Palladium-catalyzed allylic substitution between C-based nucleophiles and 6-azabicyclo[3.1.0]-hex-3-en-2-oxy derivatives: A new selectivity paradigm

Article abstract of DOI:10.1016/j.tet.2020.131182

The reaction between α-hydroxy-(or α-acetoxy) cyclopenten-aziridines (6-azabicyclo[3.1.0]hex-3-en-2-ols or acetates) and C-based nucleophiles in the presence of Pd(0)-catalysis was investigated. In all the cases studied, the reaction was totally regio- and diastereo-selective, affording a single adduct in moderate to good yields. Specifically, attack of the nucleophile at position 3 of the cyclopentene moiety, anti to the vicinal oxy group, with vinylogous ring opening of the aziridine ring, was observed. When the carbon acid is a very acidic methylene (pK_{a (DMSO)} ≤ 7.3), the resulting adduct is a zwitterion, resulting from an intramolecular proton transfer between the amino group and the carbon acid moiety taking place after the C–C bond formation. A plausible mechanism for this transformation is put forward.

Full text of DOI:10.1016/j.tet.2020.131182

Journal Pre-proof
Palladium-catalyzed allylic substitution between C-based nucleophiles and 6-
azabicyclo[3.1.0]-hex-3-en-2-oxy derivatives: A new selectivity paradigm
João A.C. Oliveira, Gredy Kiala, Filipa Siopa, Aurélie Bernard, Geoffrey Gontard, Julie
Oble, CarlosA.M. Afonso, Giovanni Poli
PII:
S0040-4020(20)30317-3
https://doi.org/10.1016/j.tet.2020.131182
TET 131182
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 8 February 2020
Revised Date: 31 March 2020
Accepted Date: 3 April 2020
Please cite this article as: Oliveira JoãAC, Kiala G, Siopa F, Bernard Auré, Gontard G, Oble J,
Afonso CM, Poli G, Palladium-catalyzed allylic substitution between C-based nucleophiles and 6-
azabicyclo[3.1.0]-hex-3-en-2-oxy derivatives: A new selectivity paradigm, Tetrahedron (2020), doi:
https://doi.org/10.1016/j.tet.2020.131182.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

.
Palladium-catalyzed allylic substitution
between C-based nucleophiles and 6-
azabicyclo[3.1.0]-hex-3-en-2-oxy derivatives:
a new selectivity paradigm
Leave this area blank for abstract info.
João A. C. Oliveira, Gredy Kiala, Filipa Siopa,* Aurélie Bernard, Geoffrey Gontard, Julie Oble,* Carlos A. M.
Afonso* and Giovanni Poli*

1
Tetrahedron
journal homepage: www.elsevier.com
Palladium-catalyzed allylic substitution between C-based nucleophiles and 6-
azabicyclo[3.1.0]-hex-3-en-2-oxy derivatives a new selectivity paradigm
:
João A. C. Oliveira,a,b Gredy Kiala,a,b Filipa Siopa,a,b Aurélie Bernard,a Geoffrey Gontard,a Julie Oble,a*
Carlos A. M. Afonsob* and Giovanni Polia*
a Sorbonne Université, Faculté des Sciences et Ingénierie, CNRS, Institut Parisien de Chimie Moléculaire, IPCM, 4 place Jussieu, 75005 Paris, France
b Research Institute for Medicines (iMed.ULisboa), Faculty of Pharmacy, Universidade de Lisboa, Av. Prof. Gama Pinto, 1649-003 Lisboa, Portugal
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
The reaction between α-hydroxy-(or α-acetoxy) cyclopenten-aziridines (6-azabicyclo[3.1.0]hex-
3-en-2-ols or acetates) and C-based nucleophiles in the presence of Pd(0)-catalysis was
investigated. In all the cases studied, the reaction was totally regio- and diastereo-selective,
affording a single adduct in moderate to good yields. Specifically, attack of the nucleophile at
position 3 of the cyclopentene moiety, anti to the vicinal oxy group, with vinylogous ring
opening of the aziridine ring, was observed. When the carbon acid is a very acidic methylene
(pKa (DMSO) ≤ 7.3), the resulting adduct is a zwitterion, resulting from an intramolecular proton
transfer between the amino group and the carbon acid moiety taking place after the C-C bond
formation. A plausible mechanism for this transformation is put forward.
Available online
Keywords:
Palladium catalysis
Bicyclic aziridine
Tsuji-Trost reaction
2009 Elsevier Ltd. All rights reserved
.
Corresponding authors.
E-mail address: filipasiopa@ff.ulisboa.pt; julie.oble@sorbonne-universite.fr; carlosafonso@ff.ulisboa.pt; giovanni.poli@sorbonne-
universite.fr.

2
Tetrahedron
1. Introduction
gram scale, e.g. 1.9 g of 6-allyl-6-azabicyclo[3.1.0]hex-3-en-2-ol
(2a) and 1.8 g of 6-butyl-6-azabicyclo[3.1.0]hex-3-en-2-ol (2b),
increasing the process productivity regarding to the reported
batch process, namely: 3:1 for 2a [32] (129.3 mg∙h−1 vs 39.4
mg∙h−1) and 18:1 for 2b [29] (87.0 mg∙h−1 vs 4.8 mg∙h−1) (Scheme
3).
Aziridines are the smallest members of the aza-heterocycle
family [1,2]. Due to their considerable ring strain, these
molecules show in general a high reactivity that makes them
major intermediates for organic synthesis. Furthermore, several
biologically active natural products [3], or analogs of them,
incorporate the aziridine motif.
In 1972, Kaplan et al. [4] reported an interesting
photochemical conversion of N-alkylpyridinium salts, such as A,
into the corresponding α-hydroxycyclopenten-aziridines (6-aza-
bicyclo[3.1.0]hex-3-en-2-ol)such as
D
(Scheme 1). The
mechanism of this photoelectrocyclization involves a π→π*
transition of the aromatic nucleus, followed by the transient
generation of the 6-aza[3.1.0]bicyclic allylic cation B, in
equilibrium with the corresponding azoniabenzvalene cation C.
Subsequent is in situ interception by the water (or alcohol)
solvent from the least hindered face affords the α-
hydroxycyclopenten-aziridines D photoproduct.
Scheme 3. Process intensification for the synthesis of α-
hydroxycyclopenten-aziridines using continuous UV-light
photoflow home-made reactors [29, 31].
Considering the peculiar structure of the above described α-
oxycyclopenten-aziridines in connection with our long-standing
interest in Pd-catalyzed allylations [33-36], we were intrigued by
the thought of investigating the behavior of such cyclic substrates
against soft carbon-based pro-nucleophiles under Pd(0) catalysis.
