Cycloaddition Reactions of Azides and Electron-Deficient Alkenes in Deep Eutectic Solvents: Pyrazolines, Aziridines and Other Surprises

Article abstract of DOI:10.1002/adsc.201901614

The reaction of organic azides and electron-deficient alkenes was investigated in a deep eutectic solvents. A series of highly substituted 2-pyrazolines was successfully isolated and their formation rationalised by DFT calculations. The critical effect of substitution was also explored; even relatively small changes in the cycloaddition partners led to completely different reaction outcomes and triazolines, triazoles or enaminones can be formed as major products depending on the alkene employed. (Figure presented.).

Full text of DOI:10.1002/adsc.201901614

Title: Cycloaddition Reactions of Azides and Electron-Deficient Alkenes
in Deep Eutectic Solvents: Pyrazolines, Aziridines and Other
Surprises
Authors: Filip Sebest, Kushal Lachhani, Chaleena Pimpasri, Luis
Casarrubios, Andrew White, Henry Rzepa, and Silvia Díez-
González
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201901614
Link to VoR: http://dx.doi.org/10.1002/adsc.201901614

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Cycloaddition Reactions of Azides and Electron-Deficient Alkenes in
Deep Eutectic Solvents: Pyrazolines, Aziridines and Other Surprises
Filip Sebest,a Kushal Lachhani,a Chaleena Pimpasri,a Luis Casarrubios,a,b Andrew J. P.
White,a Henry S. Rzepa,a and Silvia Díez-González a*
a
Imperial College London, Department of Chemistry, MSRH, White City Campus, 80 Wood Lane, London W12 0BZ
(UK)
E-mail: s.diez-gonzalez@imperial.ac.uk
Departamento de Química Orgánica I, Facultad de Ciencias Químicas, Universidad Complutense and Centro de
b
Innovación en Química Avanzada (ORFEO – CINQA), 28040 Madrid (Spain)
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.
Abstract. The reaction of organic azides and electron-deficient alkenes was
investigated in a deep eutectic solvent. A series of highly substituted 2-pyrazolines
was successfully isolated and their formation rationalised by DFT calculations. The
critical effect of substitution was also explored; even relatively small changes in the
cycloaddition partners led to completely different reaction outcomes and triazolines,
triazoles or enaminones can be formed as major products depending on the alkene
employed.
Keywords: Alkenes; Azides;
Cycloaddition; Heterocycles;
Solvent Effects;
isolated.[6] In most cases, aziridines, imines or 2-
pyrazolines are obtained instead (Scheme 1).
Electron-withdrawing substituents in triazoline 3
relatively stabilise its open-ring isomer, often
postulated as a zwitterion in the literature. This might
in turn evolve into aziridine 4 upon N2 extrusion, or
alternatively undergo a second cycloaddition with the
Introduction
Heterocycles are essential molecules both in
chemistry and biology, and accordingly there is a
relentless drive for efficient, selective and more
sustainable methodologies for accessing these highly
valuable products. 1,3-Dipolar cycloadditions
represent a convergent and convenient route to five-
membered heterocyclic molecules.[1] Compared to the
ubiquitous azide–alkyne cycloaddition reaction to
electron-poor
alkene
to
generate,
after
tautomerisation, 2-pyrazoline 6. The preferential
pathway for 1,4-disubstituted triazolines is strongly
dependent on the substitution on either cycloaddition
partner. Long reaction times (up to 150 days!)[6d] are
required for the formation of pyrazolines as these are
typically carried out at room temperature to avoid
undesired decomposition. Alternatively, this reaction
produce
triazoles,[2]
analogous
azide–alkene
cycloaddition to form triazolines has barely received
any attention and no metal-catalysed version has been
reported to date. Unactivated alkenes do not react or
only sluggishly with organic azides and
excruciatingly long reaction times (i.e. months) are
often required.[3] Furthermore, triazolines are
typically more reactive than either of the starting
materials and while this can be regarded as an
opportunity to access a range of heterocycles from
simple starting materials,[4] only a handful of these
examples are synthetically useful due to selectivity
and decomposition issues.
can be promoted by the use of an external base, either
[6a,d]
NEt3
or DABCO,[7] although these reactions
remain largely limited to acrylates.
