Samarium diiodide promoted reduction of 3,6-dihydro-2H-1,2-oxazines: Competition of 1,4-amino alcohol formation and ring contraction to pyrrole derivatives

Article abstract of DOI:10.1002/ejoc.201201210

An approach to enantiopure 1,4-amino alcohols of type 4 by samarium diiodide mediated N-O cleavage of 3,6-dihydro-2H-1,2-oxazines 3 is presented. In several cases we observed the formation of 3-methoxypyrrole derivatives 5 as byproducts in significant amo

Full text of DOI:10.1002/ejoc.201201210

FULL PAPER
DOI: 10.1002/ejoc.201201210
Samarium Diiodide Promoted Reduction of 3,6-Dihydro-2H-1,2-oxazines:
Competition of 1,4-Amino Alcohol Formation and Ring Contraction to Pyrrole
Derivatives
Marcin Jasin´ski,[a][‡] Toshiko Watanabe,[a][‡‡] and Hans-Ulrich Reissig*[a]
Keywords: Synthetic methods / Reduction / Samarium / Cleavage reactions / Ring contraction / Pyrroles / 1,2-Oxazines
An approach to enantiopure 1,4-amino alcohols of type 4 by
samarium diiodide mediated N–O cleavage of 3,6-dihydro-
2H-1,2-oxazines 3 is presented. In several cases we observed
pyrroles 5 as side-products is rationalized by a competitive
intramolecular hydrogen shift of the initially formed ring
cleavage intermediate B to C and subsequent cyclization of
the formation of 3-methoxypyrrole derivatives 5 as byprod- aldehyde E to afford F. This pathway can be disfavoured
ucts in significant amounts. For 1,2-oxazine derivative syn-
3a, up to 27% of pyrrole 5a was isolated. The examples pre-
sented show a strong dependence of the chemoselectivity on
the structure of the precursor 1,2-oxazines. The formation of
when a higher excess of samarium diiodide was employed,
which generally provided the 1,4-amino alcohols 4 in good
yields.
pounds[8] allow the efficient synthesis of various enantio-
pure cyclic or acyclic amino polyols. Whereas reactions of
samarium diiodide with tetrahydro-2H-1,2-oxazine deriva-
tives of type 1 led, in almost all tested examples, exclusively
to the desired amino alcohols 2 in excellent yields and puri-
ties, in the case of 3,6-dihydro-2H-1,2-oxazines 3, which
bear an enol ether moiety, in addition to the expected 1,4-
amino alcohols of type 4, due to competing reactions, the
formation of pyrroles 5[9] was also observed (Scheme 1). In-
termediates 4 can also be obtained by alternative reductive
processes and are crucial in the stereodivergent syntheses of
3-methoxypyrrolidines and 3-methoxy-2,5-dihydropyr-
roles.[10] More recently, we studied transformations of 1,4-
amino alcohols 4 into polyfunctionalized furan derivatives,
which were either isolated or generated as intermediates.[11]
In the present report, we demonstrate that SmI2-promoted
reduction of precursors 3 not only provide the expected
compounds but also lead to the generation of pyrroles 5.
Introduction
Since the pioneering work of Kagan and co-workers,[1]
samarium diiodide has become one of the most important
chemoselective reducing agents in organic synthesis. Its ex-
traordinary potential for the formation of challenging scaf-
folds was exploited in the preparation of various natural
products and has recently been reviewed.[2] In addition to
the formation of new C–C bonds, the use of SmI2 often
allows smooth transformations of common functional
groups. Thus, dehalogenation and deoxygenation,[3] hetero-
atom/heteroatom bond scission,[3,4] heteroatom/benzyl
cleavage, and removal of tosyl groups[5] are possible, de-
pending on the reaction conditions.
In the course of ongoing projects in the synthesis of natu-
ral products or their analogues, we studied the chemoselec-
tive N–O bond cleavage of various enantiopure 1,2-oxazine
derivatives with SmI2. These versatile heterocycles are easily
available by [3+3] cyclization of lithiated alkoxyallenes with
nitrones[6] and can be subsequently modified. As disclosed
in several reports, smooth SmI2-mediated ring opening of
monocyclic 1,2-oxazine derivatives[4d,4f,7] and bicyclic com-
[a] Institut für Chemie und Biochemie, Freie Universität Berlin,
Takustr. 3, 14195 Berlin, Germany
Fax: +49-30-83855367
E-mail: hans.reissig@chemie.fu-berlin.de
Homepage: www.bcp.fu-berlin.de/chemie/oc/reissig
[‡] Present address: Department of Organic and Applied
Chemistry, University of Łódz´,
Tamka 12, 91-403 Łódz´, Poland
[‡‡] Present address: School of Pharmaceutical Sciences,
International University of Health and Welfare,
2600-1, Kitakanemaru, Otawara, Togichi 324-8501, Japan
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201201210.
Scheme 1. Samarium diiodide induced N–O bond cleavage of tetra-
hydro-2H-1,2-oxazines 1 and of 3,6-dihydro-2H-1,2-oxazines 3.
Eur. J. Org. Chem. 2013, 605–610
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
605

M. Jasin´ski, T. Watanabe, H.-U. Reissig
FULL PAPER
The aim of this report is to call attention to the unexpected Table 1. Attempted optimization of the reaction of syn-3a with
SmI2.
formation of this latter product and to discuss a possible
mechanism.
Entry SmI2
[equiv.]
Additive
Temp. Time Yield [%]
[°C]
[h]
syn-3 syn-4a
5a
1
2
0.25
2.2
–
–
r.t.
r.t.
2
2
80
23
0
41
0
21
Results and Discussion
(ca. 2:1)
65
(ca. 2.5:1)
74
3
4
3.3
5.5
–
–
r.t.
r.t.
2
2
0
0
27
15
0
The unexpected formation of pyrrole 5a as a side product
in fair amounts was discovered during the samarium di-
iodide promoted reduction of d-glyceraldehyde-derived 3,6-
dihydro-2H-1,2-oxazine syn-3a[6c] (Scheme 2). Treatment of
syn-3a with 2.2 equivalents of freshly prepared SmI2 solu-
tion in tetrahydrofuran afforded the expected amino
alcohol syn-4a as the major component (41% yield) and the
relatively unstable pyrrole derivative 5a (21% yield) as the
minor product. Small amounts of starting material (23%)
(ca. 5:1)
0
5
6
7
3.3
2.2
2.2
–
–
–78
66
r.t.