In particular, we reasoned that such an approach could have
opened the way to a hitherto unexplored C-C functionalization of
α-oxycyclopenten-aziridines.
The potential of this seminal work remained unexploited for a
decade, until Mariano [5] found that in situ solvolytic opening of
the α-oxycyclopenten-aziridines photoproduct can take place
regio- and stereospecifically in a SN2 mode. Since the global
transformation provided a cyclopentene motif with total trans
control at C3/C4 and C4/C5 (Scheme 2), the relevance of this
method for the synthesis of aminocyclitols was immediately
apparent.
Some preliminary considerations are worthy, before
presenting the results. The cyclopentene motif in these α-
oxycyclopenten-aziridines E is bis-allylically substituted, one
allylic position being occupied by an oxy (alcohol, ether or ester)
function, the other one being substituted with the trans C–N bond
making part of the fused aziridine. In view of the known anti
approach of a Pd(0) complex to an allylic leaving group, the
initial generation of two alternative η3-allylpalladium complexes
F and G may be expected, depending on the allylic group that
prefers to leave. Furthermore, the known isomerization of η3-
allylpalladium complexes via SN2 type substitution [36] may
allow generation of intermediates H and/or I prior of the
nucleophilic trapping. Finally, each of the four η3-allylpalladium
complexes F, G, H and I may in principle give rise to two
regioisomers. As a result, prediction of product selectivity in this
reaction is far from being straightforward (Scheme 4).
Scheme 1. Kaplan’s pioneering studies: photochemical conversion
of 1-methylpyridin-1-ium chloride A into α-hydroxycyclopenten-
aziridine D[4].
In the event, the Pd(0)-catalyzed reaction between C-based
soft nucleophiles and α-hydroxycyclopenten-aziridines or
acetates took constantly place in a totally regio- and diastereo-
selective way, generating a single aminocyclopentene structure
resulting from attack of the carbon nucleophile at C3 of
intermediate G (Scheme 4), anti to the vicinal oxy group, with
vinylogous aziridine opening.
Scheme 2. Photochemical electrocyclization of N-methyl-pyridinium
chloride followed by methanolytic α-oxycyclopenten-aziridine
opening [5].
Subsequently, several groups, including those of Mariano [6-
8], Burger [9-12], Ganem [13] and Penkett [14-16] revisited and
extended the scope of this chemistry, achieving synthetically
important targets such as (+)-mannostatin
castanospermine [18], (–)-swainsonine
A
[17], (+)-
[19], the
aminocyclopentitol cores of allosamidine [20], trehazolin [21]
and 3-amino-3-deoxy sugars [22], [23-28]. More recently, we
contributed to this topic, by accomplishing the generation and the
subsequent ring opening of the resulting α-oxycyclopenten-
aziridines in water at physiological pH using heteroatom-based
nucleophiles, including the peptide hormone salmon calcitonin
(sCT) [29], and by performing the photoelectrocyclization /
nucleophilic interception sequence using different home-made
continuous UV-light photoflow reactors [30, 31]. Our process
allowed the production of α-hydroxycyclopenten-aziridines in

3
Scheme 4. Potential regio- and stereo-selectivities in the Pd(0)-
catalyzed allylic substitution between α-oxycyclopenten-aziridines
and carbon-based soft nucleophiles. In red the experimentally
observed reactivity from G. Brackets around palladium atom in a
charged or neutral complex intend to render implicit the dative
ligands.
product in 84% NMR yield (Table 1, entry 11). Here again, the
loading of the catalytic system could be reduced to [Pd(OAc)2 (5
mol%) / PPh3 (15 mol%)] recording even a slight yield increase
(Table 1, entry 12).
Table 1. Optimization of the reaction between methyl acetoacetate
3b or Meldrum's acid 3c with α-acetoxycyclopenten-aziridine 2c.a
2. Results/Discussion
2.1. Optimization of the reaction conditions
Dimethyl malonate 3a (pKa
= 13; pKa
= 15.9),
(DMSO)
(H2O)
methyl acetoacetate 3b (pKa (H2O) = 11) and Meldrum’s acid 3c
(pKa (H2O) = 5; pKa (DMSO) = 7.3) were first considered as C-based
pro-nucleophiles for the reaction with α-acetoxycyclopenten-
aziridine 2c, in turn obtained by standard acetylation of 2b. As to
the catalyst, we opted for the system [Pd(OAc)2 (10 mol%), PPh3
(30 mol%)] with or without NaH, in THF (Scheme 5) [37].
Dimethyl malonate 3a gave no reaction, whether with or without
a base, whereas methyl acetoacetate 3b displayed a modest
reactivity when working in the presence of 1.2 equiv of NaH
(4cb, 50% NMR yield). In contrast, Meldrum’s acid 3c afforded
aminocyclopentene 4cc in 63% NMR yield under base free
conditions. The above results suggest that the acidity of the
activated methylene of the nucleophile has to fit in an appropriate
pKa window to allow the allylic substitution to take place
(Scheme 5).
Entry NuH
[Pd] (x)
L (y)
Solvent (h)
Product,
yield %b
1
2
3
4
3b
3b
3b
3b
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
PPh3 (30)
dppe (20)
dppf (20)
THF (4)
4cb, 50
4cb, 41
4cb, 32
4cb, 49
THF (18)
THF (18)
MeDCHB THF (18)
(30)
5
6
3b
3b
3b
3b
3c
3c
3c
3c
[Pd(C3H5)Cl]2 (10) dppe (20)
THF (18)
Et2O (4)
4cb, 8
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (5)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (5)
PPh3 (30)
PPh3 (30)
PPh3 (5)
PPh3 (30)
PPh3 (30)
PPh3 (30)
PPh3 (15)
4cb, 59
4cb, 72
4cb, 74
4cc, 63
4cc, 73
4cc, 84
4cc, 88
7
DIOX (4)
DIOX (18)
THF (3)
8
9
10
11
12
DIOX (1.5)
Et2O (3.5)
Et2O (5)
a Reaction conditions: 2c (1.0 equiv), 3b or 3c (1.2 equiv), solvent (0.5
b
M), NaH (1.2 equiv: entry 1 to 8, or 0 equiv: entry 9 to 12).
Determined by 1H NMR analysis of the crude product using 1,4-
dinitrobenzene as internal standard. DIOX: 1-4-dioxane; Dppe: 1,2-
bis(diphenylphosphino)ethane;
dppf:
1,1′-ferrocenediyl-
bis(diphenylphosphine),
methylbiphenyl.
