Regioselective photochemical rearrangement of N-mesyloxylactams

Article abstract of DOI:10.1002/ejoc.201101222

N-Mesyloxylactams can undergo ring contraction either by C-3 (usually observed) or C-5 migration. C-5 migration can occur when the C-3 migration product possesses ring strain, but it does not usually compete with C-3 migration. The greater preference for C-3 migration is due to the carbonyl oxygen atom, which greatly stabilizes the intermediate. Studies of the photochemical rearrangement of N-mesyloxylactams showed that the lone pair of the carbonyl oxygen atom, and not the degree of the substitution atthe migrating carbon atoms, is the governing factor in the regioselectivity of the reaction. Copyright

Full text of DOI:10.1002/ejoc.201101222

FULL PAPER
DOI: 10.1002/ejoc.201101222
Regioselective Photochemical Rearrangement of N-Mesyloxylactams
Simon Pichette,[a] Samuel Aubert-Nicol,[a] Jean Lessard,*[a] and Claude Spino*[a]
Keywords: N-Mesyloxylactams / Rearrangement / Photochemistry / Ring contraction / Regiochemistry
N-Mesyloxylactams can undergo ring contraction either by
C-3 (usually observed) or C-5 migration. C-5 migration can
occur when the C-3 migration product possesses ring strain,
but it does not usually compete with C-3 migration. The
greater preference for C-3 migration is due to the carbonyl
oxygen atom, which greatly stabilizes the intermediate.
grating carbon atom make this unique rearrangement a
valuable tool for chemists in the field of C–N bond forma-
tion in cyclic substrates.
Introduction
The Lossen rearrangement is a well-known reaction that
permits the formation of a C–N bond by the stereospecific
migration of a C–C bond.[1] There are a number of known
rearrangements of this type, which include the Schmidt,[2]
Hofmann[3] and Curtius[4] rearrangements (Figure 1).[5]
Many use basic or acidic reaction conditions to convert a
carboxylic acid derivative into an isocyanate, which then
reacts with a nucleophile to give an amide derivative.
N-Mesyloxylactams possess two C–C bonds normally
aligned with the leaving group that permit two different mi-
gration pathways that potentially lead to the C-3 migration
product 6 and/or the C-5 migration product 7 (cf.
Scheme 1). Edwards and co-workers observed a fair
amount of C-5 migration product 10 when they irradiated
steroid derivatives 8a,b (Scheme 2).[6] Yet, the formation of
products that arise from C-5 migration was not encountered
in the course of our study of the photochemical rearrange-
ment of N-mesyloxylactams[7c] or N-chlorolactams[7a,7b] (cf.
Scheme 1). This inconsistency compelled us to investigate
factors that control the regiochemistry of migration in or-
der to better predict when such products are likely to arise.
Figure 1. Lossen, Schmidt, Hofmann and Curtius rearrangements.
We recently published the photochemical ring contrac-
tion of five–eight-membered N-mesyloxylactams,[6,7c] the
only rearrangement of this type that works on secondary
Results and Discussion
We started by irradiating smaller bicyclic N-mesyloxy-
amide derivatives (Scheme 1).[8] Rigid bicyclic lactam deriv- lactams that resembled Edwards’ steroidal compounds
atives also underwent this ring contraction efficiently. The (Table 1, Entries 1–4). We found that octahydroquinoline
very mild reaction conditions (254 nm light, MeOH, derivatives 13a and 13b gave a substantial amount of C-5
–78 °C) and the conservation of stereochemistry at the mi- migration products 15 along with the “normal” C-3 mi-
Scheme 1. Photochemical ring contraction of N-mesyloxylactams 5.
gration products 14a and 14b, respectively, in good com-
[a] Département de Chimie, Université de Sherbrooke,
bined yields (Entries 1 and 2). Similarly, the irradiation of
2500 boul. Université, Sherbrooke, Québec J1K 2R1, Canada
octahydroquinoline derivatives 16a or 16b gave mixtures of
bicyclic products 17a and 18 or 17b and 18, respectively
(Entries 3 and 4). Products 15 and 18 have the same stereo-
chemistry at the aminal carbon atom regardless of their
Fax: +1-819-821-8017
E-mail: jean.lessard@usherbrooke.ca
claude.spino@usherbrooke.ca
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101222.
1328
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1328–1335

Regioselective Photochemical Rearrangement of N-Mesyloxylactams
Scheme 2. Examples of the C-5 migration in steroid derivatives.
Table 1. Photochemical ring contractions of N-mesyloxylactams.[a]
[a] Reaction conditions: (i) MeOH, 254 nm, –78 °C; (ii) Et3N. [b] Combined isolated yields of C-3 and C-5 migration products. [c] Isolated
yields of C-3 migration products from three trials. [d] Isolated yield of 32 and 33. Compound 36 was isolated as a mixture with 33 and,
from the analysis of the NMR spectrum of the isolated products, we determined that the yield of 33 was 13% and that of 36 was 30 %.
Eur. J. Org. Chem. 2012, 1328–1335
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1329

S. Pichette, S. Aubert-Nicol, J. Lessard, C. Spino
FULL PAPER
provenance. Presumably, it is derived from a stereoselective to O. This difficult assessment is compounded by the fact
attack of methanol on the corresponding N-acyliminium that we do not know if all or any of these species (except
ion.
5*) is in a photochemically or vibrationally excited state.
We observed a clear decrease in the C-3/C-5 product ra- Very little is known about the photochemistry of N-chloro-
tio when the trans-fused N-mesyloxylactams 13a and 16a lactams, and nothing is known of that of their N-mesyloxy
were irradiated as opposed to the analogous cis-fused N- analogs; however, it is beyond the scope of this article to
mesyloxylactams 13b and 16b (compare Entry 1 with 2 and address photochemical issues.[13]
Entry 3 with 4). This decrease can be explained by the
Experimental results unequivocally show that the mi-
strain in the final products. The trans-fused 5,6-bicyclic gration of C-3 is greatly favoured when there is no built-in
compound 14a and 5,5-bicyclic compound 17a are more bias for the C-5 migration to occur (Table 1, Entries 5 and
strained than their cis-fused counterparts 14b and 17b, 6 and many examples in ref.[7b]). Thus, the C-5 position in
respectively. With the C-5 migration product 15 or 18, bicyclic N-mesyloxylactams 13 and 16 and in Edwards’
which are the same whether one starts with cis- or trans- steroidal substrates must have benefited somewhat from a
fused N-mesyloxylactams, it is expected that the more higher substitution at the migrating carbon atom. In the
strained products 14a and 17a will be formed in lower case of trans-fused 13a and 16a, this was combined with the
amounts than 14b and 17b, respectively. Nonetheless, the developing strain in the C-3 migration products 14a and
formation of 17a in 19% isolated yield is noteworthy as 17a to increase substantially the amount of C-5 migration
trans-fused bicyclo[3.3.0] systems are very strained[9] and product formed (6+ vs. 7+ in Scheme 4).