2
92
0
[a]
10 min
20
–
tBuOH
(2.0 equiv.)
HMPA
(18.0 equiv.)
8[b]
4[b]
1[b]
0[b]
8
2.2
r.t.
2
100[b] 0[b]
[a] A complex mixture of 4a, 5a and unidentified products was
formed. [b] Ratio estimated by H NMR spectroscopic analysis of
1
1
were also isolated from the mixture. The H NMR spec-
trum of 5a showed the characteristic sets of signals attrib-
uted to the 1,3-dioxolane, methoxy, and benzyl moieties,
and two additional doublets at δ = 5.90 and 6.46 ppm (J =
3.1 Hz each). In the 13C NMR spectrum, the signals at δ =
95.5, 111.1, 120.3, and 147.3 ppm were assigned to the car-
bon atoms of the newly formed pyrrole ring. All analytical
data supported the proposed structure of the hitherto un-
known 1-benzyl-3-methoxypyrrole derivative 5a, bearing
the chiral 1,3-dioxolanyl side chain.
the crude reaction mixture.
conversion of diastereomeric 1,2-oxazine anti-3a was signif-
icantly slower under the conditions optimized for syn-3a
(entry 1). After two hours, pyrrole derivative 5a was ob-
tained in 13% yield only, together with unchanged precur-
sor (38%) and the corresponding amino alcohol anti-4a
(38%). The two l-erythronolactone-derived 1,2-oxazines 3b
and 3c afforded allylic alcohols 4b and 4c as major products
and pyrroles 5b and 5c were formed only in 18 and 6%
yield, respectively (entries 2 and 3). The considerably lower
yield of pyrrole 5c observed in the latter case of anti-config-
ured substrate 3c can be explained by the relative configura-
tion at C-3 (compare reactions of syn-3a and anti-3a) and
the presence of the free hydroxyl group, which very likely
serves as proton source. Whereas syn-configured 3b was
fully consumed within two hours (reaction monitored by
Scheme 2. Reaction of d-glyceraldehyde-derived 2H-1,2-oxazine
syn-3a with samarium diiodide.
Although compounds such as syn-4a and 5a can be easily TLC), in the case of 3c the starting material could still be
separated by simple chromatography, we attempted to opti- detected after this time (similar to anti-3a), and complete
mize this transformation to influence the selectivity and conversion required four hours. The observed dependence
possibly obtain hints for the mechanism of the surprising of reactivity on the relative configuration of 3,6-dihydro-
pyrrole formation. The results of these experiments are 2H-1,2-oxazines has also been observed in other transfor-
summarized in Table 1. We examined the effects of the mations, for example, in hydroborations.[4d,11a]
quantities of reducing agent (entries 1–4) and of the tem-
To our surprise, the reaction of samarium diiodide with
perature (entries 5 and 6) and found that ca. 3.3 equiv. of 1,2-oxazine 3d (derived from Garner’s aldehyde) provided
SmI2 at room temperature appeared to be optimal for the the corresponding amino alcohol 4d almost exclusively,
formation of pyrrole derivative 5a. The use of a larger ex- and, after purification on silica gel, this product was iso-
cess of samarium diiodide led to a higher syn-4a/5a ratio lated in 73% yield (Table 2, entry 4). A similar result was
and enabled isolation of amino alcohol syn-4a in 74% yield. noted for fused bicyclic derivative 3e, which, after 4.5 hours,
Neither the use of tert-butanol as proton source nor HMPA furnished pyrrolidine derivative 4e in 68% yield; however,
as a Lewis base resulted in the formation of pyrrole 5a in unconsumed starting material (11%) was still present in the
1
significant amounts (entries 7 and 8). Decomposition crude mixture. Careful analysis of the H NMR spectra of
mainly occurred in the presence of a proton source, whereas the crude reaction mixtures allowed detection of the signals
addition of HMPA led to no conversion at all. This is rather of the respective pyrrole derivatives 5d and 5e in very low
surprising because the addition of this Lewis base to SmI2 amounts (less than 1% yield). In the spectra, one typical
strongly enhances its reductive power in many reactions.[12] set of doublets was detected at δ = 5.90 and 6.47 ppm (J =
Examples of 3,6-dihydro-2H-1,2-oxazines 3 that were ex- 2.9 Hz), and 5.90 and 6.36 ppm (J = 2.8 Hz), respectively.
amined under standard conditions (in general with ca. Due to the low content and limited stability it was not pos-
3 equiv. of SmI2) are presented in Table 2. Interestingly, the sible to isolate pyrroles 5d and 5e.
606
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 605–610

Reduction of 3,6-Dihydro-2H-1,2-oxazines
Table 2. Reductions of 3,6-dihydro-2H-1,2-oxazines 3 with SmI2.[a] material 3f were recovered. Subsequent experiments showed
that the use of ca. 10 equiv. SmI2 was appropriate in this
case, leading to full consumption of 3f. In contrast to model
compound syn-3a, use of the higher amount of SmI2 did
not decrease the amount of pyrrole formation and finally
allowed the isolation of 5f in 20% yield.
A plausible mechanism for the observed SmI2-mediated
transformation of 3,6-dihydro-2H-1,2-oxazines 3 into pyr-
roles is illustrated in Scheme 3. Coordination of the Lewis
acidic samarium diiodide to the 1,2-oxazine oxygen forms
adduct A, which suffers N–O bond cleavage to generate the
crucial intermediate B bearing an N-centred radical moiety.