It is well established that electron-deficient alkenes
deliver 1,4-disubstituted triazolines regioselectively
when reacted with azides.[5] However, these
compounds are particularly unstable and to date only
a handful of 1,4-disubstituted triazolines have been
Scheme 1. Evolution of disubstituted 1,2,3-triazolines.
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
We were particularly interested in the preparation
observed in all tested solvents except for DES. No or
at most traces of aziridine 4aa were produced in these
reactions, except for the one carried out on water
(Table 1, entry 6).
of pyrazolines, since they display a wide range of
biological activities as well as very promising
physical properties.[8] A reliable access to pyrazolines
from easily accessible azides and alkenes would
provide a convenient alternative to established routes
to these heterocycles, namely the cyclisation under
acidic conditions of hydrazones generated from α,β-
Table 1. Solvent screening.[a]
unsaturated aldehydes or ketones and
10]
hydrazines,[9,
cycloaddition
or the 1,3-dipolar
of electron-rich
nitrilimines and electron-deficient
alkenes.[11,
We recently reported
12]
the cycloaddition of azides and
relatively electron-rich alkenes in a
deep
eutectic
solvent
(DES)
Recov
(%)[b]
Conv (%)[b]
composed of choline chloride and
urea.[13] This strategy allowed the
isolation of Δ2-triazolines on a gram
scale, whilst minimising the use of
volatile organic solvents. DESs are
non-flammable mixtures with a much
lower melting point than any of their
t
Overall
(%)
Entry Solvent
(h)
1a
28
14
55
28
73
44
75
49
69
41
37
15
33
25
18
10
59
50
91
92
3aa 4aa 5aa 6aa
1
ChCl/urea
8
24
8
24
8
24
8
24
8
24
8
24
8
--
--
--
--
--
--
--
--
13
19
--
--
--
--
--
--
--
--
--
--
--
--
<5
<5
--
--
--
--
--
--
6
<5
6
59
69
34
52
11
28
6
24
<5
20
41
54
66
67
68
71
44
46
--
72
84
45
68
23
49
20
44
31
43
63
85
66
69
68
71
44
46
9[c]
8[c]
(1:2)
2
i-PrOH
10
14
12
21
14
20
12
23
18
22
--
<5
--
--
--
--
--
--
3
EtOH
THF
individual
components.
These
eutectic mixtures are held together
through hydrogen bonds and are
liquid at the reaction temperature.
The reported applications of deep
eutectic solvents keep on increasing
powered by their competitive cost
4
5
Toluene
Water
6
and
enhanced
environmental
and include
9
friendliness,[14]
7
ChCl/water
(1:2)
ChCl/DMU
(1:2)
Gly/Urea
(10:4)
4% NH3 in
EtOH
--
--
--
--
--
--
--
--
thermal,[15] Lewis acid-[16] and metal-
catalysed[17] cycloaddition reactions.
In this article, we present the
straightforward preparation of highly
substituted 2-pyrazolines from azides
and alkenes via transient 1,4-
disubstituted triazolines in DES. The
critical effect of substituents of
alkenes is explored and specifically,
we show how relatively small
changes can lead to 1,5-disubstituted
triazolines, enaminones or, triazoles
as major reaction products.
24
8
24
8
24
8[c]
9
10[d]
8
24
--
[a]
Reaction conditions: Azide 1a (1.0 mmol), alkene 2a (2.2 mmol) in 2 mL
of solvent. [b] 1H NMR recoveries/yields are the average of two independent
experiments and were determined with respect to 1,3,5-trimethoxybenzene as
internal standard. Decomposition products were also observed. Ethyl 4-
amino benzoate was the only reaction product.
