for the cleavage step,
and stereoselectivity for the produced isoxazolidine.
For this study, we selected an array of diversely substituted
azetidines 1–8 shown in Figure 4. These compounds were chosen
to answer the following questions: (i) is this ring expansion feasi-
ble with unsubstituted azetidines (substrate 1) and (ii) does it oc-
cur regioselectively when a nitrile or an ester is present at C-2
(substrate 211 and 312)? In the case of regioselective ring opening
at C-2, substrates 4–5,13 6,14 7,15 and 816 were selected to study
the issues of diastereoselectivity.
CN
mCPBA, Na2CO3
O
DCM
11
N
2
54%
Ph
CO2Et
12
Id
3
O
34%
N
Ph
CN
CN
O
Ph
Ph
Id
+
CN
4
(9%)
COOEt
+
13b
O
4
N
N
N
13a
N
N
1
Ph
2
Ph
3
Ph
(33%)
Me
Me
Ph
13a 13b
:
= 3 : 1
Ph
Ph
N
Me
Ph
Ph
CN
O
CN
O
N
Me
Me
Me
CN
Me
6
COOEt
Me
CN
Ph
Ph
N
Bn
Id
N
Me
4
7
5
+
5
13a
13b
N
(30%)
Me
13a : 13b = 1 : 7
CN
Ph
COOEt
N
N
Me
CO2Et
8
Me
CO2Et
Ph
Id
OH
6
O
15
+
14
N
(23%)
N
(37%)
Figure 4. Structure of selected azetidines 1–8.
Ph Bn
Bn
CO2Et
CO2Et
Ph
OH
+
Id
H2O2, NaHCO3
MeOH/CH3CN
Na2WO4(2H2O) cat.
17
O
N
7
(8%)
N
16
Me
: dr 30%
(18%)
Ph Me
O
9
N
1
CN
O
80%
Ph
Ph
Id
CONH2
8
18
: dr 66%
N
H2O2, NaOH, MeOH
15%
10
(16%)
2
O
Ph
N
Ph
Scheme 2. Reactions of azetidines with mCPBA. Yields in brackets refer to pure
isolated compounds. Ratios of isomers were determined by 1H NMR on the crude
reaction mixture.
Scheme 1. Reactions with hydrogen peroxide.

L. Menguy et al. / Tetrahedron Letters 53 (2012) 4697–4699
4699
Acknowledgments
EtO C
O
N
2
mCPBA
H
H
H
Ph
N
15
Ph
CO Et
2
Me
University of Versailles St-Quentin-en-Yvelines and CNRS are
acknowledged for financial support.
Bn
6
Bn
Figure 5. Competing Cope elimination in N-oxide derived from azetidine 6.
Supplementary data
Supplementary data (experimental procedures and character-
ization data for compounds 9–18. Crystallographic data for isoxa-
zolidine 1) associated with this article can be found, in the online
version, at http://dx.doi.org/10.1016/j.tetlet.2012.06.092.
produced in low 15% isolated yield starting from 2 (Scheme 1).
Structure of 10 was ascertained by X-ray crystallography.21
With functionalized substrates 2–8 we therefore switched to
mCPBA oxidations and results are shown in Scheme 2.
This set of experiments allows to answer all the questions ini-
tially programmed. First, this rearrangement takes place very rap-
idly and efficiently with an unsubstituted azetidine: even though
the intermediate primary carbon radical should be less stabilized,
we could not observe any transient N-oxide when the oxidation
of 1 leading to 9 with mCPBA was carried out in a NMR tube. Then,
the ring cleavage is highly regioselective when an ester or nitrile is
present at C-2, which is in agreement with the stabilization of the
References and notes
1. (a) Rayburn, C. H.; Harlan, W. R.; Hanmer, H. R. J. Am. Chem. Soc. 1950, 72,
1721–1723; (b) Malpass, J. R.; Skerry, P. S.; Rimmington, S. L. Heterocycles
2004, 62, 679–691; (c) Didier, C.; Critcher, D. J.; Walshe, N. D.; Kojima, Y.;
Yamauchi, Y.; Barrett, A. G. M. J. Org. Chem. 2004, 69, 7875–7879; (d)
Commandeur, C.; Commandeur, M.; Bathany, K.; Kaufmann, B.; Edmunds, A.
J. F.; Maienfisch, P.; Ghosez, L. Tetrahedron 2011, 67, 9899–9908.
2. (a) Davies, S. G.; Smyth, G. D. J. Chem. Soc., Perkin Trans. 1 1996, 2467–2477; (b)
Buston, J. E. H.; Coldham, I.; Mulholland, K. R Tetrahedron: Asymmetry 1998, 9,
1995–2009; (c) Blanchet, J.; Bonin, M.; Micouin, L.; Husson, H-P. Tetrahedron
Lett. 2000, 41, 8279–8283.
intermediate
a-cyano or a-ethoxycarbonyl intermediate radical.