Depending on the amount of reducing agent, this interme-
diate can undergo two reactions. A subsequent reduction of
intermediate B with a second equivalent of samarium diiod-
ide will lead to intermediate D and finally to the 1,4-amino
alcohol 4. This pathway is favoured for most compounds
studied and is facilitated by the addition of larger amounts
of samarium diiodide. Alternatively, several of the com-
pounds investigated also react through a minor pathway in-
volving an intramolecular hydrogen shift as a crucial step,
providing samarium ketyl C as intermediate. It is well-
known that samarium ketyls exist in equilibrium with the
corresponding carbonyl compounds and, hence, C may de-
liver α,β-unsaturated aldehyde E. This compound can un-
dergo a favourable 5-exo-trig-cyclization to afford interme-
diate F, which, upon elimination of water, provides pyrrole
derivative 5.[14]
According to the mechanism depicted in Scheme 3, the
pathway leading to pyrroles should require only catalytic
amounts of SmI2. However, the result presented in Table 1
(entry 1) reveals that 0.25 equiv. of the reagent is apparently
[a] Reaction conditions (if not stated otherwise): SmI2 (entry 1:
not sufficient to lead to detectable formation of 5a. Possibly,
larger amounts of the Lewis acidic samarium(II) or the gen-
erated samarium(III) species are required to trigger the
steps leading to the formation of pyrrole 5. The observation
that application of more than 3 equiv. of samarium diiodide
considerably disfavours the formation of pyrroles is in
agreement with our mechanistic proposal, because interme-
diate D should then be formed faster than samarium ketyl
C. The failed reaction of syn-3a with SmI2 in the presence
of HMPA is also significant (Table 1, entry 8). The observa-
3.3 equiv.; entries 2–5: ca. 3.0 equiv.; entry 6: ca. 10 equiv.), THF,
2 h. [b] Starting material anti-3a was recovered in 38% yield. [c] Re-
action time 4 h; yields were estimated by 1H NMR spectroscopic
analysis of the crude mixture; compound 4c could be isolated in
79% yield as reported.[11a] [d] Only traces (Ͻ1%) of the corre-
sponding pyrrole derivatives were detected in the 1H NMR spectra
of the crude mixtures. [e] Reaction time 4.5 h; starting 1,2-oxazine
3e was recovered in 11% yield. [f] Reaction time 24 h; a complex
mixture of very polar compounds derived from the expected amino
alcohol 4f was formed.
In a recent report, a series of 3,6-dihydro-1,2-oxazines tion that no conversion at all was observed is in agreement
such as 3f, bearing an l-erythronolactone-derived auxiliary with the mechanism depicted in Scheme 3, in which (revers-
at the nitrogen atom, were presented as attractive precur- ible) coordination of the 1,2-oxazine oxygen to SmI2 lead-
sors for the synthesis of highly functionalized hydroxylated ing to complex A is proposed as the initial step. The pres-
compounds.[13] Under all conditions examined, the reaction ence of the very strong Lewis base HMPA[15] completely
of 3f with samarium diiodide furnished only moderate prevents this step and, hence, both pathways leading to 4
amounts of pyrrole 5f and numerous highly polar side- and 5 are apparently unfavourable. The differences in rates
products (Table 2, entry 6). The expected allylic alcohol 4f and the dependence of the 4/5 ratio on configuration and
was not detected, which was probably due to the existence structure of starting materials 3 cannot be explained by sim-
of ring-chain tautomers of this compound allowing a range ple assumptions. The formation of pyrroles by ring contrac-
of destructive subsequent reactions. Only pyrrole 5f could tion[16] of 3,6-dihydro-1,2-oxazines has been described in
be isolated by chromatography as a less polar fraction. The the literature. Conversion under basic[17] or acidic condi-
first experiment with 3f was conducted using ca. 6.0 equiv. tions[18] and at elevated temperatures and pressure[19] are
of SmI2, and, after five hours, the dark-blue solution be- known. Nevertheless, all the methods described seem to have
came yellow, indicating complete consumption of the reduc- limitations. Alternatively, photochemical reactions of certain
ing agent. However, significant amounts (37%) of starting 3,6-dihydro-2H-1,2-oxazines may also furnish pyrroles.[20,21]
Eur. J. Org. Chem. 2013, 605–610
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
607

M. Jasin´ski, T. Watanabe, H.-U. Reissig
FULL PAPER
Bruker AC 300 and FTIR Nicolet 5SXC instruments, respectively.
All reactions were performed under argon in flame-dried flasks
with addition of the components by using a syringe.
Reaction of syn-3a with SmI2: 1,2-Oxazine syn-3a (150 mg,
0.50 mmol) in tetrahydrofuran (5 mL) was treated with a freshly
prepared solution of samarium diiodide [obtained by reaction of
samarium metal (271 mg, 1.80 mmol) with 1,2-diiodoethane
(465 mg, 1.65 mmol) in tetrahydrofuran (10 mL) and stirring at
room temp. for 2 h], and stirred at ambient temperature for 2 h.
The reaction mixture was quenched with 10% aqueous Na2S2O3
(30 mL), extracted with CH2Cl2 (3ϫ30 mL) and filtered through a
pad of Celite. The organic layer was dried with MgSO4, filtered
and evaporated to give a pale-yellow oil (140 mg). Purification by
column chromatography (silica gel; hexane/ethyl acetate, 3:1 gradi-
ent to pure ethyl acetate, then ethyl acetate/methanol, 10:1) af-
forded unstable 5a (39 mg, 27%, first eluted) and 4a (100 mg, 65%,
second eluted) as colourless oils.
(E)-(4S,4ЈS)-4-Benzylamino-4-(2Ј,2Ј-dimethyl-1Ј,3Ј-dioxolan-4Ј-yl)-
3-methoxybut-2-en-1-ol (syn-4a): [α]2D2 = +1.3 (c = 0.50, CHCl3). 1H
NMR (CDCl3, 300 MHz): δ = 1.34, 1.35 (2ϫs, 2 ϫ3 H, 2 Me),
2.69 (br. s, 2 H, NH, OH), 3.54 (s, 3 H, OMe), 3.57 (d, J = 8.0 Hz,
1 H, 4-H), 3.63 (d, J = 13.3 Hz, 1 H, CH2Ph), 3.76 (dd, J = 6.0,
8.5 Hz, 1 H, 5Ј-H), 3.88 (d, J = 13.3 Hz, 1 H, CH2Ph), 3.94 (dd, J
= 6.3, 8.5 Hz, 1 H, 5Ј-H), 4.04 (d, J = 7.4 Hz, 2 H, 1-H), 4.35 (ddd,
J = 6.0, 6.3, 8.0 Hz, 1 H, 4Ј-H), 5.02 (t, J ≈ 7.4 Hz, 1 H, 2-H),
7.20–7.38 (m, 5 H, Ph) ppm. 13C NMR (CDCl3, 75.5 MHz): δ =
25.4, 26.7 (2ϫq, Me), 50.9 (t, CH2Ph), 54.5 (d, C-4), 57.3 (t, C-1),
59.9 (q, OMe), 66.4 (t, C-5Ј), 76.6 (d, C-4Ј), 101.2 (d, C-2), 109.4
(s, C-2Ј), 127.1, 128.2, 128.4, 139.5 (3 ϫ d, s, Ph), 157.0 (s, C-
3) ppm. IR (film): ν = 3400–3330 (O–H, N–H), 3080–2820
˜
(=C–H, C–H), 1660 (C=C) cm–1. C17H25NO4 (307.4): calcd. C
66.43, H 8.20, N 4.56; found C 66.26, H 8.30, N 4.88.