[c]
[d]
It has been reported in the literature that urea
containing DESs might decompose when heated
originating ammonia as by-product.[19] To confirm
Results and Discussion
Preparation of 2-pyrazolines: Strong solvent effects
were observed when a first screening was performed
with ethyl 4-azidobenzoate 1a and methyl vinyl
ketone 2a as the model substrates (Table 1).[18] In a
first screening, the highest conversion to pyrazoline
6aa was obtained in a choline chloride/urea DES
after 24 h, presumably due to the presence of basic
urea, although acceptable conversions were also
obtained in protic solvents such as isopropanol or
pure water in the absence of a base (Table 1, entries 1,
2 and 6). Triazoline 3aa was only detected in toluene
(Table 1, entry 5), while significant amounts (up to
whether ammonia played a role in our reactions, we
next tested several DESs known to produce different
amounts of ammonia. DESs with no urea
(ChCl/water, Table 1, entry 7) or reported to form
similar amounts of ammonia than ChCl/urea
(ChCl/DMU, Table 1, entry 8) led to comparable
results. However, a mixture of glycerol and urea,
expected to produce 10 times more ammonia,
actually inhibited the formation of pyrazoline 6aa
(Table 1, entry 9). To confirm this trend a
commercially available solution 4% of ammonia in
ethanol was also screened and no cycloadduct, only
an aniline by-product was then formed (Table 1, entry
1
23% by H NMR) of diazo compound 5aa were
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
Table 2. Synthesis of pyrazolines with methyl vinyl
10). These results clearly show that the improved
yields in pyrazoline in DES are not due to the
incidental formation of ammonia under the reaction
conditions.
ketone.[a]
Since incomplete conversions were observed in all
solvents after 24 h, higher reaction temperatures were
also tested with ChCl/urea as DES. Identical results
were obtained at 50 ºC, but slightly better conversion
into pyrazoline 6aa was obtained at 60 ºC after 16 h
1
(78% by H NMR). Longer reaction times or higher
temperatures resulted in no further conversions and
formation of decomposition products. Hence, the
scope of the reaction was explored at 60 ºC in DES
with a slight excess of the alkene partner.
Pyrazolines 6 were isolated in good yields from
the reaction of various aryl and heteroaryl azides with
methyl vinyl ketone (Table 2, entries 1–11). The
exceptions were reactions with 1-azido-2-
trifluoromethylbenzene 1g and 3-azidopyridine 1k,
where even if some of the starting azide could be
recovered, significant decomposition was also
observed (Table 2, entries 7 and 11). Nevertheless,
increased steric hindrance was well tolerated in
general and pyrazolines from 2-chlorophenyl azide or
mesityl azide were successfully prepared (Table 2,
entries 8 and 9). Adamantyl azide also reacted under
these conditions, although a lower yield in pyrazoline
6la was obtained, partially due to incomplete
conversion of the starting azide. All other screened
alkyl azides, such as benzyl azide (Table 2, entry 13),
led to the complete decomposition of the reaction
mixture, even when the reactions were carried out at
lower temperatures.
In these reactions no triazoline or aziridine
products were evidenced and only traces of the
corresponding diazo intermediates were observed by
1H NMR. Interestingly, the reaction of 1-azido-3-
chlorobenzene 1f also delivered 1,4-disubstituted
1
triazole 7fa (5% by H NMR), which was isolated in
4% yield and fully characterised. Formation of
analogous triazoles was not observed in any other
reaction.
A different electron-poor alkene, diethyl vinyl
phosphonate, also produced the expected pyrazoline
6nb in an excellent yield when reacted with 4-
azidobenzonitrile at 70 ºC for 24 h (Scheme 2).
[a]
Reaction conditions: Azide 1X (1.0 mmol), alkene 2a
[b]
(2.2 mmol) in 2 mL DES, choline chloride/urea (1:2).
1H NMR recoveries were determined with respect to 1,3,5-
[c]
trimethoxybenzene as internal standard.
are the average of two independent experiments, H NMR
yields are provided in brackets and were determined with
Isolated yields
1
[d]
respect to 1,3,5-trimethoxybenzene as internal standard.
Triazole 7fa was also isolated from this reaction.
The structure of 6nb was confirmed by X-ray
crystallography (Figure 1). The pyrazoline ring in
6nb has a small envelope deformation with N1 lying
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
Table 3. Cycloaddition of azides and methyl acrylate.[a]
ca. 0.14 Å out of the {N2,C3,C4,C5} plane, which
atoms are coplanar within ca. 0.01 Å.
Scheme 2. Cycloaddition of 4-azidobenzonitrile and
diethyl vinyl phosphonate. DES = choline chloride/urea
(1:2).
[a]
Reaction conditions: Azide 1X (1.0 mmol), alkene 2c
[b]
(2.2 mmol) in 2 mL DES, choline chloride/urea (1:2).
Isolated yields are the average of two independent
1
experiments, H NMR yields are provided in brackets and
were determined with respect to 1,3,5-trimethoxybenzene
as internal standard. [c] Reaction carried out for 16 h.