normally very difficult to prepare.[10]
Entries 5 and 6 of Table 1 lend some support to these
The fact that N-mesyloxylactams 13a–b and 16a–b bear affirmations. Substitution alone is not enough to explain
no substituents α to the carbonyl group is no accident. We the observed C-3/C-5 product ratio as indicated by the ab-
have previously shown that the yield of the C-3-rearranged sence of products 21 and 24a,b. Yet, if C-6 is substituted
product 6 increased with increasing substitution at C-3 (cf. such that a secondary carbocation is formed at that position
Scheme 1). Substrates with no α-substituents typically gave upon C-5 migration, then the formation of product 27 is
35–50% yield of rearranged products, whereas substrates increased (Entry 7). In accordance with our above assertion
with one or two substituents α to the carbonyl group gave on the development of strain in bicyclic products, N-mes-
70–90% yield of products.[7b] In all of the examples in our yloxylactam 25, which is not constrained by the presence of
previous studies, the C-5 position was occupied by a meth- a second ring, produced less of the C-5 migration product
ylene group, and no trace of the corresponding C-5 mi- 27 upon irradiation than its bicyclic homologues 13 and 16.
gration product was ever observed.
If the stability of the developing carbocation were the
We know from previous experiments that the rearrange- controlling element in C-3 vs. C-5 migration, we ought to
ment is not radical in nature.[7b] Although the radical spe- be able to substitute the C-6 position with a heteroatom
cies 5· is possibly formed from the activated complex 5*,[11] to favour the migration of C-5. N-Mesyloxyimides do not
the stereospecificity of the migration precludes radical in- undergo the photolytic ring contraction, presumably be-
termediates during migration. However, it is not certain cause of the inherently high energy associated with a de-
that a true N-acylnitrenium ion 5+ is formed from 5* or 5·, veloping biacylnitrenium ion. We therefore prepared N-
as these species are known to be high-energy cations mesyloxylactam 28 substituted with a methoxy group at C-
(Scheme 3).[12] An alternative is that migration occurs either 6 (Entry 8). Despite the fact that 29 was difficult to isolate
directly from 5* by heterolytic cleavage of the N–O bond (it was partly converted during the reaction to 34 and 35),
or via 5· by a concomitant single electron transfer from N we were able to determine that 39–46% of the products iso-
Scheme 3. Mechanistic pathways of the photochemical rearrangement.
1330
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1328–1335

Regioselective Photochemical Rearrangement of N-Mesyloxylactams
Scheme 4. C-3 and C-5 migration pathways.
Figure 2. Side products formed during the photolysis of N-mesyloxylactams 28 and 31.
lated had undergone a C-3 migration (cf. 34 and 35 in Fig-
ure 2). We did not observe 30 in the crude mixture, but it is
possible that it was too unstable to survive under the reac-
tion conditions. Thus, 31 was constructed in the hope that
its migration products would be more stable (Entry 9). This
was indeed the case and, to our surprise, we were able to
determine a 4:1 ratio of products, still in favour of the C-3-
migration product 32 over the C-5-migration product 33.
Compound 36, which resulted from the loss (or migration?)
of a proton, was also formed in 30% yield (Figure 2).
There can be little doubt that N-acylimminium ions[14]
7b+ and 7c+ are more stable than azaacylium ion[15] 6a+ due
Figure 3. Carbocations and possible transition states involved in
the photochemical rearrangement of N-mesyloxylactams.
to the neighbouring methoxy or p-methoxyphenyl groups
(Figure 3). Yet, even in these cases, C-3 migration was fav-
oured over C-5 migration (Entries 8 and 9, Table 1). We
must therefore conclude that the stability of the developing
The results discussed so far show how effective the amide
carbocation plays a minor role in the outcome of the mi- carbonyl oxygen lone pair of electrons is in the promotion
gration reaction. To be true, this implies that the cation of the rearrangement, which provides a bias for C-3 mi-
character of the transition state (TS) must be weak. It be- gration. It follows that replacing the carbonyl oxygen atom
comes possible to argue that the carbonyl electron pair in or otherwise impeding its donation should remove this bias
TS-6*, which is fixed and perfectly aligned with σ* of the or even reverse it. Complexation of the carbonyl oxygen
migrating bond, assists the departure of the migrating C-3 atom with a Lewis acid seemed like a good way to cancel
carbon atom (Figure 3). There is no delocalization in TS- its electron-donating ability. Several Lewis acids were tested
7* (R = H or Me), and the delocalization from the methoxy in stoichiometric amounts, which included BF3·Et2O,
lone pair (R = OMe) or p-methoxyphenyl group (R = p- Yb(OTf)3 and Ti(iPrO)4, but, surprisingly, in no case was a
MeOPh) is less favourable from an entropy point of view, significant change in ratio observed. We verified that com-
1
because the oxygen electron pairs are not fixed in the proper plexation occurred by H NMR spectroscopy of the start-
alignment. The energy difference between TS-6* and TS-7* ing material. This observation, coupled to the fact that all
will depend on other factors, such as the developing strain the reactions carried out in the presence of a Lewis acid
in the product and the electronics of the migrating carbon were slower than usual, seemed to indicate that the re-
atom (i.e. its migration ability).
arrangement took place only on the uncomplexed substrate.
Eur. J. Org. Chem. 2012, 1328–1335
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1331

S. Pichette, S. Aubert-Nicol, J. Lessard, C. Spino
FULL PAPER
The slower reaction rate could come from a lower concen- attempts to interfere with the carbonyl oxygen atom have
tration of free carbonyl group upon the addition of a Lewis not been successful so far. However, we are currently de-
acid. The absorption band of the N-mesyloxyamide 13b is veloping a series of useful substrates where proper align-
blueshifted when complexed to BF3·Et2O, which slowed the ment of the leaving mesyloxy group with the C-2–C-3 bond
reaction.[16] However, the latter argument does not explain is not possible. C-5 migration in such systems should pre-
why the C-3 migration was still favoured.
vail, and those results will be reported in due course.