(4ЈS)-1-Benzyl-2-(2Ј,2Ј-dimethyl-1Ј,3Ј-dioxolan-4Ј-yl)-3-methoxy-
pyrrole (5a): [α]2D2 = –0.65 (c = 0.30, CHCl3). 1H NMR (CDCl3,
300 MHz): δ = 1.37, 1.43 (2ϫs, 2 ϫ3 H, 2 Me), 3.75 (s, 3 H, OMe),
3.96 (dd, J = 6.7, 8.2 Hz, 1 H, 5Ј-H), 4.17 (t, J ≈ 8.2 Hz, 1 H, 5Ј-
H), 5.07 (d, J = 16.1 Hz, 1 H, CH2Ph), 5.08–5.18 (m, 1 H, 4Ј-H),
5.19 (d, J = 16.1 Hz, 1 H, CH2Ph), 5.90, 6.46 (2ϫd, J = 3.1 Hz, 2
H, 4-H, 5-H), 7.04 (br d, J = 6.7 Hz, 2 H, Ph), 7.20–7.36 (m, 3 H,
Ph) ppm. 13C NMR (CDCl3, 75.5 MHz): δ = 25.5, 26.5 (2ϫq, Me),
51.3 (t, CH2Ph), 58.5 (q, OMe), 66.6 (t, C-5Ј), 69.5 (d, C-4Ј), 95.5
(d, C-4), 108.5 (s, C-2Ј), 111.1 (s, C-2), 120.3 (d, C-5), 126.5, 127.4,
Scheme 3. Proposed mechanism for the SmI2-mediated reaction of
3,6-dihydro-2H-1,2-oxazines 3 leading to amino alcohols 4 or pyr-
roles 5.
128.6, 138.4 (3ϫd, s, Ph), 147.3 (s, C-3) ppm. IR (film): ν = 3150–
˜
2800 (=C–H, C–H), 1580 (C=C) cm–1. C17H21NO3 (287.4): calcd.
C 71.06, H 7.37, N 4.87; found C 71.04, H 7.39, N 5.19.
Conclusions
Reaction of 1,2-Oxazine anti-3a with SmI2: By a procedure similar
to that used for syn-3a, 1,2-oxazine anti-3a (100 mg, 0.33 mmol) in
tetrahydrofuran (4 mL) was added to a freshly prepared solution
of samarium diiodide [obtained from samarium metal (177 mg,
1.18 mmol) and 1,2-diiodoethane (304 mg, 1.08 mmol) in tetra-
hydrofuran (8 mL)]. After workup and purification by column
chromatography (silica gel; hexane/ethyl acetate, 3:1 gradient to
ethyl acetate, then ethyl acetate/methanol, 10:1) afforded pyrrole
derivative 5a (12 mg, 13%), unconsumed anti-3a (38 mg, 38%), and
amino alcohol anti-4a (38 mg, 38%) as a brown oil.
In this report we described samarium diiodide induced
reactions of 3,6-dihydro-2H-1,2-oxazines 3 leading to the
expected 1,4-amino alcohols 4 and the fairly unstable elec-
tron-rich pyrrole derivatives 5 as minor components. We
present a mechanistic proposal for the formation of the pyr-
roles that involves an interesting internal redox reaction as a
key step. The ability to predict which precursor 1,2-oxazine
derivatives will provide the pyrroles in reasonable amount
is not yet possible. The reported results again emphasize the
versatility (and partial unpredictability) of reactions using
1,2-oxazines as key compounds.[22]
(E)-(4R,4ЈS)-4-Benzylamino-4-(2Ј,2Ј-dimethyl-1Ј,3Ј-dioxolan-4Ј-yl)-
3-methoxybut-2-en-1-ol (anti-4a): [α]2D2 = –3.0 (c = 0.50, CHCl3). 1H
NMR (CDCl3, 300 MHz): δ = 1.32, 1.38 (2ϫs, 2 ϫ3 H, 2 Me),
2.02 (br. s, 2 H, NH, OH), 3.40 (d, J = 9.2 Hz, 1 H, 4-H), 3.51 (d,
J = 13.1 Hz, 1 H, CH2Ph), 3.61 (s, 3 H, OMe), 3.75 (d, J = 13.1 Hz,
1 H, CH2Ph), 3.82–4.06 (m, 4 H, 1-H, 5Ј-H), 4.20 (dd, J = 6.2,
8.5 Hz, 1 H, 4Ј-H), 5.18 (t, J ≈ 8.2 Hz, 1 H, 2-H), 7.20–7.37 (m, 5
Experimental Section
For general information, see Jasin´ski et al.;[11a] the NMR and IR
spectra of compounds syn-4a, anti-4a and 5a were recorded with
608
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 605–610

Reduction of 3,6-Dihydro-2H-1,2-oxazines
H, Ph) ppm. 13C NMR (CDCl3, 75.5 MHz): δ = 25.3, 26.3 (2ϫq,
Me), 51.2 (t, CH2Ph), 54.7 (d, C-4), 56.9 (t, C-1), 58.9 (q, OMe),
69.0 (t, C-5Ј), 75.6 (d, C-4Ј), 100.9 (d, C-2), 109.5 (s, C-2Ј), 127.1,
1.34 (2ϫs, 2 ϫ3 H, 2 Me), 3.31 (dd, J = 4.5, 11.9 Hz, 1 H, 5Ј-
CH2), due to overlapping the signals of 5Ј-CH2 (1 H), OMe and
5Ј-H could not be assigned, 4.97, 5.20 (2ϫd, J = 15.9 Hz, 2 H,
CH2Ph), 5.37 (d, J = 7.4 Hz, 1 H, 4Ј-H), 5.91, 6.45 (2ϫd, J =
128.2, 128.3, 140.0 (3ϫd, s, Ph), 157.5 (s, C-3) ppm. IR (film): ν =
˜
3450–3340 (O–H, N–H), 3150–2820 (=C–H, C–H), 1660 3.1 Hz, 2 H, 4-H, 5-H), 7.01–7.04 (m, 2 H, Ph) ppm, 3 H signals
(C=C) cm–1. C17H25NO4 (307.4): calcd. C 66.43, H 8.20, N 4.56;
found C 65.97, H 8.64, N 4.83.