Computational studies: In our previous report on
the cycloaddition of azides and electron-rich alkenes,
1,5-disubstituted triazolines were the typical major
reaction products, while regioisomeric 1,4-triazolines
were never observed, and instead, were shown to
evolve to either aziridine or imine derivatives.[5]
Indeed, the presence of an EWG in the starting olefin
not only switches the regioselectivity of the reaction,
but also leads to a different reaction pathway with
pyrazolines 6 as major products. A triazoline
opening/cycloaddition sequence is often postulated to
rationalise these reactions (see Scheme 1),[6a,d,7] but to
date no computational mechanistic studies are
available for this transformation.
A plausible reaction profile to rationalise the
formation of pyrazolines 6 from azides and alkenes
was elucidated through DFT calculations.[20] First, the
initial cycloaddition of 4-chlorophenyl azide 1b and
methyl vinyl ketone 2a was computed and a clear
preference for the formation of the 1,4-disubstituted
triazoline 3ba was found, as expected for an electron-
deficient alkene (Scheme 3, TS11,4 vs TS11,5). For
completion, the evolution of the 1,5-disubstituted
regioisomer 3ba’ was explored, and it resulted in the
formation of imine by-products through a nitrogen
extrusion/migration sequence. Considering the higher
Figure 1. Structure of pyrazoline 6nb (50% probability
ellipsoids). Most hydrogens are omitted for clarity.
Selected bond lengths (ꢀ) and angles (°): N(1)–N(2)
1.362(3), N(2)–C(3) 1.290(3), C(3)–C(4) 1.505(3), C(4)–
C(5) 1.554(3), C(5)–N(1) 1.484(3); C(5)–N(1)–N(2)
112.03(16); N(1)–N(2)–C(3) 109.55(18); N(2)–C(3)–C(4)
114.05(19); C(3)–C(4)–C(5) 101.31(17); N(1)–C(5)–C(4)
102.04(16).
Next, methyl acrylate was used as cycloaddition
partner (Table 3). Unsurprisingly, a small loss of
regioselectivity was observed with a less polarised
alkene and traces of the corresponding 1,5-
disubstituted triazolines were detected in some cases
1
by H NMR[6a,b]. In these reactions both pyrazolines
and aziridines were formed, except with 3-
azidopyridine and adamantyl azide, which
exclusively produced the corresponding pyrazolines
6kc and 6lc (Table 3, entries 7 and 8). Full azide
conversions were obtained in these reactions but
decomposition was sometimes problematic, and for
example, only 49% of 3-chlorophenyl azide 1b was
converted into cyclic products (Table 3, entry 5).
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
We previously reported[13] that
such transition states could lead to
the formation of either an aziridine
or an imine derivative. An
alternative retro-cycloaddition to a
diazo compound and an imine (TS23,
Scheme 4) is a non-productive but
non-competitive pathway (G‡
=
333
31.3 kcal/mol). For the present study,
no transition state could be located
for the direct cleavage of the N1–N2
bond in 3ba, so we then attempted
the initiation of the process by a
proton shift from C4 to N1. While a
direct hydrogen migration appeared
to be energetically unviable (TS24,
G‡
=
52.2 kcal/mol, see
333
repository data for details[19]), an
IRC calculation on this transition
state nonetheless confirmed the
desired cleavage of the N1–N2 bond,
a single concerted (albeit non-
yielding in
Scheme 3. Initial cycloaddition step and evolution of 1,5-
disubstituted triazoline 3ba’. G‡ values are provided
synchronous) step to the diazo compound 5ba.
333
It is well known that participation of solvent
molecules in such proton migrations can lower the
activation barriers.[21] Transition states including one
to three water molecules as proton transfer agents, as
well as one or two choline chloride molecules were
located (TS25–TS29)[19] but the resulting activation
free energies ranged between 40.1 and 52.2 kcal/mol,
values significantly higher than the previous 25.8
kcal/mol required for TS21, as well as for the
deprotonation of triazoline 3ba with urea (TS210,
in kcal/mol.
free energy barriers located for these processes,
neither triazoline 3ba’ nor its decomposition products
are expected to be formed in considerable amounts,
as confirmed experimentally.
We then studied the possible evolution paths for
triazoline 3ba. Structures of all calculated transition
states (TS21 to TS212) are available via the data
repository,[19] while the most energetically viable
structures are depicted in Scheme 4. Firstly, N2 loss
may occur through two diastereomeric transition
states (TS21 and TS22) according to an asynchronous
concerted retro-cycloaddition mechanism.