MeDCHB: 2-dicyclohexylphosphino-2′-
2.2. Scope of the reaction
With the optimized reaction conditions in hand, we passed to
evaluate the scope of this allylic substitution starting with the use
of the base-free conditions, associated to the more acidic active
methylenes (Table 2). The reaction between the free alcohol
aziridine 2b and Meldrum’s acid 3c was also successful,
affording the corresponding allylic product 4bc, although its
isolated yield was lower than that of the reaction starting from the
corresponding aziridine acetate 2c (Table 2, compare entries 1
and 2). Use the α-acetoxycyclopenten-aziridine 2d carrying a
different substituent at the nitrogen atom gave with Meldrum’s
acid the expected product in fair yield of 52 % (Table 2, entry 3).
Reaction of the α-acetoxycyclopenten-aziridine 2c with the
methylated Meldrum’s acid 3d gave also the expected product
4cd, although with a low yield. It should be noted that, in contrast
to the previously obtained products, product 4cd cannot stay as
its inner salt form, due to the lack of a second acidic proton. This
is also confirmed by the very inferior polarity of 4cd with respect
to that of the other substitution products. The reaction between
Scheme 5. Preliminary allylic substitution tests between methyl
acetoacetate 3b or Meldrum’s acid 3c and α-acetoxycyclopenten-
aziridine 2c.
The influence of the nature of the palladium catalyst, the
ligand as well as the solvent were investigated next. The results
are presented in Table 1.
In the reaction between methyl acetoacetate 3b and the α-
acetoxycyclopenten-aziridine 2c in presence of NaH, the use of
phosphines other than PPh3 (mono- or bidentate) did not allow
further improvements (Table 1, entries 1-4). Switch to {[Pd(η3-
C3H5)Cl]2 (10 mol%)/ dppe (20 mol%)} as catalytic system led to
an important yield drop (Table 1, entry 5). Variation of the
solvent was next addressed. While use of Et2O led to an NMR
yield essentially comparable to that of the experiment in THF
(Table 1, entries 6 and 1), 1,4-dioxane significantly increased this
result (Table 1, entry 7). Finally, the catalytic loading could be
reduced to [Pd(OAc)2 (5 mol%) / PPh3 (15 mol%)] in 1,4-
dioxane, without a significant yield erosion of the product 4cb
(Table 1, entry 8). Control experiments carried out by omitting
the Pd source as well as the ligand afforded no trace of product 4,
giving mainly recovered starting materials, which confirmed the
need of the catalytic system for the allylic substitution (see SI).
N,N-dimethyl barbituric acid (pKa
= 4) and the α-
(H2O)
acetoxycyclopenten-aziridine 2c or the hydroxy derivative 2b,
paralleled the behavior observed with Meldrum’s acid, giving in
both cases the expected allylic substitution product with better
yields in the case of the acetate partner (Table 2, compare entries
5 and 6). Finally, the two reactions could also be extended to
ethyl nitroacetate (pKa
entries 7 and 8), although the corresponding products were
obtained in a modest yield. At last, the dimedone bearing also a
= 5.79; pKa
= 9.1) (Table 2,
The influence of the solvent was also studied for the reaction
with Meldrum’s acid 3c under base free conditions (Table 1,
entries 9-11). In this case, use Et2O afforded 4cc as the sole
(H2O)
(DMSO)

4
Tetrahedron
very acidic methylene (pKa (H2O) = 5.2; pKa (DMSO) = 11.2) was
structural assignment and unveiled the relative trans
stereochemistry around the three contiguous stereogenic centers
of the cyclopentene structure [39].
tested, but unfortunately led to an unanalyzable complex mixture.
Table 2. Substrate scope of the Pd(0)-catalyzed allylic substitution
under base-free conditions.a
Analogies among all the 1H and 13C NMR spectra of the
allylation products allowed to tentatively assign the same trans
relative configuration to all the other obtained products.
Furthermore, this X-ray structure reveals that 4ce exists as an
inner salt, (Figure 1). This feature is in line with the high polarity
observed for this product and the other products of allylic
substitution derived from the most acidic active methylenes.
However, the equilibrium between the non-ionic and the inner
salt forms in the allylic substitution products depends on the
degree of acidity of the proton left on the active methylene
moiety after the substitution, and only those deriving from the
most acidic pro-nucleophiles are found as inner salts [40].
Figure 1. Top: Single-crystal X-ray structural analysis of the adduct
4ce. Indicated distance between an oxygen atom of the N-dimethyl
barbituric fragment and a calculated H atom on the N atom. Bottom:
equilibrium between the non-ionic and the inner salt forms of the
allylic substitution product, whose position depends on the acidity of
the proton left on the active methylene moiety after the substitution.
a
b
Isolated yields. Isolated product with traces of triphenylphosphine oxide
2.4. Proposal of the mechanism
(in parenthesis, NMR yield using 1,4-dinitrobenzene as internal standard)
We then evaluated the allylations of the less acidic active
methylenes, which required the use of NaH as base. In the event,
methyl acetoacetate 3b reacted with the acetoxyaziridine 2c as
well as with the free alcohol 2b to give the corresponding
allylation products 4cb [38] and 4bb in 74% NMR yield and
65% isolated yield, respectively (Scheme 6). Finally, other less
acidic active methylenes (pKa (H2O) = 11-13), such as malonitrile,
disulfone or methyl (phenylsulfonyl)acetate, were tested and
showed no reactivity or led to intractable mixtures (See SI).
To gain insight into the chronology of the proton transfers
involved in the transformation under study, we ran 1H NMR
spectra of equimolar mixtures of the couples [Meldrum’s acid 3c
/ aziridine 2b] and [Meldrum’s acid 3c / NEt3] in CDCl3 (Figure
2). Interestingly, while the spectrum of the couple [Meldrum’s
acid 3c / aziridine 2b] showed no significant deviation with
respect to the simple superposition of the spectra of the two
single components, the one of [Meldrum’s acid 3c / NEt3]
showed the absence of the methylene peak of Meldrum’s acid
(Figure 2). The above result unequivocally shows that, in contrast
to the behavior of a classical amine such as NEt3, aziridine 2b is
not capable of deprotonating to an apparent extent the active
methylene. This result, which is in line with the known weak
basicity of the aziridine nitrogen atom [41-43], indicates that in
our reaction deprotonation of the active methylene has to take
place only after aziridine opening (Scheme 7).
Scheme 6. Pd(0)-catalyzed allylic substitution in the presence of
base with methyl acetoacetate 3b.
2.3. Structural assignment of the allylation products
In all the cases studied, product NMR analysis allowed to
unambiguously establish that the reaction gave rise to a single
product, constantly resulting from attack of the carbon
nucleophile to the olefinic C3 carbon atom vicinal to the oxy
function with vinylogous aziridine opening. The single-crystal X-
ray structural analysis of the adduct 4ce confirmed the above

5
Scheme 8. Proton transfer equilibria after the oxidative addition step.