of Ph overlapped with other signals.
(E)-(4R,4ЈR)-4-Benzylamino-4-(3Ј-tert-butoxycarbonyl-2Ј,2Ј-di-
methyl-1Ј,3Ј-oxazolidin-4Ј-yl)-3-methoxybut-2-en-1-ol (4d): By a
procedure similar to that used with 1,2-oxazine 3b, compound 3d
(85 mg, 0.21 mmol) in tetrahydrofuran (7 mL) was added to a sa-
marium diiodide solution (ca. 0.1 m in THF, 6.3 mL, ca.
0.63 mmol) and stirred for 3 h. After standard workup, the crude
product was purified by chromatography (silica gel; dichlorometh-
ane/methanol, 40:1) to give amino alcohol 4d (62 mg, 73%) as a
colourless oil. [α]2D2 = –37.6 (c = 1.28, CHCl3). 1H NMR ([D6]-
DMSO, 500 MHz, 90 °C): δ = 1.39 (s, 9 H, tBu), 1.40, 1.42 (2ϫs,
2ϫ3 H, 2 Me), 2.13, 4.09 (2ϫbr. s, 2 H, NH, OH), 3.51 (d, J =
13.6 Hz, 1 H, CH2Ph), 3.51 (s, 3 H, OMe), 3.63 (d, J = 7.3 Hz, 1
H, 4-H), 3.73 (d, J = 13.6 Hz, 1 H, CH2Ph), 3.76 (dd, J = 5.9,
9.0 Hz, 1 H, 5Ј-H), 3.88 (dd, J = 6.5, 12.2 Hz, 1 H, 1-H), 3.91 (d,
J = 9.0 Hz, 1 H, 5Ј-H), 3.97 (dd, J = 7.7, 12.2 Hz, 1 H, 1-H), 4.05
(br t, J ≈ 6.4 Hz, 1 H, 4Ј-H), 4.89 (br t, J ≈ 7.1 Hz, 1 H, 2-H),
7.17–7.21, 7.25–7.31 (2ϫm, 5 H, Ph) ppm. 13C NMR ([D6]DMSO,
126 MHz, 90 °C): δ = 24.3, 27.2 (2ϫq, Me), 28.6 (q, tBu), 51.2 (t,
CH2Ph), 54.9 (q, OMe), 57.0 (t, C-1), 58.2 (d, C-4), 59.8 (d, C-4Ј),
64.9 (t, C-5Ј), 79.6 (s, tBu), 93.6 (d, C-2Ј), 103.4 (s, C-2), 127.0,
128.4*, 141.5 (2ϫd, s, Ph), 152.7 (s, C=O), 155.2 (s, C-3) ppm;
Reaction of 3b with SmI2: To a solution of samarium diiodide (ca.
0.1 m in THF, 10.5 mL, ca. 10.5 mmol; prepared by the method of
Imamoto[23]), a solution of 1,2-oxazine 3b (161 mg, 0.36 mmol) in
degassed tetrahydrofuran (8 mL) was added dropwise at room
temp. The mixture was stirred for 2 h, quenched with sat. aqueous
sodium potassium tartrate (10 mL) and extracted with diethyl ether
(3ϫ20 mL). The combined organic layers were dried (MgSO4), fil-
tered and the solvents removed under reduced pressure. Purifica-
tion by column chromatography (alumina; hexane/dichlorometh-
ane, 3:2) afforded 5b (28 mg, 18%, Ͼ90% purity, first eluted) and
amino alcohol 4b (97 mg, 60%) as colourless oils.
(E)-(4R,4ЈR,5ЈS)-4-Benzylamino-4-(5Ј-tert-butyldimethylsiloxy-
methyl-2Ј,2Ј-dimethyl-1Ј,3Ј-dioxolan-4Ј-yl)-3-methoxybut-2-en-1-ol
(4b): [α]2D2 = –15.4 (c = 1.30, CHCl3). 1H NMR (CDCl3, 500 MHz):
δ = 0.039, 0.042, 0.88 (3ϫs, 2 ϫ3 H, 9 H, TBS), 1.37, 1.45 (2ϫs,
2ϫ3 H, 2 Me), 3.54 (s, 3 H, OMe), 3.60 (dd, J = 5.0, 11.1 Hz, 1
H, CH2OTBS), 3.61 (d, J = 12.8 Hz, 1 H, CH2Ph), 3.72 (dd, J =
4.3, 11.1 Hz, 1 H, CH2OTBS), 3.84 (d, J = 12.8 Hz, 1 H, CH2Ph),
3.93 (d, J = 8.7 Hz, 1 H, 4-H), 4.06 (dd, J = 7.4, 12.6 Hz, 1 H, 1-
H), 4.13 (td, J ≈ 4.4, 5.9 Hz, 1 H, 5Ј-H), 4.17 (dd, J = 7.6, 12.6 Hz,
1 H, 1-H), 4.64 (dd, J = 6.2, 8.7 Hz, 1 H, 4Ј-H), 5.06 (t, J ≈ 7.5 Hz,
1 H, 2-H), 7.21–7.25, 7.29–7.31 (2ϫm, 5 H, Ph) ppm; OH and NH
signals could not be detected. 13C NMR (CDCl3, 126 MHz): δ =
–5.3*, 18.4, 26.0 (q, s, q, TBS), 25.3, 27.6 (2 ϫ q, Me), 48.7 (t,
CH2Ph), 54.5 (q, OMe), 56.5 (d, C-4), 57.3 (t, C-1), 62.4 (t, 5Ј-
CH2), 75.1 (d, C-4Ј), 78.0 (d, C-5Ј), 100.6 (d, C-2), 108.0 (s, C-2Ј),
127.1, 128.2, 128.4. 139.6 (3ϫd, s, Ph), 158.8 (s, C-3) ppm; * higher
* higher intensity. IR (ATR): ν = 3515–3320 (O–H, N–H), 3085–
˜
2820 (=C–H, C–H), 1695 (C=O), 1655 (C=C), 1390, 1250, 1170,
1080 (C–O) cm–1. HRMS (ESI-TOF): calcd. for C22H35N2O5 [M +
H]+ 407.2546; found 407.2538.