G‡ = 25.3 kcal/mol). Overall, the most viable
333
pathway located from 3ba is TS212, which involves
two molecules of water as well as a proton and a
chlorine anion to form diazo compound 5ba. The free
energy barrier (computed for [H+] = 0.036 M = 1 atm
at 333 K) is sufficiently low (G‡
= 15.6
333
kcal/mol) that even
a
much smaller proton
concentration of e.g. 10-7 M would still lead to a
facile barrier (G‡333 = 24.1 kcal/mol) for the proton
transfer. Furthermore, TS212 would lead to diazo
compound 5ba in a single step with no other
intermediates involved. From there,
a second
cycloaddition process with methyl vinyl ketone
would yield the observed pyrazoline 6ba after a final
facile proton shift (Scheme 5).
Overall, our calculations indicate favourable free
energy barriers for the reactions corresponding to our
experimental results. A first thermal cycloaddition
step leads to 1,4-disubstituted triazoline 3ba as a
single regioisomer, which is kinetically unstable and
readily evolves to the corresponding diazo compound
5ba in the highly polar, protic reaction media
employed in these transformations. A straightforward
second cycloaddition/tautomerisation sequence then
delivers pyrazoline 6ba, the only experimentally
observed product under the reported conditions for
the model substrates.
Scheme 4. Selected located transition states for the
formation of diazo compound 4ba. G‡
values are
333
provided in kcal/mol.
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
8nd was in accordance with the literature.[23] The
structure of 3nd' was found to contain two
crystallographically independent molecules (3nd'-A
and 3nd'-B) in the asymmetric unit. An envelope
deformation of the triazoline ring was observed in
both cases and the bond lengths and angles on the
heterocyclic ring for this 1,5,5-trisubstituted
triazoline were essentially identical to those reported
for 1,5-disubstituted triazolines except for a longer
N(1)–C(5) bond and smaller N(1)–C(5)–C(4) angle,
which might be attributed to the additional
substituent in the 5-position.[24]
Scheme 5. Complete reaction profile with G‡333 for the
formation of pyrazoline 6ba.
Further reactivity with electron-poor alkenes:
With a mechanistic profile charted computationally,
we next explored experimentally the effect of
introducing additional substituents on the alkene
moiety. This had dramatic effects on the outcome of
the reaction, and no pyrazolines were obtained from
substituted enones. Furthermore, either higher
reaction temperatures or longer reaction times were
required for these substrates. The reaction of 4-
azidobenzonitrile with a gem-disubstituted 3-methyl-
3-buten-2-one allowed isolation of four different
products: a triazoline, an aziridine, and two
enaminones (Scheme 6). The significant formation of
the first two products shows that the presence of the
gem-methyl substituent has a significant impact both
on the regioselectivity of the azide-alkene
cycloaddition and on the preferred evolution path for
the transient 1,4,4-trisubstituted triazoline.
3nd’-A
E-8nd
Figure 2. Structure of one (3nd’-A) of the two
independent molecules present in the crystal of triazoline
3nd’, and enaminone E-8nd (50% probability ellipsoids).
Most hydrogens are omitted for clarity.
Mechanistically, aziridine 4nd, and enaminones
8nd are generated by different evolution of the open-
ring isomer of unstable triazoline 3nd after nitrogen
extrusion. Our previously reported calculations
showed that aziridines 4 were solely formed from
1,4-disubstituted triazolines after the electrocyclic
ring closure of a singlet biradical intermediate.
Enaminones 8nd would be formed from 3nd’ or 3nd
through similar intermediate after a 1,2-hydride (or
Scheme 6. Cycloaddition of 4-azidobenzonitrile with 3-
methyl-3-buten-2-one. DES = choline chloride/urea (1:2).
[a]
1H NMR yields/recovery are the average of two
Textmarke nicht
independent experiments and were determined with respect
to 1,3,5-trimethoxybenzene as internal standard.
acyl) shift and tautomerisation.[13Fehler!
definiert.]