Our next approach was to replace the carbonyl oxygen
atom by a sulfur atom, for which the stabilization of a posi-
tive charge is less than that of oxygen[17] (Scheme 5). We Experimental Section
believed that N-mesyloxythiolactam 37 would be able to ab-
General: All reactions were performed under argon in flame-dried
sorb light and undergo the rearrangement. After its irradia-
tion at 254 nm in MeOH at –78 °C, we isolated only the
demesyloxythiolactam 38 in 32% yield.
glassware. Solvents were distilled from potassium/benzophenone
ketyl (Et2O, tetrahydrofuran), from calcium hydride (CH2Cl2, tolu-
ene, benzene, triethylamine) or from 4 Å molecular sieves (MeOH,
1
DMF) prior to use. H NMR spectra were recorded with a 300 or
400 MHz spectrometer. NMR samples were dissolved in CDCl3
(unless specified otherwise), and chemical shifts are reported in
ppm relative to the residual protic solvent. 13C NMR spectra were
recorded with a 75.5 or 100.7 MHz spectrometer. NMR samples
were dissolved in CDCl3 (unless specified otherwise), and chemical
shifts are reported in ppm relative to the solvent. MS analyses were
performed with a GC system spectrometer (30 m length, 25 μ OD,
DB-5ms column) coupled with a mass spectrometer or with a ZAB-
1F micromass spectrometer. HRMS was performed with a ZAB-
1F micromass spectrometer. Reactions were monitored by TLC on
0.25 mm silica gel coated glass plates UV 254. Silica gel (230–
400 mesh) was used for flash chromatography.
Scheme 5. Irradiation of N-mesyloxythiolactam 37.
Nonetheless, this surprising result gave us new insights
into the reaction mechanism. Lactam 38 would be formed
by hydrogen abstraction from methanol by an intermediate
radical 39 if the synchronous rearrangement/SET (39 Ǟ 40)
is not fast enough (Scheme 6). The larger energy gap be-
tween the thioamidyl radical 39 and the N-thioacyl ion 40
could explain this difference in the reaction rate.
General Procedure for Rearrangements: Mesylated hydroxamic acid
(1.00 equiv.) was dissolved in methanol (35 mL). The solution was
transferred to a quartz cell, cooled to –78 °C and irradiated with
254 nm light. After irradiation at –78 °C, the reaction mixture was
transferred to
a round-bottomed flask, and triethylamine
(1.10 equiv.) was added. After stirring at room temperature for
15 min, the solvent was removed under reduced pressure. The resi-
due was dissolved in dichloromethane and water, the phases were
separated, and the aqueous layer was extracted with dichlorometh-
ane (twice). The organic extracts were combined, dried with anhy-
drous magnesium sulfate, filtered and concentrated. The crude
product was purified by flash chromatography on silica gel.
Methyl trans-3a-Methyloctahydro-1H-indole-1-carboxylate (14a):
Carbamate 14a was synthesized according to the general procedure.
The reaction was carried out on a 1.15 mmol scale, the reaction
time was 2 h, the solution was stirred at room temperature for 18 h,
the eluent for the flash chromatography was a mixture of ethyl
acetate and hexanes (20:80 and 50:50), and the product was ob-
tained in 39% yield as a colourless oil. 1H NMR (300 MHz,
CDCl3): δ = 3.61 (s, 3 H), 3.36 (dd, J = 10.4, 3.8 Hz, 2 H), 2.83
(dd, J = 11.8, 3.0 Hz, 1 H), 2.55 (br. s, 1 H), 1.77–1.12 (m, 9 H),
0.82 (s, 3 H) ppm. 13C NMR (75.5 MHz, CDCl3): δ = 156.8 (s),
66.1 (d), 51.7 (q), 45.3 (t), 40.9 (s), 37.7 (t), 36.2 (t), 25.0 (t), 24.7
Scheme 6. Proposed mechanism for the irradiation of N-mesyloxy-
thiolactam 37.
(t), 20.6 (t), 16.8 (q) ppm. IR (neat): ν = 2940, 2927, 2856, 1708,
˜
1456, 1125, 772 cm–1. MS: m/z (%) = 196 (100) [M – H]+, 182 (85)
[M – Me]+. HRMS: calcd. for C11H18NO2 [M – H]+ 196.1337;
found 196.1342.
Conclusions
We have shown that C-5 migration in the photochemical
ring contraction of N-mesyloxylactams is exceptional and
does not usually compete with C-3 migration. We have
shown that the carbonyl oxygen atom of the lactam is the
responsible factor behind this effect. C-5 migration will oc-
cur when the C-3 migration product possesses ring strain,
yet even the very strained trans-5–5-fused ring system was
cis-5-Methoxy-9a-methylhexahydro-1H-pyrrolo[1,2-a]azepin-3(2H)-
one (15): Lactam 15 was synthesized according to the general pro-
cedure. The reaction was carried out on a 1.15 mmol scale, the
reaction time was 2 h, the solution was stirred at room temperature
for 18 h, the eluent for the flash chromatography was a mixture of
ethyl acetate and hexanes (25:75 and 50:50), and the product was
obtained in 16% yield as a colourless oil. 1H NMR (300 MHz,
not enough to shut out C-3 migration completely. Our CDCl3): δ = 5.37 (dd, J = 8.2, 6.0 Hz, 1 H), 3.21 (s, 3 H), 2.40 (dd,
1332 Eur. J. Org. Chem. 2012, 1328–1335
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Regioselective Photochemical Rearrangement of N-Mesyloxylactams
J = 8.8, 7.2 Hz, 2 H), 2.06–1.92 (m, 1 H), 1.85–1.49 (m, 8 H), 1.34
(s, 3 H), 1.27–1.12 (m, 1 H) ppm. 13C NMR (75.5 MHz, CDCl3):
δ = 176.1 (s), 83.0 (d), 63.9 (s), 55.8 (q), 39.5 (t), 34.9 (t), 33.7 (t),
1.40 (m, 8 H), 1.15 (s, 3 H) ppm. 13C NMR (75.5 MHz, CDCl3):
Rotamer A: δ = 155.4 (s), 69.6 (d), 52.1 (q), 50.3 (s), 46.5 (t), 39.1
(t), 38.1 (t), 33.8 (t), 26.3 (q), 24.9 (t) ppm; Rotamer B: δ = 155.4
(s), 69.0 (d), 52.1 (q), 49.4 (s), 46.1 (t), 39.1 (t), 37.5 (t), 32.9 (t),
29.8 (t), 25.8 (q), 25.0 (t), 22.7 (t) ppm. IR (neat): ν = 2936, 2861,
˜
1695, 1461, 1403, 1337, 1081 cm–1. MS: m/z (%) = 197 (1) [M]+,
182 (50) [M – Me]+, 166 (40), 98 (100). HRMS: calcd. for
C11H19NO2 [M]+ 197.1416; found 197.1412. M.p. 40–44 °C.