Top: when using an hydroxycyclopenten-aziridine. Bottom: when
using an acetoxycyclopenten-aziridine.
1
Figure 2. Top: H NMR spectrum of an equimolar mixture of the
couple [Meldrum’s acid 3c / aziridine 2b]; bottom: 1H NMR
spectrum of an equimolar mixture of the couple [Meldrum’s acid 3c /
NEt3].
3. Conclusion
A plausible mechanism for this annulation, exemplified for the
reaction between N,N-dimethyl barbituric acid 3e and aziridine
2c, is presented in Scheme 7.
In conclusion, in this study we have developed the first Pd-
catalyzed allylic substitution between 2-oxy-cyclopenten-
aziridines and active methylenes. In all the cases studied, the C-
nucleophile substituted the position C3 of the cyclopentene
moiety, anti to the allylic oxy group, with vinylogous ring
opening of the aziridine ring, according to a syn SN2’ mode. This
new selectivity paradigm adds to the already known ones,
enriching the chemistry of the 2-oxy-cyclopenten-aziridines.
Interestingly, the equilibrium between the non-ionic and the inner
salt forms in the allylic substitution products depends on the
acidity of the proton left on the active methylene moiety after the
substitution (Scheme 9).
Scheme 7. Proposed mechanism for the Pd(0)-catalyzed allylic
substitution of aziridine 2c.
Coordination of the Pd(0) complex by the alkene function of
the substrate from the face syn to the acetate group triggers
aziridine opening with concomitant formation of the zwitterionic
η3-allylpalladium complex J (Scheme 7). Following proton
Scheme 9. Reactivity modes in the allylic substitution of 2-oxy
cyclopenten-aziridines.
4. Experimental section
transfer between
J and the active methylene generates
intermediate K, ready for an outer sphere C-allylation to afford
intermediate L. Finally, Pd(0) decoordination and proton transfer
generates the final allylation product 4ce and regenerates the
Pd(0) active catalyst.
4.1. General considerations
All reactions were carried out under an argon atmosphere by
standard syringe and septa techniques. Et2O was dried on a
Mbraun purification system MB SPS-800. 1,4-Dioxane was
distilled from Na / benzophenone keep under N2 atmosphere.
NMR spectra (1H, 13C) were recorded on a Bruker Fourier 300
or Bruker AM 300 MHz or Bruker AVANCE 400 MHz
spectrometer. NMR experiments were carried out at room
temperature in CDCl3 or D2O. Chemical shifts are given in parts
per million (ppm). The proton signal of residual non-deuterated
A little variant of the above mechanism has to be taken into
account when the cyclopenten-aziridines carrying the free alcohol
are used (Scheme 8). In this case, after the oxidative addition,
intramolecular proton transfer from the alcohol function to amide
anion (M ⇆ N) competes with the direct deprotonation of the
active methylene (P + O ⇆ P+H+ + O-H+) (Scheme 8).

6
Tetrahedron
solvent (δ 7.26 ppm for CHCl3) was used as an internal
(101 MHz, CDCl3) δ 167.8, 143.3, 121.0, 102.1, 80.0, 74.8, 70.8,
51.0, 44.9, 28.9, 25.7, 25.6, 20.0, 13.71. IR (cm-1): 2965, 2184,
2139, 2047, 2030, 2007, 1541, 1460, 1400, 1371, 1259, 1203,
1164, 1134, 1105, 1061, 998, 914, 816, 778, 730, 645, 586, 522,
425, 391. HRMS (ESI) calcd for C15H24NO5 [M+H]+: 298.1649;
found: 298.1650.
1
reference for H spectra. The carbon signal of deuterated solvent
(δ 77.16 ppm for CDCl3) was used as an internal reference for
13C spectra. Coupling constants (J) are given in Hertz (Hz). For
previously unknown compounds,
a
combination of 2D
experiments (HSQC, COSY and HMBC) were often used to
complete assignment of 1H and 13C signals. IR spectra were
5-((4R*,5R*)-5-acetoxy-4-((3-
acetoxypropyl)ammonio)cyclopent-2-en-1-yl)-2,2-dimethyl-4-
oxo-4H-1,3-dioxin-6-olate 4dc.
recorded with
a
Tensor 27 (ATR Diamond) Bruker
spectrophotometer. IR spectra were reported as characteristic
bands (cm-1). High-resolution mass spectra (HRMS) were
recorded using a mass spectrometer MicroTOF from Bruker with
an electron spray ion source (ESI) and a TOF detector or using a
mass spectrometer from Thermo Fisher Scientific with an
electron spray ion source (ESI) and a LTQ Orbitrap as detector at
Institut Parisien de Chimie Moléculaire. Melting points were
measured in capillary tubes on a Stuart Scientific SMP3
apparatus and are uncorrected.
From aziridine 2d (51.1 mg, 0.21 mmol) and Meldrum’s acid
(36.2 mg, 0.25 mmol). The mixture was stirred at room
temperature for 3 h. The reaction crude product was purified by
flash chromatography on
silica gel eluting
with
dichloromethane/acetone (80/20; 60/40) to afford 41.4 mg of 4dc
as an orange oil (52 %). 1H NMR (300 MHz, CDCl3) δ 10.20 (s,
1H), 9.63 (s, 1H), 6.06 (dd, J = 5.6, 2.7 Hz, 1H), 5.74 (s, 1H),
5.26 (s, 1H), 4.20 – 4.12 (m, 2H), 4.02 (m, 1H), 3.79 (s, 1H),
3.41 (s, 1H), 3.09 (s, 1H), 2.12 – 2.04 (m, 8H), 1.62 (s, 6H).13C
NMR (75 MHz, CDCl3) δ 171.4, 171.1, 167.3, 142.6, 121.6,
102.2, 79.5, 77.6, 74.1, 69.6, 61.3, 47.9, 43.2, 29.8, 26.2, 25.7,
21.3, 21.0. IR (cm-1): 2927, 1727, 1556, 1455, 1404, 1371, 1245,
1166, 1101, 1018, 874, 808, 779, 727, 521. HRMS m/z calcd for
C18H26NO8 [M+H]+: 384.1653; found 384.1650.