On the other hand, the reaction of 2-cyclohexen-
1-one led to a similar outcome with the formation of
two enaminones but no heterocycles were evidenced
in the crude mixture, only diazo intermediate 5ae
(Scheme 7). This reaction was quite sluggish and
22% of the starting azide was recovered even after
heating for 48 h. The formation of related ring-
contracted enaminones from azides and 2-
cyclohexen-1-one has been reported in the presence
of a Lewis acid.[25] However, under such conditions
The spectroscopic data for all of these products was
in accordance with that reported for similar
1
compounds, and notably the H NMR spectra of
isomeric enaminones 8nd displayed chemical shifts
of the N–H protons at 12.69 and 6.51 ppm,
respectively, as expected due to the influence of
intramolecular hydrogen bonding in Z-8nd.[22] The
structures of 3nd’ and E-8nd were further confirmed
by X-ray crystallography (Figure 2), and that of Z-
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
the Z isomer was the major product in most reported
Triazole 10nf most probably originates from the
expected 1,4-disubstituted triazoline followed by
deprotonation at C4 and conjugate addition of a
second molecule of vinyl phenyl sulfone. This step
would be particularly supported by the stabilising
effect of the sulfonyl group onto the corresponding
carbanion.[27] An alkylation step at C4 would also
prevent any isomerisation to linear diazo compounds
since there would no longer be a proton available at
that position. An elimination step from a transient
1,4,4-trisubstituted triazoline (Scheme 5)[28] would
generate triazole 10nf and phenylsulfonate, which
would then form compound 11f upon reaction with
the starting vinyl sulfone.[29]
examples.
Scheme 7. Reaction of ethyl 4-azidobenzoate with 2-
cyclohexen-1-one. DES = choline chloride/urea (1:2). [a] 1
H
NMR yields are the average of two independent
experiments and were determined with respect to 1,3,5-
trimethoxybenzene as internal standard.
Conclusion
A range of trisubstituted 2-pyrazolines has been
prepared from the corresponding azides and electron-
deficient alkenes in DES and fully characterised. The
reaction media not only reduces the amount of
volatile organic solvent required, but also
circumvents the need for an external base to promote
the formation of the desired pyrazolines. Furthermore,
the reaction times can be dramatically reduced in
DES since higher temperatures than in the previously
reported solvents (typically MeOH or toluene) are
now possible. This is most probably due to a
stabilising effect of the solvent on the transient
A last alkene was tested, phenyl vinyl sulfone,
which led to an unexpected reaction outcome. In this
case, no pyrazoline or any of the previously obtained
products were observed and instead, 1,4-disubstituted
triazole 10nf and 1,2-bis(phenylsulfonyl)ethane 11f
were formed (Scheme 8). Disappointingly, only small
amounts of pure products could be obtained since
10nf and 11f could hardly be separated either by
column chromatography, or recrystallisation and
most of triazole 10nf (72%) was isolated as a 57:43
mixture with compound 11f. Fortunately, both
structures could still be fully confirmed by X-ray
crystallography. That of triazole 10nf is depicted in
Figure 3, while data collected for 11f was in
accordance with the literature.[26]
triazolines,
which
minimises
undesired
decomposition and promotes the formation of
cycloadducts.[13] Most remarkably, previous reports in
organic solvents used basic conditions to promote the
formation of pyrazolines, but in contrast, our DFT
calculations points to formal acidic media (a
combination of H+ and Cl–) as key for this reaction in
DES.
The impact of the substitution on the alkene of
choice can only be described as dramatic and while
pyrazolines were formed preferentially (or even as
the only products) from mono-substituted alkenes, the
formation of other heterocycles, namely aziridines,
triazolines or triazoles can become competitive or
even the preferred reaction path. In order to fully
exploit the potential of these transformations a better
understanding of the factors controlling the reaction
selectivity is needed. Efforts in this regard are
currently ongoing in our laboratory and will be
reported in due course.
Scheme 8. Reaction of 4-azidobenzonitrile with phenyl
vinyl sulfone. DES = choline chloride/urea (1:2).
Experimental Section
General Considerations: All chemicals were obtained
from commercial sources and used without further
purification. The Deep Eutectic Solvent (DES) used in this
work were prepared following a reported procedure.[30]
Melting points were determined on an Electrothermal
Gallenhamp apparatus and are uncorrected. Infrared
spectra were recorded using a Perkin Elmer 100 series FT-
Figure 3. Structure of triazole 10nf (50% probability
ellipsoids). Most hydrogens are omitted for clarity.