26.3 (q), 24.9 (t) ppm. IR (neat): ν = 2946, 2867, 1723, 1712, 1690,
˜
1447, 1350, 1243, 1189, 1122, 1086 cm–1. MS: m/z (%) = 183 (15)
[M]+, 168 (60), 140 (100), 81 (30). HRMS: calcd. for C10H17NO2
[M]+ 183.1259; found 183.1254.
Methyl cis-3a-Methyloctahydro-1H-indole-1-carboxylate (14b):
Carbamate 14b was synthesized according to the general procedure.
The reaction was carried out on a 0.72 mmol scale, the reaction
time was 2 h, the solution was stirred at room temperature for 18 h,
Methyl 3,3-Dimethylpyrrolidine-1-carboxylate (20): Carbamate 20
was synthesized according to the general procedure. The reaction
was carried out on a 0.39 mmol scale, the reaction time was 45 min,
the eluent for the flash chromatography was a mixture of ethyl the solution was stirred at room temperature for 15 min, the eluent
acetate and hexanes (25:75 and 50:50), and the product was ob-
tained in 46% yield as a colourless oil. 1H NMR (300 MHz,
CDCl3): δ = 3.67 (s, 3 H), 3.60–3.23 (m, 3 H), 2.27–1.84 (m, 2 H),
1.65–1.03 (m, 8 H), 0.98 (s, 3 H) ppm. 13C NMR (75.5 MHz,
for the flash chromatography was a mixture of ethyl acetate and
hexanes (10:90 and 50:50), and the product was obtained in 37%
yield as a colourless oil. 1H NMR (300 MHz, CDCl3): Rotamer A:
δ = 3.67 (s, 3 H), 3.45 (t, J = 7.0 Hz, 2 H), 3.12 (s, 2 H), 1.62 (q,
CDCl3): Rotamer A: δ = 155.9 (s), 62.6 (d), 52.0 (q), 43.7 (t), 40.3 J = 7.0 Hz, 2 H), 1.05 (s, 6 H) ppm; Rotamer B: δ = 3.66 (s, 3 H),
(s), 34.7 (t), 33.1 (t), 28.9 (t), 27.8 (q), 23.7 (t), 21.7 (t) ppm; Rot-
amer B: δ = 155.5 (s), 62.6 (d), 52.0 (q), 43.5 (t), 39.5 (s), 34.5 (t),
3.38 (t, J = 7.0 Hz, 2 H), 3.05 (s, 2 H), 1.62 (q, J = 7.0 Hz, 2 H),
1.05 (s, 6 H) ppm. 13C NMR (75.5 MHz, CDCl3): Rotamer A: δ =
155.7 (s), 58.8 (t), 52.2 (q), 45.3 (t), 39.2 (t), 37.4 (s), 26.0 (q) ppm;
31.6 (t), 28.1 (t), 27.6 (q), 23.3 (t), 21.7 (t) ppm. IR (neat): ν =
˜
2958, 2936, 2861, 1704, 1452, 1386, 1107 cm–1. MS: m/z (%) = 197 Rotamer B: δ = 155.6 (s), 58.5 (t), 52.2 (q), 44.9 (t), 38.4 (t), 29.6
(30) [M]+, 182 (100) [M – Me]+·, 140 (80), 122 (30). HRMS: calcd.
for C11H19NO2 [M]+ 197.1416; found 196.1418.
(s), 26.0 (q) ppm. IR (neat): ν = 3013, 2964, 2876, 1685, 1460, 1396,
˜
1195, 1118 cm–1. MS: m/z (%) = 157 (30) [M]+, 142 (50) [M –
CH3]+, 101 (100). HRMS: calcd. for C8H15NO2 [M]+ 157.1103;
found 157.1108.
Methyl trans-3a-Methylhexahydrocyclopenta[b]pyrrole-1(2H)-carb-
oxylate (17a): Carbamate 17a was synthesized according to the ge-
neral procedure. The reaction was carried out on a 0.81 mmol scale,
the reaction time was 3 h, the solution was stirred at room tempera-
ture for 18 h, the eluent for the flash chromatography was a mixture
of ethyl acetate and hexanes (20:80 and 75:25), and the product
was obtained in 16% yield as a colourless oil. 1H NMR (300 MHz,
CDCl3): δ = 3.95–3.62 (m, 2 H), 3.63 (s, 3 H), 3.33–3.22 (m, 1 H),
2.15–1.85 (m, 2 H), 1.64–1.23 (m, 6 H), 0.87 (s, 3 H) ppm. 13C
NMR(75.5 MHz, CDCl3): Rotamer A: δ = 156.9 (s), 70.9 (d), 52.7
(t), 52.0 (q), 50.9 (s), 33.5 (t), 30.0 (t), 26.3 (t), 23.3 (t), 18.3 (q)
ppm; Rotamer B: δ = 156.9 (s), 70.9 (d), 52.7 (t), 52.0 (q), 50.2 (s),
Methyl trans-2,4-Dimethylpyrrolidine-1-carboxylate (23a): Carb-
amate 23a was synthesized according to the general procedure. The
reaction was carried out on a 0.50 mmol scale, the reaction time
was 1 h, the solution was stirred at room temperature for 15 min,
the eluent for the flash chromatography was a mixture of dichloro-
methane and hexanes (75:25) followed by a mixture of dichloro-
methane and methanol (10:90). The product was obtained in 67%
1
yield as a colourless oil. H NMR (300 MHz, CDCl3): Mixture of
rotamers: δ = 4.07–3.80 (m, 2 H), 3.65 (s, 6 H), 3.60–3.42 (m, 2 H),
2.97–2.75 (m, 2 H), 2.43–2.24 (m, 2 H), 1.71–1.51 (m, 4 H), 1.17
(d, J = 6.3 Hz, 3 H), 1.12 (d, J = 6.3 Hz, 3 H), 1.00 (d, J = 6.4 Hz,
6 H) ppm. 13C NMR (75.5 MHz, CDCl3): Rotamer A: δ = 155.8
(s), 76.6 (q), 53.6 (t), 53.0 (d), 40.5 (t), 31.2 (d), 20.3 (q), 17.6 (q)
ppm; Rotamer B: δ = 155.2 (s), 76.6 (q), 53.3 (t), 51.9 (d), 41.2 (t),
33.5 (t), 30.0 (t), 26.3 (t), 22.7 (t), 18.3 (q) ppm. IR (neat): ν =
˜
2960, 2935, 2897, 2875, 1689, 1460, 1368, 1132 cm–1. MS: m/z (%)
= 183 (15) [M]+, 182 (20), 168 (65) [M – CH3]+, 140 (100), 81 (20).