4.2. General base-free procedure in Et2O
In a Schlenk equipped with a stir bar and purged under an argon
atmosphere, was prepared a solution of palladium acetate (5
mol%) and triphenylphosphine (15 mol%) in distilled diethyl
ether (0.5 mL). The solution was let stirring for 10 min at room
temperature. Then, a solution of aziridine (50 mg, 1.0 equiv) in
distilled diethyl ether (0.5 mL) was added and let stirring for
another 10 min. The nucleophile (1.2 equiv) was added and the
mixture was stirred at room temperature (25 °C). The reaction
mixture was dissolved in methanol, and the solvent was removed
(1R*,2R*)-2-(butylamino)-5-(2,2,5-trimethyl-4,6-dioxo-1,3-
dioxan-5-yl)cyclopent-3-en-1-yl acetate 4cd.
From palladium acetate (3.0 mg, 10 mol%) and
triphenylphosphine (10.1 mg, 30 mol%), aziridine 2c (25.5 mg,
1.0 equiv) and methyl meldrum acid 3d (24.4 mg, 1.2 equiv). The
mixture was stirred at room temperature for 5 h. The reaction
mixture was dissolved in methanol, and the solvent was removed
under reduced pressure. The reaction crude was purified by flash
column chromatography dichloromethane/methanol (99/1; 98/2)
to afford the product 4cd as a yellow oil (11.8 mg, 26%). This
product was isolated with trace amounts of triphenylphosphine
1
under reduced pressure. The reaction crude was analyzed by H
NMR and the crude product was purified by silica gel
chromatography.
5-((1R*,4R*,5R*)-5-acetoxy-4-(butylammonio)cyclopent-2-en-
1-yl)-2,2-dimethyl-4-oxo-4H-1,3-dioxin-6-olate 4cc.
From aziridine 2c (50 mg, 0.26 mmol, 1 equiv) and Meldrum’s
acid (44 mg, 0.31 mmol, 1.2 equiv). The mixture was stirred at
room temperature for 5 h. The crude product was purified by
1
oxide (around 20%), NMR yield of 65%. H NMR (400 MHz,
CDCl3) δ 5.91 – 5.89 (m, 1H), 5.67 (d, J = 6.1 Hz, 1H), 5.53 (dd,
J = 4.0, 3.1 Hz, 1H), 3.71 (dd, J = 4.5, 2.5 Hz, 1H), 3.37 – 3.36
(m, 1H), 2.70 – 2.57 (m, 2H), 2.07 (s, 3H), 1.76 (s, 3H), 1.72 (s,
3H), 1.68 (s, 3H), 1.45 – 1.32 (m, 4H), 0.91 (t, J = 7.2 Hz, 3H).
13C NMR (101 MHz, CDCl3) δ 170.3, 169.5, 135.3, 127.8, 105.4,
78.2, 71.1, 60.9, 51.5, 46.8, 32.5, 30.3, 28.2, 21.3, 21.2, 20.5,
14.1. IR (cm-1): 2973, 1740, 1618, 1233, 1046, 880, 723, 542.
HRMS m/z calcd for C18H28NO6 [M+H]+: 354.1911; found
354.1912.
flash
chromatography on silica gel eluting
with
dichloromethane/acetone (60/40) to afford 61 mg of 4cc (71%
yield) as a brown solid. m.p. 137.0-137.2 °C.1H NMR (400
MHz, CDCl3) δ 9.87 (s, 1H), 9.55 (s, 1H), 5.98 (dd, J = 5.7, 2.7
Hz, 1H), 5.73 (dd, J = 5.5, 2.6 Hz, 1H), 5.28 (s, 1H), 3.97 (t, J =
2.5 Hz, 1H), 3.75 (s, 1H), 3.28 (d, J = 9.7 Hz, 1H), 2.94 (s, 1H),
2.05 (s, 3H), 1.69 (p, J = 7.6 Hz, 2H), 1.59 (s, 6H), 1.42 – 1.36
(m, 2H), 0.92 (t, J = 7.3 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ
170.7, 167.3, 141.8, 122.1, 102.0, 79.4, 74.0, 69.8, 48.3, 46.0,
28.8, 25.7, 21.3, 20.0, 13.7. IR (cm-1): 3427, 2962, 2935, 2870,
2459, 2250, 2184, 2139, 2057, 2031, 2013, 1970, 1933, 1737,
1653, 1548, 1460, 1399, 1369, 1235, 1202, 1165, 1133, 1103,
1049, 1023, 993, 911, 805, 777, 744, 644, 604, 578, 520, 422,
396. HRMS (ESI) calcd for C17H26NO6 [M+H]+: 340.1755;
found: 340.1756.
5-((1R*,4R*,5R*)-5-acetoxy-4-(butylammonio)cyclopent-2-en-
1-yl)-1,3-dimethyl-2,6-dioxo-1,2,3,6-tetrahydropyrimidin-4-
olate 4ce.
From aziridine 2c (50 mg, 0.26 mmol, 1 equiv) and 1,3-
dimethylbarbituric acid 3e (48 mg, 0.31 mmol, 1.2 equiv). The
mixture was stirred at room temperature for 18 h. The crude
product was purified by flash chromatography on silica gel
eluting with dichloromethane/acetone (40/60; 30/70; 20/80) to
afford 63 mg of 4ce (83% yield) as a brown solid. m.p. 240-
5-((1R*,4R*,5R*)-4-(butylammonio)-5-hydroxycyclopent-2-
en-1-yl)-2,2-dimethyl-4-oxo-4H-1,3-dioxin-6-olate 4bc.
From aziridine 2b (50 mg, 0.32 mmol, 1 equiv) and Meldrum’s
acid (56 mg, 0.39 mmol, 1.2 equiv). The mixture was stirred at
room temperature for 22 h. The crude product was purified by
1
242 °C. H NMR (300 MHz CDCl3) δ 10.58(s, 1H), 9.48 (s,
1H), 5.99 (dd, J = 5.7, 2.7 Hz, 1H), 5.63 (d, J = 5.7 Hz, 1H), 5.19
(s, 1H), 4.31(s, 1H), 3.77 (s, 1H), 3.67 (s, 1H), 3.21 (s, 6H), 3.03
(s, 1H), 2.02 (s, 3H), 1.83 – 1.57 (m, 2H), 1.46 (ddd, J = 8.5, 7.3,
5.8 Hz, 2H), 0.95 (t, J = 7.3 Hz, 3H). 13C NMR (75 MHz, CDCl3)
δ 171.1, 164.9, 153.0, 142.7, 121.5, 85.6, 80.1, 69.4, 48.0, 45.7,
39.5, 28.9, 28.6, 27.9, 21.2, 19.9, 13.7. IR (cm-1): 2924, 2854,
1735, 1675, 1573, 1436, 1374, 1314, 1240, 1165, 1067, 1023,
776, 518, 483. HRMS (ESI) calcd for C17H26N3O5 [M+H]+:
352.1867; found: 352.1867.
flash
chromatography on silica gel eluting
with
dichloromethane/acetone (20/80) to afford 42 mg of 4bc (44%
yield) as a brown solid. m.p. 155.2-155.4 °C.1H NMR (400
MHz, CDCl3) δ 9.71 (s, 1H), 9.15 (s, 1H), 6.03 (dd, J = 5.8, 2.7
Hz, 1H), 5.70 – 5.67 (m, 1H), 4.20 (s, 1H), 3.82 (s, 1H), 3.77 (t, J
= 2.5 Hz, 1H), 2.94 – 2.91 (m, 2H), 1.72 – 1.63 (m, 2H), 1.59 (s,
6H), 1.39 (h, J = 7.2 Hz, 2H), 0.92 (t, J = 7.3 Hz, 3H). 13C NMR

7
5-((1R*,4R*,5R*)-4-(butylammonio)-5-hydroxycyclopent-2-
en-1-yl)-1,3-dimethyl-2,6-dioxo-1,2,3,6-tetrahydropyrimidin-
4-olate 4be.