IR spectrometer, equipped with
a beam-condensing
accessory (samples were sandwiched between diamond
compressor cells). NMR spectra were measured on Bruker
AVANCE 400 spectrometers (1H: 400 MHz, 13C: 101
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
MHz, 19F: 377 MHz, 31P: 162 MHz) at 20 °C. The
chemical shifts (δ) are given in ppm relatively to a
tetramethylsilane (0.00 ppm), CDCl3 (77.2 ppm), DMSO-
d6 (39.5 ppm), 1-fluorobenzene (–113.15 ppm) or
phosphoric acid (0.00 ppm). The multiplicity is given as br,
s, d, t, q, sept, and m for broad, singlet, doublet, triplet,
References
[1] a) 1,3-Dipolar Cycloaddition Chemistry, A. Padwa, ed.,
Wiley-Interscience, New York, 1984, Vol. 1 and 2; b)
R. Huisgen, Proc. Chem. Soc. 1961, 357–369.
1
quartet, septet, and multiplet. Assignments of some H and
13C NMR signals rely on COSY, HSQC, HMBC and/or
DEPT-135 experiments. Single crystal X-ray diffraction
data was collected using Xcalibur PX Ultra A and Agilent
Xcalibur 3 E diffractometers, and the structures were
refined using the SHELXTL[31] and SHELX-2013[32]
program systems. Mass spectra (MS) were recorded on a
Micromass Autospec Premier, Micromass LCT Premier or
a VG Platform II spectrometer using EI, CI or ESI
techniques at the Mass Spectrometry Service of Imperial
College London. CCDC 1883194–1883197 contains the
supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge
[2] For selected, complementary reviews, see: a) C. J.
Pickens, S. N. Johnson, M. M. Pressnall, M. A. Leon, C.
J. Berkland, Bioconjugate Chem. 2018, 29, 686–701; b)
S. Chassaing, V. Bénéteau, P. Pale, Catal. Sci. Technol.
2016, 6, 923–957; c) E. Haldón, M. C. Nicasio, P. J.
Pérez, Org. Biomol. Chem. 2015, 13, 9528–9550; d) S.
Díez-González, Catal. Sci. Technol., 2011, 1, 166–178;
e) Special issue on Click chemistry, Chem. Soc. Rev.
2010, 39, 1221–1408; f) M. Meldal, C. W. Tornøe,
Chem. Rev. 2008, 108, 2952–3015.
Crystallographic
Data
Centre
via
www.ccdc.cam.ac.uk/data_request/cif. FAIR data for
NMR spectra, computational and crystallographic data, see
DOI: 10.14469/hpc/4842.
[3] a) P. Scheiner, Tetrahedron 1967, 24, 349–356; b) P.
K. Kadaba, S. B. Edelstein, J. Org. Chem. 1990, 55,
5891–5894; c) P. K. Kadaba, B. Stanovnik, M. Tisler,
Adv. Heterocyc. Chem. 1984, 37, 217–349; d) G.
L’Abbé, Chem. Rev. 1969, 69, 345–363.
General procedure for the formation of pyrazolines:
Azide (1.0 equiv.) and alkene (2.2 equiv.) were stirred in
deep eutectic solvent (choline chloride/urea = 1:2; 0.5 M)
between 60–80 °C for 16 h, unless stated otherwise. The
reaction mixture was cooled down to room temperature,
diluted with water and EtOAc (2 mL each per mmol of
azide) and the organic layer was separated. The aqueous
layer was extracted twice with EtOAc (2 mL per mmol of
azide), and the combined organic layers were washed with
brine, dried over MgSO4, filtered and concentrated under
reduced pressure to give the crude product, which was
purified by column chromatography (eluent: petroleum
ether/EtOAc 10→50% gradient unless stated otherwise;
reaction crude was dry-loaded onto stationary phase).
Occasionally, the product recovered after column
[4] a) J. Bourgois, M. Bourgois, F. Texier, Bull. Soc. Chim.
Fr. 1978, 9–10, 485–527; b) P. Scheiner, Sel. Org.
Transform. 1970, 1, 328–362.
[5] R. Huisgen, in 1,3-Dipolar Cycloaddition Chemistry,
ed. A. Padwa, Wiley-Interscience, New York, 1984,
vol. 1, pp. 1–176.
[6] a) R. Huisgen, G. Szeimies, L. Möbius, Chem. Ber.