(E)-(3aЈR,4ЈS,6aЈS)-3-(2Ј,2Ј-Dimethyltetrahydro-[1,3]dioxolo[4,5-c]-
pyrrol-4Ј-yl)-3-methoxyprop-2-en-1-ol (4e): By a procedure similar
to that used with 1,2-oxazine 3b, compound 3e (18.3 mg,
0.080 mmol) in tetrahydrofuran (2 mL) was added to a samarium
diiodide solution (ca. 0.1 m in THF, 2.4 mL, ca. 0.24 mmol) and
stirred for 4.5 h. After typical workup, the crude mixture was puri-
fied by column chromatography (silica gel; dichloromethane/meth-
anol, 15:1) to give unconsumed 3e (2.0 mg, 11 %) and amino
alcohol 4e (12.7 mg, 68%), which was isolated as a colourless oil.
intensity. IR (ATR): ν = 3490–3295 (O–H, N–H), 3090–2830
˜
(=C–H, C–H), 1655 (C=C), 1215, 1090 (C–O) cm–1. HRMS (ESI-
TOF): calcd. for C24H42NO5Si [M + H]+ 452.2827; found 452.2849.
(4ЈR,5ЈS)-1-Benzyl-2-(5Ј-tert-butyldimethylsiloxymethyl-2Ј,2Ј-di-
methyl-1Ј,3Ј-dioxolan-4Ј-yl)-3-methoxypyrrole (5b): [α]2D2 = –14.9 (c
= 1.29, CHCl3). 1H NMR (CDCl3, 500 MHz): δ = –0.04, –0.02,
0.84 (3ϫs, 2 ϫ3 H, 9 H, TBS), 1.34, 1.41 (2ϫs, 2 ϫ3 H, 2 Me),
3.38 (dd, J = 4.6, 10.8 Hz, 1 H, CH2OTBS), 3.66 (dd, J = 7.3,
10.8 Hz, 1 H, CH2OTBS), 3.73 (s, 3 H, OMe), 4.30 (td, J ≈ 4.6,
7.6 Hz, 1 H, 5Ј-H), 4.93, 5.22 (2ϫd, J = 15.8 Hz, 2 H, CH2Ph),
5.43 (d, J = 7.9 Hz, 1 H, 4Ј-H), 5.88, 6.35 (2ϫd, J = 3.1 Hz, 2 H,
4-H, 5-H), 7.04–7.06, 7.21–7.32 (2ϫm, 5 H, Ph) ppm. 13C NMR
(CDCl3, 126 MHz): δ = –5.4*, 18.4, 25.9 (q, s, q, TBS), 23.5, 26.1
(2ϫq, Me), 51.8 (t, CH2Ph), 58.2 (q, OMe), 63.8 (t, 5Ј-CH2), 70.7
(d, C-4Ј), 78.9 (d, C-5Ј), 95.4 (d, C-4), 107.7 (s, C-2Ј), 110.9 (s, C-
2), 119.6 (d, C-5), 126.6, 127.3, 128.5, 138.6 (3ϫd, s, Ph), 146.9 (s,
1
[α]2D2 = +54.4 (c = 1.10, CHCl3). H NMR ([D6]DMSO, 500 MHz,
70 °C): δ = 1.24, 1.41 (2ϫs, 2 ϫ3 H, 2 Me), 2.81 (br d, J ≈ 12.2 Hz,
1 H, 6Ј-H), 3.04 (dd, J = 4.9, 12.2 Hz, 1 H, 6Ј-H), 3.44 (s, 3 H,
OMe), 3.95 (br d, J ≈ 2.0 Hz, 1 H, 4Ј-H), 3.99 (dd, J = 7.3, 12.3 Hz,
1 H, 1-H), 4.03 (dd, J = 7.1, 12.3 Hz, 1 H, 1-H), 4.57 (dd, J = 2.0,
6.0 Hz, 1 H, 3aЈ-H), 4.64–4.67 (m, 1 H, 6aЈ-H), 4.69 (t, J ≈ 7.2 Hz,
1 H, 2-H) ppm; OH and NH signals could not be detected. 13C
NMR ([D6]DMSO, 126 MHz, 70 °C): δ = 24.3, 26.3 (2ϫq, Me),
52.8 (t, C-6Ј), 53.8 (q, OMe), 56.0 (t, C-1), 62.5 (d, C-4Ј), 81.4 (d,
C-3aЈ), 83.9 (d, C-6aЈ), 99.1 (d, C-2), 110.7 (s, C-2Ј), 157.4 (s, C-
3) ppm. IR (ATR): ν = 3440–3180 (O–H, N–H), 3020–2825 (=C–
˜
C-3) ppm; * higher intensity. IR (ATR): ν = 3065–2835 (=C–H,
˜
H, C–H), 1660 (C=C), 1210, 1040 (C–O) cm–1. HRMS (ESI-TOF):
calcd. for C11H20NO4 [M + H]+ 230.1387; found 230.1385.
C–H), 1580 (C=C), 1250, 1090, 1070 cm–1. HRMS (ESI-TOF):
calcd. for C24H37NNaO4Si [M + Na]+ 454.2384; found 454.2375.