3287, 2918, 1640, 1544, 1314, 1255, 1204, 1143, 890, 676.
HRMS (ESI) calcd for C13H23N2O5 [M+H]+: 287.1601; found:
287.1603.
4.3. General procedure with NaH in 1,4-dioxane
From aziridine 2c (50 mg, 0.32 mmol, 1 equiv) and 1,3-
dimethylbarbituric acid 3e (61 mg, 0.39 mmol, 1.2 equiv). The
mixture was stirred at room temperature for 18 h. The crude
product was purified by flash chromatography on silica gel
eluting with dichloromethane/acetone (90/10); Acetone and
acetone/methanol (95/5); to afford 38 mg of 4be (42% yield) as a
In a Schlenk tube under an argon was prepared a solution of
Pd(OAc)2 (5 mol%,) and PPh3 (15 mol%) in 0.5 mL of distilled
1,4-dioxane. After 10 min of stirring, the aziridine (50 mg, 1
equiv) dissolved in 0.5 mL of distilled 1,4-dioxane was added,
and the mixture was stirred for another 10 minutes. A solution of
the nucleophile (1.2 equiv) deprotonated by sodium hydride (1.2
equiv) in 0.5 ml of 1,4-dioxane was added to the previous
mixture. The mixture was stirred at room temperature (25 °C)
overnight (18 h), and then dissolved with methanol. The solvent
was evaporated, and the crude product was purified by column
chromatographic to afford the desired product.
1
brown solid. m.p. 211-215 °C. H NMR (400 MHz, CDCl3) δ
10.43 (s, 1H), 9.23 (s, 1H), 5.95(dd, J = 5.7, 2.7 Hz, 1H), 5.68 (d,
J = 5.7 Hz, 1H), 4.15 (s, 1H), 3.99 (t, J = 2.4 Hz, 1H), 3.85 (d, J
= 2.1 Hz, 1H), 3.18 (s, 6H), 3.01 – 2.94 (m, 2H), 1.71 (q, J = 7.6
Hz, 2H), 1.43 (p, J = 7.2 Hz, 2H), 0.92 (t, J = 7.3 Hz, 3H).13C
NMR (75 MHz, CDCl3) δ 163.7, 152.7, 143.0, 121.1, 86.5, 79.5,
70.5, 51.5, 44.7, 28.9, 27.9, 19.9, 13.7. IR (cm-1): 3350, 2958,
2925, 2854, 1661, 1565, 1430, 1373, 1311, 1262, 1241, 1065,
1014, 963, 888, 796, 774, 732, 701, 582, 517, 485, 419. HRMS
(ESI) calcd for C15H24N3O4 [M+H]+: 310.1761; found: 310.1760.
Methyl 2-((1R*,4R*,5R*)-5-acetoxy-4-(butylamino)cyclopent-
2-en-1-yl)-3-oxobutanoate 4cb.
From aziridine 2c (50.5 mg, 0.26 mmol, 1 equiv) and methyl
acetoacetate (35.7 mg, 0.39 mmol, 1.2 equiv). The purification of
the crude product (NMR 74% using 1,4-nitroacetate as internal
standard) on silica gel eluting with cyclohexene/ethylacetate
(60/40; 30/70) did not allow to isolate pure the product 4cb
Ethyl 2-((1R*,4R*,5R*)-5-acetoxy-4-(butylamino)cyclopent-2-
en-1-yl)-2-nitroacetate 4cf.
From aziridine 2c (50 mg, 0.26 mmol, 1 equiv) and ethyl
nitroacetate 3f (40.9 mg, 0.31 mmol, 1.2 equiv). The mixture was
stirred at room temperature for 5 h. The crude product was
purified by flash chromatography on silica gel eluting with
cyclohexane/ethyl acetate (70/30) followed by crystallization
with cold diethyl ether and filtration to afford 20 mg of 4cf (24%
yield) as a brown solid. This product was difficult to isolate pure
due to the presence of triphenylphosphine oxide (71% NMR
without
triphenylphosphine
oxide
contamination.
1/1
1
diastereoisomeric mixture H NMR (400 MHz, CDCl3) δ 5.81
(dq, J = 4.9, 2.3 Hz, 1H, dia A+B), 5.71 (ddd, J = 5.5, 3.4, 1.7
Hz, 1H, dia A+B), 4.97 (t, J = 4 Hz, 1H, dia A), 4.89 (t, J = 4 Hz,
1H, dia B), 3.88 (s, 1H, dia A), 3.86 (s, 1H, dia B), 3.74 (s, 3H,
dia A), 3.72 (s, 3H, dia B), 3.30 – 3.27 (m, 1H, dia A+B), 2.66 –
2.60 (m, 1H, dia A+B), 2.25 (s, 3H, dia A), 2.24 (s, 3H, dia B),
2.06 (s, 3H, dia A), 2.05 (s, 3H, dia B), 1.49 – 1.24 (m, 4H, dia
A+B), 0.91 (t, J = 7.2 Hz, 3H, dia A+B). 13C NMR (75 MHz,
CDCl3) δ 202.0, 201.7, 171.2, 170.9, 169.0, 133.4, 133.0, 132.9,
132.0, 132.0, 131.5, 131.5, 82.3, 81.6, 70.2, 69.9, 62.5, 52.7,
52.5, 50.7, 50.3, 47.3, 47.2, 32.6, 32.6, 30.4, 29.9, 21.2, 20.5,
20.5, 14.1. IR (cm-1): 2954, 2925, 1742, 1676, 1536, 1424, 1377,
1323, 1241, 1197, 1166, 1140, 1107, 1037, 785, 767, 750.
HRMS (ESI) calcd for C16H26NO5 [M+H]+: 312.1805; found:
312.1802.