HRMS: calcd. for C10H17NO2 [M]+ 183.1259; found 183.1254.
30.3 (d), 20.9 (q), 17.6 (q) ppm. IR (neat): ν = 2964, 2937, 2817,
˜
cis-5-Methoxy-8a-methylhexahydroindolizin-3(5H)-one (18): Lac-
tam 18 was synthesized according to the general procedure. The
reaction was carried out on a 0.86 mmol scale, the reaction time
was 3.5 h, the solution was stirred at room temperature for 18 h,
1701, 1450, 1379, 1188 cm–1. MS: m/z (%) = 157 (5) [M]+, 142 (100)
[M – Me]+. HRMS: calcd. for C8H15NO2 [M]+ 157.1103; found
157.1107.
the eluent for the flash chromatography was a mixture of ethyl Methyl cis-2,4-Dimethylpyrrolidine-1-carboxylate (23b): Carbamate
acetate and hexanes (25:75 and 75:25), and the product was ob-
tained in 16% yield as a colourless oil. 1H NMR (300 MHz,
CDCl3): δ = 5.28 (d, J = 3.8 Hz, 1 H), 3.27 (s, 3 H), 2.63–2.50 (m,
1 H), 2.32 (J = 17.2, 9.8, 2.1 Hz, ddd. 1 H), 2.05–1.33 (m, 8 H),
1.37 (s, 3 H) ppm. 13C NMR (75.5 MHz, CDCl3): δ = 175.0 (s),
78.6 (d), 59.1 (s), 55.3 (q), 39.0 (t), 35.9 (t), 29.9 (t), 29.3 (t), 24.4
23b was synthesized according to the general procedure. The reac-
tion was carried out on a 0.42 mmol scale, the reaction time was
30 min, the solution was stirred at room temperature for 15 min,
the eluent for the flash chromatography was a mixture of ethyl
acetate and hexanes (15:85 and 50:50), and the product was ob-
tained in 52% yield as a colourless oil. 1H NMR (300 MHz,
(q), 15.7 (t) ppm. IR (neat): ν = 2942, 2874, 2849, 2842, 1701, 1690, CDCl3): Mixture of rotamers: δ = 3.90–3.76 (m, 2 H), 3.65 (s, 3
˜
1672, 1393, 1079 cm–1. MS: m/z (%) = 183 (5) [M]+, 168 (100), 152 H), 2.85–2.68 (dd, J = 10.2, 10.2 Hz, 1 H), 2.28–2.14 (m, 1 H),
(100), 140 (30), 98 (50). HRMS: calcd. for C10H17NO2 [M]+
183.1259; found 183.1254.
2.13–1.94 (m, 1 H), 1.33–1.15 (m, 3 H), 1.14–1.03 (m, 1 H), 0.99
(d, J = 6.5 Hz, 3 H) ppm. 13C NMR (75.5 MHz, CDCl3): Rot-
amer A: δ = 155.8 (s), 54.2 (q), 53.6 (t), 51.9 (d), 42.6 (t), 32.5 (d),
20.6 (q), 17.1 (q) ppm; Rotamer B: δ = 155.3 (s), 54.2 (q), 53.8 (t),
Methyl trans-3a-Methylhexahydrocyclopenta[b]pyrrole-1(2H)-carb-
oxylate (17b): Carbamate 17b was synthesized according to the ge-
neral procedure. The reaction was carried out on a 0.86 mmol scale,
the reaction time was 3.5 h, the solution was stirred at room tem-
perature for 18 h, the eluent for the flash chromatography was a
mixture of ethyl acetate and hexanes (25:75 and 75:25), and the
product was obtained in 37% yield as a colourless oil. 1H NMR
51.9 (d), 43.3 (t), 32.3 (d), 21.6 (q), 17.1 (q) ppm. IR (neat): ν =
˜
2960, 2930, 2874, 1705, 1450, 1383, 1203, 1106, 769 cm–1. MS: m/z
(%) = 157 (10) [M]+, 142 (100) [M – Me]+. HRMS: calcd. for
C8H15NO2 [M]+ 157.1103; found 157.1099.
Methyl 2,3,3-Trimethylpyrrolidine-1-carboxylate (26): Carbamate
(300 MHz, CDCl3): δ = 3.67 (s, 3 H), 3.65–3.36 (m, 3 H), 1.94– 26 was synthesized according to the general procedure. The reac-
Eur. J. Org. Chem. 2012, 1328–1335
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1333

S. Pichette, S. Aubert-Nicol, J. Lessard, C. Spino
FULL PAPER
tion was carried out on a 1.27 mmol scale, the reaction time was
4.25 h, the solution was stirred at room temperature for 15 min, the
eluent for the flash chromatography was a mixture of ethyl acetate
and hexanes (10:90 and 40:60), and the product was obtained in
56% yield as a colourless oil. 1H NMR (400 MHz, CDCl3): δ =
3.64 (s, 3 H), 3.51–3.44 (m, 1 H), 3.42–3.24 (m, 2 H), 1.70 (ddd, J
= 10.9, 10.9, 10.9 Hz, 1 H), 1.50 (dt, J = 10.9, 5.5 Hz, 1 H), 1.09–
088 (m, 3 H), 0.96 (s, 3 H), 0.94 (s, 3 H) ppm. 13C NMR
(100.7 MHz, CDCl3): Rotamer A: δ = 156.1 (s), 62.7 (d), 52.3 (q),
44.6 (t), 40.9 (s), 36.9 (t), 27.7 (q), 23.2 (q), 16.8 (q) ppm; Rot-
amer B: δ = 155.8 (s), 62.3 (d), 52.2 (q), 44.3 (t), 40.1 (s), 35.9 (t),
(3) [M – CH3]+, 156 (20) [M – OMe]+, 124 (15), 113 (60), 74 (100).