Finally, the configuration of the migrating carbon stereocenter is
retained only in one case, starting from azetidine 6. Starting from
diastereoisomers 4 and 5, up to 25% of epimerization is observed,
but the ratio of epimers depends on the configuration in the start-
ing material. This suggests a partial retention of configuration,
though relative configuration in isoxazolidines 13a and 13b could
not be clearly ascertained by NOE experiments. This degree of epi-
merization also depends on the nature of the substituent on the
radical intermediate (nitrile or ester), since no epimerization is ob-
served starting from azetidine ester 6. Furthermore, the nature of
the nitrogen substituent is also a crucial parameter, since extensive
epimerization is observed starting from N-Me azetidine 7, while no
epimerization is observed starting with N-Bn analogue 6. Note that
in the former case, major epimer retains the stereochemistry of the
starting azetidine based on the examination of the coupling con-
stants (see Supplementary data). These experiments also allowed
to put in light a possible competing process when the N-oxide is
stereoselectively produced in a cis relationship to the 4-methyl
substituent. In this case, a Cope elimination yielding hydroxyl-
amine 15 and 17 can occur. This syn elimination is observed with
esters 6 and 7, but not with nitriles 4 and 5, suggesting that the
homolytic cleavage leading to isoxazolidine is easier in the latter
case (Fig. 5).
3. (a) Schöllkopf, U.; Patsch, M.; Shäfer, H. Tetrahedron Lett. 1964, 36, 2515–2520;
(b) Lorand, J. P.; Grant, R. W.; Samuel, P. A.; O’Connell, S. E.; Zaro, J. Tetrahedron
Lett. 1969, 46, 4087–4088; (c) Lorand, J. P.; Grant, R. W.; Samuel, P. A.;
O’Connell, S. E.; Zaro; Pilotte, J.; Wallace, R. W. J. Org. Chem. 1973, 38, 1813–
1821; Khutier, A.-H.; Al-Mallah, K.; Hanna, S. Y.; Abdulla, N-A. J. Org. Chem.
1987, 52, 1710–1713.
4. (a) Zhang, X. Comput. theor. chem. 2011, 966, 383–390; (b) Molina, J. M.; El-
Bergmi, R.; Dobado, J. A.; Portal, D. J. Org. Chem. 2000, 65, 8574–8581.
5. (a) Brewster, J. H.; Kline, M. W. J. Am. Chem. Soc. 1952, 74, 5179–5182; (b)
Schöllkopf, U.; Ludwig, U.; Ostermann, G.; Patsch, M. Tetrahedron Lett. 1969, 39,
3415–3418.
6. Brandi, A.; Cardona, F.; Cicchi, S.; Cordero, F. M.; Goti, A. Chem. Eur. J. 2009, 15,
7808–7821.
7. (a) Kurihara, T.; Ohuchi, K.; Kawamoto, M.; Harusawa, S.; Yoneda, R. Chem. Lett.
1991, 1781–1784; (b) Kuruhara, T.; Sakamoto, Y.; Takai, M.; Ohuchi, K.;
Harusawa, S.; Yoneda, R. Chem. Pharm. Bull. 1993, 41(7), 1221–1225; (c)
Yoneda, R.; Sakamoto, Y.; Oketo, Y.; Harusawa, S.; Kurihara, T. Tetrahedron
1996, 52, 14563–14576.
8. Yoneda, R.; Araki, L.; Harusawa, S.; Kurihara, T. Chem. Pharm. Bull. 1998, 46,
853–856.
9. Couty, F.; Durrat, F.; Evano, G.; Marrot, J. Eur. J. Org. Chem. 2006, 4214.
10. (a) Couty, F.; Evano, G. Synlett 2009, 3053–3064. and references cited therein;
(b) Couty, F.; Drouillat, B.; Lemée, F. Eur. J. Org. Chem. 2011, 794–801.
11. Couty, F.; David, O.; Larmanjat, B.; Marrot, J. J. Org. Chem. 2007, 72, 1058–1061.
12. Compound 3 was obtained by treating 2 with ethanol and H2SO4.
13. Agami, C.; Couty, F.; Evano, G. Tetrahedron: Asymmetry 2002, 297–302.
14. Couty, F.; Durrat, F.; Prim, D. Tetrahedron Lett. 2003, 44, 5209–5212.
15. Drouillat, B.; Couty, F.; David, O.; Evano, G.; Marrot, J. Synlett 2008, 1345–1348.
16. Couty, F.; Evano, G.; Vargas-Sanchez, M.; Bouzas, G. J. Org. Chem. 2005, 70,
9028–9031.
17. Sumitomo Chemical Company, European Patent no 0 040 896 1979.
18. Sheng, M. N.; Zajacek, J. G. J. Org. Chem. 1968, 33, 588–590.
19. Caron, S.; Do, N. M.; Sieser, J. E. Tetrahedron Lett. 2000, 41, 2299–2302.
20. O’Neil, I. A.; Potter, A. J. Chem. Commun. 1998, 1487–1488.
21. Data has been deposited at the Cambridge Crystallographic Data Centre and
allocated the deposition number CCDC 873260. See Supplementary data.
In conclusion, we have demonstrated that [1,2] Meisenheimer
rearrangement in azetidine N-oxides occurs very rapidly and regio-
selectively. The stereospecificity of the reaction is however low and
depends on several parameters, such as the nature of the interme-
diate carbon radical, and the nature of the substituents on the
azetidine ring.