Reaction of 3c with SmI2: By a procedure analogous to that used
with 1,2-oxazine 3b, compound 3c (30 mg, 0.089 mmol) in tetra-
hydrofuran (1 mL) was treated with samarium diiodide (ca. 0.1 m
(3aЈS,4ЈS,6aЈS)-1-(2Ј,2Ј-Dimethyltetrahydrofuro[3,4-d][1,3]dioxol-
4Ј-yl)-3-methoxy-2-phenylpyrrole (5f): By a procedure similar to
that used with 1,2-oxazine 3b, compound 3f (126 mg, 0.38 mmol)
in THF, 2.7 mL, ca. 0.27 mmol) to quantitatively yield a mixture in tetrahydrofuran (10 mL) was added to a samarium diiodide solu-
1
of crude 4c and 5c (30 mg, 94:6 ratio based on the H NMR spec-
trum) after standard workup. Isolation and characterisation data
of 4c was described elsewhere.[11a] Compound 5c was not isolated
tion (ca. 0.1 m in THF, 38 mL, ca. 3.8 mmol) and stirred for 24 h.
After typical workup and purification by column chromatography
(silica gel; dichloromethane) compound 5f (24 mg, 20%) was ob-
1
but the yield (6%) was estimated from the H NMR spectrum of
tained as colourless crystals; m.p. 132–134 °C. [α]2D2 = +150.6 (c =
the crude mixture. 1H NMR of 5c (CDCl3, 500 MHz): δ = 1.25, 1.00, CHCl3). H NMR (CDCl3, 500 MHz): δ = 1.33, 1.49 (2ϫs,
1
Eur. J. Org. Chem. 2013, 605–610
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
609

M. Jasin´ski, T. Watanabe, H.-U. Reissig
FULL PAPER
[7] a) B. Bressel, B. Egart, A. Al-Harrasi, R. Pulz, H.-U. Reissig,
I. Brüdgam, Eur. J. Org. Chem. 2008, 467–474; b) V. Dekaris,
H.-U. Reissig, Synlett 2010, 42–46.
[8] A. Al-Harrasi, H.-U. Reissig, Angew. Chem. 2005, 117, 6383–
6387; Angew. Chem. Int. Ed. 2005, 44, 6227–6231.
[9] For other methods furnishing pyrrole derivatives starting from
alkoxyallenes, see: a) M. Okala Amombo, A. Hausherr, H.-U.
Reissig, Synlett 1999, 1871–1874; b) O. Flögel, H.-U. Reissig,
Synlett 2004, 895–897; c) O. Flögel, J. Dash, I. Brüdgam, H.
Hartl, H.-U. Reissig, Chem. Eur. J. 2004, 10, 4283–4290.
[10] R. Pulz, T. Watanabe, W. Schade, H.-U. Reissig, Synlett 2000,
983–986.
2ϫ3 H, 2 Me), 3.73 (s, 3 H, OMe), 4.07–4.14 (m, 2 H, 6Ј-H), 5.03
(m, 2 H, 3aЈ-H, 6aЈ-H), 5.82 (s, 1 H, 4Ј-H), 6.09, 6.34 (2ϫd, J =
3.3 Hz, 2 H, 4-H, 5-H), 7.27–7.31, 7.40–7.44, 7.50–7.53 (3ϫm, 5
H, Ph) ppm. 13C NMR (CDCl3, 126 MHz): δ = 24.7, 26.2 (2ϫq,
Me), 58.5 (q, OMe), 73.3 (t, C-6Ј), 80.7 (d, C-6aЈ), 84.7 (d, C-3aЈ),
90.3 (d, C-4Ј), 98.0 (d, C-4), 112.9 (s, C-2Ј), 115.2 (d, C-5), 119.0
(s, C-2), 126.8, 128.4, 130.0, 130.3 (3ϫd, s, Ph), 145.4 (s, C-3) ppm.
IR (ATR): ν = 3065–2830 (=C–H, C–H), 1210, 1080 (C–O) cm–1.
˜
ESI-TOF (m/z): calcd. for C18H21NNaO4 [M + Na]+ 338.1363;
found 338.1361.
[11] a) M. Jasin´ski, D. Lentz, E. Moreno-Clavijo, H.-U. Reissig,
Eur. J. Org. Chem. 2012, 3304–3316; b) V. Dekaris, B. Bressel,
H.-U. Reissig, Synlett 2010, 47–50.
Supporting Information (see footnote on the first page of this arti-
1
cle): Copies of H and 13C NMR spectra of compounds prepared.
[12] a) M. Shabangi, R. A. Flowers II, Tetrahedron Lett. 1997, 38,
1137–1140; b) M. Shabangi, J. M. Sealy, J. R. Fuchs, R. A.
Flowers II, Tetrahedron Lett. 1998, 39, 4429–4432; for reviews,
see: c) A. Dahlén, G. Hilmersson, Eur. J. Inorg. Chem. 2004,
3393–3403; d) R. A. Flowers II, Synlett 2008, 1427–1439.
[13] M. Jasin´ski, D. Lentz, H.-U. Reissig, Beilstein J. Org. Chem.
2012, 8, 662–674.
[14] The formation of intermediate C and finally of product 5 can
be classified as an internal redox reaction under N–O bond
cleavage, which is known for other N-O heterocycles under dif-
ferent conditions, see: a) S.-I. Murahashi, Y. Kodera, T. Ho-
somi, Tetrahedron Lett. 1988, 29, 5949–5952; b) M. Meske, J.
Prakt. Chem. 1997, 339, 426–433; c) P. Bayón, P. de March, M.
Figueredo, J. Font, Tetrahedron 1998, 54, 15691–15700; d) F.
Machetti, F. M. Cordero, F. De Sarlo, A. Brandi, Tetrahedron
2001, 57, 4995–4998; e) C. Chevrier, A. Defoin, Synthesis 2003,
1221–1224; f) B. Dugovicˇ, T. Wiesenganger, L. Fisˇera, C.
Hametner, N. Prónayová, Heterocycles 2005, 65, 591–605; g) B.
Dugovicˇ, L. Fisˇera, H.-U. Reissig, Eur. J. Org. Chem. 2008,
277–284; h) A. Al-Harrasi, L. Bouché, R. Zimmer, H.-U. Reis-
sig, Synthesis 2011, 109–118.
[15] For replacement of the toxic and carcinogenic HMPA, see: M.