HRMS: calcd. for C9H16NO3 [M – H]+ 186.1130; found 186.1133.
Methyl 5-(4-Methoxyphenyl)-2,2-dimethylpyrrolidine-1-carboxylate
(32): Carbamate 32 was synthesized according to the general pro-
cedure. The reaction was carried out on a 0.92 mmol scale, the
reaction time was 30 min, the solution was stirred at room tempera-
ture for 15 min, the eluent for the flash chromatography was a mix-
ture of ethyl acetate and hexanes (10:90 and 20:80), and the product
was obtained in 47% yield as a colourless oil. 1H NMR (300 MHz,
CDCl3): Rotamer A: δ = 7.20–7.03 (m, 2 H), 6.84 (d, J = 8.7 Hz,
2 H), 5.06–4.91 (m, 1 H), 3.79 (s, 3 H), 3.48 (s, 3 H), 2.40–2.20 (m,
2 H), 2.02–1.53 (m, 2 H), 1.67 (s, 3 H), 1.43 (s, 3 H) ppm; Rot-
amer B: δ = 7.20–7.03 (m, 2 H), 6.84 (d, J = 8.7 Hz, 2 H), 5.06–
4.91 (m, 1 H), 3.69 (s, 3 H), 3.48 (s, 3 H), 2.40–2.20 (m, 2 H), 2.02–
1.53 (m, 2 H), 1.58 (s, 3 H), 1.37 (s, 3 H) ppm. 13C NMR
27.5 (q), 23.0 (q), 16.0 (q) ppm. IR (CHCl ): ν = 2960, 2875, 1689,
˜
3
1456, 1382, 1340, 1294, 1122, 1051 cm–1. MS: m/z (%) = 171 (15)
[M]+, 156 (100), 115 (80), 97 (15). HRMS: calcd. for C9H17NO2
[M]+ 171.1259; found 171.1264.
1-(1-Methoxyethyl)-5,5-dimethylpyrrolidin-2-one (27): Lactam 27 (100.7 MHz, CDCl3): Rotamer A: δ = 158.3 (s), 155.1 (s), 136.8 (s),
was synthesized according to the general procedure. The reaction
was carried out on a 1.27 mmol scale, the reaction time was 4.25 h,
the solution was stirred at room temperature for 15 min, the eluent
for the flash chromatography was a mixture of ethyl acetate and
hexanes (10:90 and 40:60), and the product was obtained in 10%
126.6 (d), 113.7 (d), 63.5 (s), 62.5 (d), 55.5 (q), 51.9 (q), 38.9 (t),
32.3 (t), 27.3 (q), 25.2 (q) ppm; Rotamer B: δ = 158.3 (s), 156.1 (s),
136.4 (s), 126.8 (d), 113.8 (d), 63.5 (s), 62.1 (d), 55.5 (q), 52.1 (q),
40.2 (t), 31.7 (t), 28.2 (q), 26.5 (q) ppm. IR (CHCl ): ν = 3051,
˜
3
2995, 2935, 1689, 1615, 1587, 1446, 1372, 1298, 1174, 1086 cm–1.
yield as a colourless oil. 1H NMR (300 MHz, CDCl3): δ = 5.33 (q, MS: m/z (%) = 263 (60) [M]+, 248 (90) [M – Me]+, 207 (100), 173
J = 6.4 Hz, 1 H), 3.25 (s, 3 H), 2.42 (t, J = 8.8 Hz, 1 H), 2.41 (t, J
= 6.8 Hz, 1 H), 1.82 (t, J = 8.2 Hz, 2 H), 1.50 (d, J = 6.4 Hz, 3 H),
1.40 (s, 3 H), 1.37 (s, 3 H) ppm. 13C NMR (100.7 MHz, CDCl3): δ
= 175.9 (s), 81.1 (d), 61.7 (s), 55.8 (q), 36.3 (t), 30.2 (t), 28.5 (q),
(75). HRMS: calcd. for C15H21NO3 [M]+ 263.1521; found
263.1523.
6-(4-Methoxyphenyl)-3,3-dimethyl-3,4-dihydropyridin-2(1H)-one (36)
and 1-[Methoxy(4-methoxyphenyl)methyl]-3,3-dimethylpyrrolidin-2-
one (33): Lactams 36 and 33 were synthesized according to the
general procedure. The reaction was carried out on a 0.92 mmol
scale, the reaction time was 30 min, the solution was stirred at room
temperature for 15 min, the eluent for the flash chromatography
was a mixture of ethyl acetate and hexanes (10:90 and 20:80), and
a mixture of the products (2:1) was obtained in 43% yield as a
28.2 (q), 20.8 (q) ppm. IR (CHCl ): ν = 2992, 2942, 2826, 1664,
˜
3
1449, 1403, 1354, 1245, 1189, 1139, 1093, 1051, 910, 847 cm–1. MS:
m/z (%) = 171 (10) [M]+, 156 (100), 115 (60), 98 (35), 84 (20).
HRMS: calcd. for C9H17NO2 [M]+ 171.1259; found 171.1264.
Methyl 5-Methoxy-2,2-dimethylpyrrolidine-1-carboxylate (29) and
Methyl 5,5-Dimethoxy-2-methylpent-2-ylcarbamate (34): Carb-
amates 29 and 34 were synthesized according to the general pro-
cedure. The reaction was carried out on a 0.31 mmol scale, the
reaction time was 1 h, the solution was stirred at room temperature
for 15 min, the solvent was removed under reduced pressure and
the crude mixture was directly purified by flash chromatography.
The eluent was a mixture of ethyl acetate and hexanes (15:85), and
the products were obtained in 17% yield for 29 (not characterized
1
beige solid. Lactam 36: H NMR (300 MHz, CDCl3): δ = 7.33 (d,
J = 8.6 Hz, 2 H), 7.11 (s, 1 H), 6.89 (d, J = 8.6 Hz, 2 H), 5.27 (td,
J = 4.7, 1.6 Hz, 1 H), 3.81 (s, 3 H), 2.30 (d, J = 4.7 Hz, 2 H), 1.21
(s, 6 H) ppm. 13C NMR (100.7 MHz, CDCl3): δ = 177.4 (s), 160.1
(s), 136.1 (s), 127.8 (s), 127.4 (d), 114.4 (d), 100.4 (d), 55.6 (q), 37.1
(s), 36.1 (t), 24.7 (q) ppm. IR (CHCl ): ν = 3544–3023 (br.), 2995,
˜
3
2967, 2939, 2872, 2844, 1689, 1657, 1611, 1460, 1446, 1287, 1178,
because of instability) and in 29% yield for 34 as colourless oils.