1
yield). m.p. 46-50 °C. 1/1 diastereoisomeric mixture H NMR
(400 MHz, CDCl3) δ 5.99 – 5.92 (m, 1H, dia A+B), 5.80 (d, J =
6.0 Hz, 1H, dia A), 5.76 (d, J = 6.1 Hz, 1H, dia B), 5.55 (d, J =
6.4 Hz, 1H, dia A+B), 5.12 – 5.07 (m, 1H, dia A), 5.05 – 5.01
(m, 1H, dia B), 4.35 – 4.24 (m, 2H, dia A+B), 3.79 (s, 1H, dia
A+B), 3.48 (s, 1H, dia A), 3.43 (s, 1H, dia B), 2.69 – 2.57 (m,
2H, dia A+B), 2.13 – 2.95 (m, 3H, dia A+B), 1.50 – 1.40 (m, 2H,
dia A+B), 1.39 – 1.27 (m, 5H, dia A+B), 0.91 (t, J = 7.2 Hz, 3H,
dia A+B). 13C NMR (101 MHz, CDCl3) δ 171.3, 171.1, 135.6,
135.3, 128.7, 128.2 89.8, 88.9, 80.6, 80.6, 69.9, 69.8, 63.3, 63.3,
52.9, 52.8, 47.4, 47.3, 32.6, 32.5, 21.1, 21.1, 20.5, 14.1, 13.9. IR
(cm-1): 3287, 2920, 2852, 1640, 1547, 1313, 1255, 1204, 697.
HRMS (ESI) calcd for C15H25N2O6 [M+H]+: 329.1707; obtained:
329.1706.
Methyl 2-((1R*,4R*,5R*)-4-(butylamino)-5-
hydroxycyclopent-2-en-1-yl)-3-oxobutanoate 4bb.
From aziridine 2b (50 mg, 0.32 mmol, 1 equiv) and acetoacetate
(45 mg, 0.39 mmol, 1.2 equiv). The crude product was purified
by flash chromatography on silica gel eluting with
dichloromethane/acetone (40/60) to afford 57 mg of 4bb (65%
Ethyl 2-((1R*,4R*,5R*)-4-(butylamino)-5-hydroxycyclopent-
2-en-1-yl)-2-nitroacetate 4bf.
1
yield) as a brown oil. 1/1 diastereoisomeric mixture H NMR
From aziridine 2b (50 mg, 0.32 mmol, 1 equiv) and ethyl
nitroacetate 3f (52 mg, 0.39 mmol, 1.2 equiv). The mixture was
stirred at room temperature for 1 h. The crude product was
purified by flash chromatography on silica gel eluting with
cyclohexane/ethyl acetate (1:1) and cyclohexane/ethyl
acetate/methanol (49.8/49.8/0.2; 49.5/49.5/1) to afford 30 mg of
4bf (32% yield) as a brown solid. m.p. 40-44 °C. 1/1
(400 MHz, CDCl3) δ 5.80 (dt, J = 6.2, 2.1 Hz, 1H, dia A+B), 5.66
(dt, J = 6.1, 1.8 Hz, 1H, dia A), 5.58 (dt, J = 6.1, 1.9 Hz, 1H, dia
B), 4.05 (t, J = 5.1 Hz, 1H, dia A), 3.95 (t, J = 5.2 Hz, 1H, dia B),
3.82 – 3.75 (m, 1H, dia A+B), 3.75 (s, 3H, dia A), 3.74 (s, 3H,
dia B), 3.68 (d, J = 8.9 Hz, 1H, dia A), 3.65 (d, J = 9.1 Hz, 1H,
dia B), 3.23 – 3.14 (m, 1H, dia A+B), 2.78 – 2.72 (m, 2H, dia
A+B), 2.27 (s, 3H, dia A+B), 1.62 – 1.47 (m, 2H, dia A+B), 1.43
– 1.27 (m, 2H, dia A+B), 0.91 (t, J = 7.4 Hz, 3H, dia A+B). 13C
NMR (101 MHz, CDCl3) δ 203.6, 202.7, 170.0, 169.3, 132.4,
131.6, 131.4, 130.7, 81.5 (2C), 71.4, 63.1, 62.4, 52.9, 52.8, 52.1,
51.8, 47.0, 31.7, 31.4, 30.3, 29.7, 20.5, 14.0, 14.0. IR (cm-1):
3348, 2956, 2930, 2872, 2126, 2006, 1978, 1713, 1647, 1577,
1529, 1460, 1424, 1375, 1327, 1253, 1212, 1185, 1163, 1106,
1055, 1029, 978, 939, 898, 860, 789, 765, 751, 683, 613, 533,
473, 392. HRMS (ESI) calcd for C14H24NO4 [M+H]+: 270.1700;
found: 270.1700.
1
diastereoisomeric mixture H NMR (300 MHz, CDCl3) δ 5.68
(d, J = 3.1 Hz, 1H, dia A), 5.58 (d, J = 3.0 Hz, 1H, dia B), 5.05
(d, J = 7.7 Hz, 1H, dia A), 4.96 (d, J = 8.7 Hz, 1H, dia B), 4.30
(p, J = 7.1 Hz, 2H, dia A+B), 4.02 (s, 1H, dia A+B), 3.80 – 3.73
(m, 1H, dia A+B), 3.00 – 2.93 (m, 2H, dia A+B), 2.78 – 2.66 (m,
1H, dia A+B), 2.47 – 2.34 (m, 1H, dia A+B), 1.52 (p, J = 7.1 Hz,
2H, dia A+B), 1.41 – 1.32 (m, 2H, dia A+B), 1.30 (td, J = 7.1,
2.5 Hz, 3H, dia A+B), 0.92 (t, J = 7.3 Hz, 3H, dia A+B). 13C
NMR (101 MHz, CDCl3) δ 200.8, 200.7, 163.3, 163.2, 147.7,
147.7, 114.1, 113.5, 91.4, 90.8, 77.3, 77.0, 76.7, 63.2, 63.2, 43.8,
37.6, 37.2, 37.1, 36.7, 30.9, 20.1, 13.9, 13.9, 13.7. IR (cm-1):

8
Tetrahedron
[31]
M. A. G. Fortunato, C.-P. Ly, F. Siopa, C. A. M. Afonso, Methods
Protoc. 2 (2019) 67.
Acknowledgments
[32]
C. Colombo, B. M. Pinto, A. Bernardi, A. J. Bennet, Org. Biomol.
Chem.14 (2016) 6539–6553.