Berndt, A. Hölemann, A. Niermann, C. Bentz, R. Zimmer, H.-
U. Reissig, Eur. J. Org. Chem. 2012, 1299–1302.
[16] For a review on pyrrole formation by ring contraction, see: U.
Joshi, M. Pipelier, S. Naud, D. Dubreuil, Curr. Org. Chem.
2005, 9, 261–288.
[17] a) J. Firl, G. Kresze, Chem. Ber. 1966, 99, 3695–3706; b) G.
Kresze, H. Braun, Tetrahedron Lett. 1969, 10, 1743–1746.
[18] a) K. F. McClure, S. J. Danishefsky, J. Org. Chem. 1991, 56,
850–853; b) G.-Q. Shi, M. Schlosser, Tetrahedron 1993, 49,
1445–1456.
[19] a) F. Ragaini, S. Cenini, E. Borsani, M. Dompé, E. Gallo, Or-
ganometallics 2001, 20, 3390–3398; b) F. Ragaini, S. Cenini, D.
Brignoli, M. Gasperini, E. Gallo, M. Moret, J. Org. Chem.
2003, 68, 460–466.
Acknowledgments
Support of this work by the Foundation for Polish Science (post-
doctoral fellowship to M. J.), the Deutsche Forschungsgemein-
schaft, and Bayer HealthCare is most gratefully acknowledged. We
also thank Dr. R. Zimmer for help during the preparation of this
manuscript.
[1] a) P. Girard, J. L. Namy, H. B. Kagan, J. Am. Chem. Soc. 1980,
102, 2693–2698. For selected reviews on samarium diiodide
promoted reactions, see: b) G. A. Molander, C. R. Harris,
Chem. Rev. 1996, 96, 307–338; c) G. A. Molander, C. R. Harris,
Tetrahedron 1998, 54, 3321–3354; d) H. B. Kagan, Tetrahedron
2003, 59, 10351–10372; e) M. Berndt, S. Gross, A. Hölemann,
H.-U. Reissig, Synlett 2004, 422–438; f) K. Gopalaiah, H. B.
Kagan, New J. Chem. 2008, 32, 607–637; g) M. Szostak, M.
Spain, D. Parmar, D. J. Procter, Chem. Commun. 2012, 48, 330–
346.
[2] a) D. J. Edmonds, D. Johnston, D. J. Procter, Chem. Rev. 2004,
104, 3371–3403; b) K. C. Nicolaou, S. P. Ellery, J. S. Chen, An-
gew. Chem. 2009, 121, 7276–7301; Angew. Chem. Int. Ed. 2009,
48, 7140–7165; c) C. Beemelmanns, H.-U. Reissig, Chem. Soc.
Rev. 2011, 40, 2199–2210; d) M. Szostak, D. J. Procter, Angew.
Chem. 2011, 123, 7881–7883; Angew. Chem. Int. Ed. 2011, 50,
7737–7739.
[3] D. J. Procter, R. A. Flowers II, T. Skrydrup, in: Organic Synthe-
sis Using Samarium Diiodide, A Practical Guide, Royal Society
of Chemistry Publishing, Cambridge, UK, 2010, chapter 4, pp.
36–68.
[4] For samarium diiodide promoted N–O bond cleavage, see: a)
J. L. Chiara, C. Destabel, P. Gallego, J. Marco-Contelles, J.
Org. Chem. 1996, 61, 359–360; b) S. H. Jung, J. E. Lee, H. Y.
Koh, Bull. Korean Chem. Soc. 1998, 19, 33–35; c) G. E. Keck,
T. T. Wager, S. F. McHardy, Tetrahedron 1999, 55, 11755–
11772; d) R. Pulz, A. Al-Harrasi, H.-U. Reissig, Org. Lett.
2002, 4, 2353–2355; e) J. Revuelta, S. Cicchi, A. Brandi, Tetra-
hedron Lett. 2004, 45, 8375–8377; f) B. Bressel, H.-U. Reissig,
Org. Lett. 2009, 11, 527–530.
[20] a) P. Scheiner, O. L. Chapman, J. P. Lassila, J. Org. Chem. 1969,
34, 813–816; b) R. S. Givens, D. J. Choo, S. N. Merchant, R. P.
Stitt, B. Matuszewski, Tetrahedron Lett. 1982, 23, 1327–1330.
[21] For the conversion of related 1,2-diazines into pyrroles, see: a)
D. L. Boger, R. S. Coleman, J. S. Panek, D. Yohannes, J. Org.
Chem. 1984, 49, 4405–4409; b) D. L. Boger, J. S. Panek, M.
Patel, Org. Synth. 1992, 70, 79–92; c) D. L. Boger, C. W. Boyce,
M. A. Labroli, C. A. Sehon, Q. Jin, J. Am. Chem. Soc. 1999,
121, 54–62.
[5] a) T. Ankner, G. Hilmersson, Tetrahedron 2009, 65, 10856–
10862; b) T. Ankner, G. Hilmersson, Org. Lett. 2009, 11, 503–
506.
[6] a) W. Schade, H.-U. Reissig, Synlett 1999, 632–634; b) R. Pulz,
S. Cicchi, A. Brandi, H.-U. Reissig, Eur. J. Org. Chem. 2003,
1153–1156; c) M. Helms, W. Schade, R. Pulz, T. Watanabe, A.
Al-Harrasi, L. Fisˇera, I. Hlobilová, G. Zahn, H.-U. Reissig,
Eur. J. Org. Chem. 2005, 1003–1019; d) V. Dekaris, R. Pulz, A.
Al-Harrasi, D. Lentz, H.-U. Reissig, Eur. J. Org. Chem. 2011,
3210–3219.
[22] For reviews, see: a) M. Brasholz, H.-U. Reissig, R. Zimmer,
Acc. Chem. Res. 2009, 42, 45–56; b) F. Pfrengle, H.-U. Reissig,
Chem. Soc. Rev. 2010, 39, 549–557; c) L. Bouché, H.-U. Reis-
sig, Pure Appl. Chem. 2012, 84, 23–36.
[23] T. Imamoto, M. Ono, Chem. Lett. 1987, 501–502.
Received: September 8, 2012
Published Online: December 5, 2012
610
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 605–610