1079 cm–1. MS: m/z (%) = 231 (90) [M]+, 216 (100), 203 (35), 188
1
Carbamate 34: H NMR (400 MHz, CDCl3): δ = 4.65 (br. s, 1 H), (65), 162 (35), 146 (40), 134 (50). HRMS: calcd. for C14H17NO
4.34 (t, J = 5.5 Hz, 1 H), 3.59 (br. s, 3 H), 3.31 (s, 6 H), 1.75–1.64 [M]+ 231.1259; found 231.1263. Lactam 33: H NMR (300 MHz,
1
(m, 2 H), 1.63–1.53 (m, 2 H), 1.27 (s, 6 H) ppm. 13C NMR CDCl3): δ = 7.26 (d, J = 8.2 Hz, 2 H), 6.86 (d, J = 8.2 Hz, 2 H),
(100 MHz, CDCl3): δ = 155.5 (s), 104.9 (d), 53.0 (q), 52.5 (s), 51.7 6.19 (s, 1 H), 3.79 (s, 3 H), 3.40 (s, 3 H), 3.23–3.11 (m, 1 H), 2.87–
(q), 35.3 (t), 29.9 (t), 27.6 (t), 27.3 (q) ppm. IR (CHCl ): ν = 3411–
2.74 (m, 1 H), 1.91–1.66 (m, 2 H), 1.23 (s, 3 H), 1.14 (s, 3 H) ppm.
13C NMR (100.7 MHz, CDCl3): δ = 181.1 (s), 159.5 (s), 129.9 (s),
126.2 (d), 113.9 (d), 82.6 (d), 55.9 (q), 55.5 (q), 41.7 (s), 38.0 (t),
˜
3
3137 (br.), 3017, 2960, 2837, 1720, 1509, 1269, 1206, 1121,
1058 cm–1. MS: m/z (%) = 204 (2) [M – CH3]+, 188 (10) [M –
OMe]+, 172 (25), 156 (40), 116 (100), 97 (15). HRMS: calcd. for 34.2 (t), 24.9 (q), 24.5 (q) ppm. IR (CHCl ): ν = 3544–3023 (br.),
˜
3
C9H18NO4 [M – CH3]+ 204.1236; found 204.1239.
2995, 2967, 2939, 2872, 2844, 1689, 1657, 1611, 1460, 1446, 1287,
1178, 1079 cm–1. MS: m/z (%) = 263 (5) [M]+, 248 (50) [M –
Me]+, 232 (20), 135 (100). HRMS: calcd. for C15H21NO3 [M]+
263.1521; found 263.1526.
4-Isocyanato-1,1-dimethoxy-4-methylpentane (35): Isocyanate 35
was synthesized according to the general procedure. The reaction
was carried out on a 0.58 mmol scale, the reaction time was 30 min,
the solution was stirred at room temperature for 18 h, the solvent
was removed under reduced pressure, the crude product was di-
rectly purified by flash chromatography, the eluent was a mixture
of ethyl acetate and hexanes (15:85), and the product was obtained
cis-4a-Methyloctahydroquinoline-2(1H)-thione (38): Thiolactam 38
was synthesized according to the general procedure. The reaction
was carried out on a 0.49 mmol scale, the reaction time was 30 min,
the solution was stirred at room temperature for 15 min, the solvent
was removed under reduced pressure, the crude mixture was di-
rectly purified by flash chromatography, the eluent was a mixture
1
in 17% yield as a colourless oil. H NMR (400 MHz, CDCl3): δ =
4.36 (t, J = 5.4 Hz, 2 H), 3.32 (s, 6 H), 1.76–1.60 (m, 2 H), 1.59–
1.51 (m, 2 H), 1.34 (s, 6 H) ppm. 13C NMR (100 MHz, CDCl3): δ of ethyl acetate and hexanes (5:95), and the product was obtained
1
= 122.6 (s), 104.5 (d), 58.1 (s), 53.1 (q), 38.6 (t), 30.4 (q), 28.0 (t)
in 30% yield as a colourless oil. H NMR (300 MHz, CDCl3): δ =
ppm. IR (CHCl ): ν = 3013, 2982, 2939, 2837, 2259, 1463, 1445, 8.64 (br. s, 1 H), 3.16–3.08 (m, 1 H), 2.93 (t, J = 6.6 Hz, 2 H),
˜
3
1386, 1364, 1125, 1072 cm–1. MS: m/z (%) = 186 (3) [M – H]+, 172
1.90–1.79 (m, 1 H), 1.72 (dt, J = 13.6, 6.6 Hz, 1 H), 1.62–1.32 (m,
1334
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1328–1335

Regioselective Photochemical Rearrangement of N-Mesyloxylactams
6 H), 1.31–1.14 (m, 2 H), 1.03 (s, 3 H) ppm. 13C NMR (100.7 MHz,
CDCl3): δ = 202.3 (s), 60.7 (d), 36.4 (t), 33.1 (t), 31.4 (t), 30.7 (s),
Blackwell, G. Grue-Sørensen, O. E. Edwards, Can. J. Chem.
1997, 75, 857.
[7]
[8]
a) A. Drouin, J. Lessard, Tetrahedron Lett. 2006, 47, 4285; b)
D. K. Winter, A. Drouin, J. Lessard, C. Spino, J. Org. Chem.
2010, 75, 2610; c) D. K. Winter, A. Drouin, S. Pichette, S. Aub-
ert-Nicol, J. Lessard, C. Spino, J. Org. Chem. 2011, 76, 164.
The rearrangement of cyclic imides under basic conditions is
possible, but the rearrangement takes place presumably via an
acyclic intermediate; see, for example: X. Jianga, J. Wanga, J.
Hua, Z. Gea, Y. Hua, H. Hua, D. F. Covey, Steroids 2001, 655–
662.