The authors would like to acknowledge Horizon 2020
ERANet-LAC project CelluloseSynThech for financial support
(ref. ELAC2014/BEE-0341), as well as CNRS, Sorbonne
Université and Labex Michem (Investissements d'Avenir
programme, ref. ANR-11-IDEX-0004-02), Fundação para a
Ciência e Tecnologia (SFRH/BPD/88666/2012, PTDC/QUI-
QOR/32008/2017, UID/DTP/04138/2019) and COMPETE
Programme (SAICTPAC/0019/2015). Support through CMST
COST Action, CA15106 (CHAOS) is also gratefully
acknowledged.
[33]
[34]
[35]
[36]
For a recent example, see: Y. Liu, J. Oble, G. Poli, Beilstein J. Org.
Chem. 15 (2019) 1107–1115;
For a recent example, see: Y. Liu, Z. Mao, A. Pradal, P-Q. Huang, J.
Oble, G. Poli, Org. Lett. 20 (2018) 4057–4061.
For an account, see: C. Kammerer, G. Prestat, D. Madec, G. Poli,
Acc. Chem. Res. 47 (2014) 3439–3447.
For a chapter review, see: C. Kammerer-Pentier, F. Liron, G. Prestat,
G. Poli, Selectivity in Palladium Catalyzed Allylic Substitution. In
Transition Metal Catalyzed Enantioselective Allylic Substitution in
Organic Synthesis; Kazmaier, U., Ed.; Topics in Organometallic
Chemistry; Springer-Verlag: Berlin Heidelberg, 2012; Vol. 38, pp
1−64.
Supplementary Material
[37]
[38]
See Supporting Information for detailed optimization.
We were not able to totally remove traces of triphenylphosphine
oxide from the isolated compound 4cb.
Crystallographic data was deposited at the Cambridge
Crystallographic Data Centre with number CCDC 1980421.
The ratio of neutral to zwitterionic form is expected to be rather
context dependent and may be different in going from solid state to
solution, or in solvents with differing dielectric constants.
T. Ohwada, H. Hirao, A. Ogawa, J. Org. Chem. 69 (2004) 7486-
7494.
Supplementary data to this article can be found online at
https://doi.org/.
[39]
[40]
References and notes
[1]
[2]
G. S. Singh, Heterocyclic Chem. 129 (2019) 245–335.
M. C. McMills, S. C. Bergmeier, 1.02 - Aziridines and azirines:
fused-ring derivatives. In Comprehensive Heterocyclic Chemistry
III, A. R. Katritzky, C. A. Ramsden, E. F. V. Scriven, R. J. K.
Taylor, Eds. Elsevier: Oxford, 2008; pp 105–172.
S. Vandekerckhove, M. D’hooghe, Bioorg. Med. Chem. 21 (2013)
3643–3647.
[41]
[42]
[43]
R. Hadjadj-Aoul, A. Bouyacoub, A. Krallafa, F. Volatron, J. Mol.
Struct. THEOCHEM 849 (2008) 8-16.
I. Radić, N. Despotović, R. Vianello, Croat. Chem. Acta. 85 (2012)
495-504.
[3]
[4]
L. Kaplan, J. W. Pavlik, K. E. Wilzbach, J. Am. Chem. Soc. 94
(1972) 3283–3284.
[5]
U. C. Yoon, S. L. Quillen, P. S. Mariano, Tetrahedron Lett. 23
(1982) 919–922.
[6]
U. C. Yoon, S. L. Quillen, P. S. Mariano, R. Swanson, J. L.
Stavinoha, E. Bay, J. Am. Chem. Soc. 105 (1983) 1204–1218.
R. Ling, M. Yoshida, P. S. Mariano, J. Org. Chem. 61 (1996) 4439-
4449.
[7]
[8]
S. J. Cho, R. Ling, A. Kim, P. S. Mariano, J. Org. Chem. 65 (2000)
1574–1577.
[9]
F. Glarner, S. R. Thornton, D. Schärer, G. Bernardinelli, U. Burger,
Helv. Chim. Acta 80 (1997) 121–127.
[10]
[11]
[12]
[13]
[14]
E. A. Acar, F. Glarner, U. Burger, Helv. Chim. Acta 81 (1998) 81
1095–1104.
F. Glarner, B. Acar, I. Etter, T. Damiano, E. A. Acar, G.
Bernardinelli, U. Burger, Tetrahedron 56 (2000) 4311–4316.
R. A. King, H. P. Lüthi, H. F. Schaefer III, F. Glarner, U. Burger,
Chem. Eur. J. 7 (2001) 1734–1742.
M. A. Clark, R. C. Schoenfeld, B. Ganem, J. Am. Chem. Soc. 123
(2001) 10425–10426.
C. S. Penkett, I. D. Simpson, Tetrahedron Lett. 42 (2001) 1179–
1181.
[15]
[16]
[17]
[18]
[19]
C. S. Penkett, Synlett 1 (1999) 93–95.
C. S. Penkett, I. D. Simpson, Tetrahedron 55 (1999) 6183–6204.
R. Ling, P. S. Mariano, J. Org. Chem. 63 (1998) 6072–6076.
Z. Zhao, L. Song, P. S. Mariano, Tetrahedron 61 (2005) 8888–8894.
L. Song, E. N. Duesler, P. S. Mariano, J. Org. Chem. 69 (2004)
7284–7293.
[20]
[21]
[22]
[23]
[24]
H. Lu, P. S. Mariano, Y. Lam, Tetrahedron Lett. 42 (2001) 4755–
4757.
X. H. Feng, E. N. Duesler, P. S. Mariano, J. Org. Chem. 70 (2005)
5618–5623.
H. Lu, Z. Su, L. Song, P. S. Mariano, J. Org. Chem. 67 (2002) 3525–
3528.
For reviews, see: G. Poli, G. Prestat, Chemtracts: Org. Chem. 15
(2002) 506–513, and the next five references.
T. Damiano, D. Morton, A. Nelson, Org. Biomol. Chem. 5 (2007)
2735–2752.
[25]
[26]
J. Zou, P. S. Mariano, Photochem. Photobiol. Sci. 7 (2008) 393–404.
S. Sowmiah, J. M. S. S. Esperança, L. P. N. Rebelo, C. A. M.
Afonso, Org. Chem. Front. 5 (2018) 453–493.
P. S. Mariano, Tetrahedron 39 (1983) 3845–3879.
P. S. Mariano, Acc. Chem. Res. 16 (1983) 130–137.
J. R. Vale, F. Siopa, P. S. Branco, C. A. M. Afonso, Eur. J. Org.
Chem. (2016) 2048–2053.
[27]
[28]
[29]
[30]
F. Siopa, J. P. M. António, C. A. M. Afonso, Org. Process Res. Dev.
22 (2018) 551–556.

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:
Carlos A. M. Afonso