29.3 (t), 25.7 (q), 21.8 (t), 21.0 (t) ppm. IR (CHCl ): ν = 3354,
˜
3
3280–3111 (br.), 3059, 2942, 2865, 1565, 1530, 1449, 1301,
1093 cm–1. MS: m/z (%) = 199 (1) [M]+, 183 (100) [M – Me]+, 150
(50) [M – (Me + SH)]+, 88 (50). HRMS: calcd. for C10H17NOS
[M]+ 199.1031; found 199.1027.
CCDC-840317 (for 15) and -840318 (for 49a) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[9]
S. Chang, D. McNally, S. Shary-Tehrany, M. J. Hickey, R. H.
Boyd, J. Am. Chem. Soc. 1970, 92, 3109–3118.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures and characterization for all new
compounds in the synthesis of N-mesyloxylactams 13a, 13b, 16a,
16b, 19, 22a, 22b, 25, 28 and 31.
[10]
See, for example: a) E. Piers, A. M. Kaller, Tetrahedron Lett.
1996, 37, 5857–5860; b) L. Van Hijfte, R. D. Little, J. L. Pe-
tersen, K. D. Moeller, J. Org. Chem. 1987, 52, 4647–4661.
Radical reactions are commonplace when N-chlorolactams are
irradiated; nothing is known about N-mesyloxylactams; see: a)
J. Lessard, Y. Couture, M. Mondon, D. Touchard, Can. J.
Chem. 1984, 62, 101–104; b) B. Daoust, J. Lessard, Tetrahedron
1999, 55, 3495–3514; R. Sutcliffe, D. Griller, J. Lessard, K. U.
Ingold, J. Am. Chem. Soc. 1981, 103, 624–628.
Acylnitrenium ions are formed if an electron-donating group,
such as a methoxy group, is present on the nitrogen atom. For
the stability of such cations, see: a) M. Kawase, T. Kitamura,
Y. Kikugawa, J. Org. Chem. 1989, 54, 3394–3403; b) Y. Kiku-
gawa, M. Kawase, J. Am. Chem. Soc. 1984, 106, 5728–5729; c)
S. A. Glover, A. P. Scott, Tetrahedron 1989, 45, 1763–1776.
[11]
[12]
Acknowledgments
We thank the Natural Sciences and Engineering Research Council
of Canada (NSERC), the Fonds québécois de la recherché sur la
nature et les technologies (FQRNT) and the Université de Sher-
brooke for financial support.
[1] a) W. Lossen, Ann. Chem. Pharm. 1869, 313–325; b) L. Kürti,
B. Czakó in Strategic Applications of Named Reactions in Or-
ganic Synthesis, Elsevier Academic Press, Burlington, MA,
USA, 2005, pp. 266–267; c) P. A. S. Smith in Molecular Re-
arrangements (Ed.: P. de Mayo), Interscience Publ., New York,
1963, vol. 1, pp. 457.
[2] a) K. F. Z. Schmidt, Angew. Chem. 1923, 35, 511; b) H. Wolff,
Org. React. 1946, 307–336; c) G. I. Koldobskii, G. F. Treshch-
enko, E. S. Gerasimova, L. I. Bagal, Russ. Chem. Rev. 1971,
835; d) G. I. Koldobskii, V. A. Ostrovskii, B. V. Gidaspov, Russ.
Chem. Rev. 1978, 1084–1094; e) J. Aube, G. L. Milligan, J. Am.
Chem. Soc. 1991, 113, 8965–8966.
[3] a) E. S. Wallis, J. F. Lane, Org. React. 1946, 267–306; b) P. A. S.
Smith, Trans. N. Y. Acad. Sci. 1969, 504–515; c) S. S. Jew, H. J.
Park, M. S. Park, Y. S. Cho, Ind. Chem. Libr. 1991, 147–153.
[4] a) E. Buchner, T. Curtius, Ber. Dtsch. Chem. Ges. 1885, 2371–
2377; b) Y. Yukawa, Y. Tsuno, J. Am. Chem. Soc. 1959, 81,
2007–2012; c) E. Fahr, L. Neumann, Angew. Chem. 1965, 77,
591; d) S. Linke, G. T. Tisue, W. Lwowski, J. Am. Chem. Soc.
1967, 89, 6308–6310; e) A. Rauk, P. F. Alewood, Can. J. Chem.
1977, 55, 1498–1510.
[13] To the best of our knowledge, the photochemistry of N-chlo-
rolactams has not been studied.
[14] For reviews on N-acyliminium chemistry, see: a) W. N. Speck-
amp, M. J. Moolenaar, Tetrahedron 2000, 56, 3817–3856; b) A.
Padwa, Pure Appl. Chem. 2003, 75, 47–62; c) B. E. Maryanoff,
H.-C. Zhang, J. H. Cohen, I. J. Turchi, C. A. Maryanoff, Chem.
Rev. 2004, 1431–1628.
[15] Azaacylium ions have only been observed in the gas phase; see,
for example: a) E. C. Meurer, A. A. Sabino, M. N. Eberlin,
Anal. Chem. 2003, 75, 4701–4709; b) L. A. B. Moraes, M. N.
Eberlin, J. Mass Spectrom. 2002, 162–168; c) A. B. Lemos, R.
Sparrapan, M. N. Eberlin, J. Mass Spectrom. 2003, 305–314;
d) G. E. Reid, S. E. Tichy, J. Perez, R. A. J. O’Hair, R. J. Simp-
son, H. I. Kenttaemaa, J. Am. Chem. Soc. 2001, 123, 1184–
1192; e) T. W. Bentley, J. Org. Chem. 2008, 73, 6251–6257.
[16] Hypsochromic shifts have been observed for n–π* transitions in
protic solvents: a) P. Klán, J. Wirz in Photochemistry of organic
compounds, John Wiley & Sons Ltd, Chippenham, Wiltshire,
UK, 2009, p. 294; b) N. J. Turro in Modern Molecular Photo-
chemistry, University Science Books, Reading, MA, USA,
1991, pp. 107–108.
[5] For general books and reviews, see: a) L. Kürti, B. Czakó, Stra-
tegic Applications of Named Reactions in Organic Synthesis, El-
sevier Academic Press, Burlington, MA, USA, 2005.
[6] Edwards and co-workers observed this rearrangement in the
course of their work on steroid derivatives; see: B. A.
[17] C. F. Rodriquez, A. C. Hopkinson, J. Mol. Struct. 1987, 55–
68.
Received: August 19, 2011
Published Online: January 9, 2012
Eur. J. Org. Chem. 2012, 1328–1335
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1335