1966, 99, 475–490; b) G. Szeimies, R. Huisgen, Chem.
Ber. 1966, 99, 491–503; c) F. Texier, R. Carrié,
Tetrahedron Lett. 1969, 10, 823–826; d) W. Broeckx,
N. Overbergh, C. Samyn, G. Smets, G. L’Abbé,
Tetrahedron 1971, 27, 3527–3534.
chromatography
required
trituration
with
pentane/Et2O/MeOH (49:49:2) at −18 °C unless stated
otherwise.
6aa: Following the general procedure at 60 °C from ethyl
4-azidobenzoate (1.50 g, 7.8 mmol) and methyl vinyl
ketone (1.40 mL, 17.3 mmol), 6aa was isolated as a yellow
solid (1.89 g, 73%) after column chromatography (eluent:
MeOH/EtOAc 0→20% gradient)
[7] J. S. Yadav, B. V. Subba Reddy, V. Geetha, Synlett
2002, 513–515.
[8] B. Varghese, S. N. Al-Busafi, F. O. Suliman, S. M. Z.
and trituration with Et2O. Mp
Al-Kindy, RSC Adv. 2017, 7, 46999–47016.
121.3–123.7 °C; Rf
=
0.37
(petroleum ether/EtOAc = 50:50);
IR: νmax 3357 (m, N–H), 2982,
[9] For representative examples, see: a) E. Fischer, O.
Knoevenagel, Justus Liebigs Ann. Chem. 1887, 239,
194–206; b) P. J. Slavish, J. E. Price, Q. Jiang, X. Cui,
S. W. Morris, T. R. Webb, Bioorg. Med. Chem. Lett.
2011, 21, 4592–4596; c) Y. Li, L. Wei, J.-P. Wan, C.
Wen, Tetrahedron 2017, 73, 2323–2328.
1706 (m, C=O), 1670 (s, C=O),
1596 (s), 1565 (m), 1539 (s,
C=N), 1501, 1477, 1412 (m),
1363 (m), 1341 (m), 1287 (s), 1263 (s), 1209 (m), 1178 (s),
1124 (m), 1062 (m), 1022 (m), 951 (m), 882, 849 (m), 769
(m), 702, 679, 657, 639 cm–1; 1H NMR (CDCl3, 400
MHz): δ 7.87 (d, J = 9.0 Hz, 2H, H2), 7.00 (br s, 1H, H17),
6.59 (d, J = 9.0 Hz, 2H, H3), 4.40–4.29 (br m, 1H, H16),
4.32 (q, J = 7.0 Hz, 2H, H14), 3.55 (dd, J = 13.5; 5.0 Hz,
1H, H5), 3.47 (dd, J = 13.5; 8.0 Hz, 1H, H5), 3.12 (d, J =
18.0 Hz, 1H, H7), 3.00 (d, J = 18.0 Hz, 1H, H7), 2.45 (s,
3H, H10), 2.30 (s, 3H, H12), 1.36 (t, J = 7.0 Hz, 3H, H15);
13C NMR (CDCl3, 101 MHz): δ 207.3 (C11), 194.0 (C9),
166.5 (C13), 151.3 (C4), 150.0 (C8), 131.6 (C2), 120.3 (C1),
112.1 (C3), 77.7 (C6), 60.4 (C14), 46.9 (C5), 37.0 (C7), 25.6
(C10), 25.3 (C12), 14.4 (C15); HRMS (ES+) calculated for
C17H22N3O4: 332.1610, found: 332.1613 ([M + H]+).
[10] For attempts to render this transformation greener,
see: a) K. Longhi, D. N. Moreira, M. R. B. Marzari, V.
M. Floss, H. G. Bonacorso, N. Zanatta, M. A. P.
Martins, Tetrahedron Lett. 2010, 51, 3193–3196; b) S.
N. Shelke, G. R. Mhaske, V. D. B. Bonifácio, M. B.
Gawande, Bioorg. Med. Chem. Lett. 2012, 22, 5727–
5730; c) Z. Daneshfar and A. Rostami, RCS Adv. 2015,
5, 104695–10470; d) V. Marković, M. D. Joksović,
Green. Chem. 2015, 17, 842–847.
[11] For representative examples, see: a) M. Chiericato, P.
Dalla Croce, G. Carganico, S. Maiorana, J. Heterocycl.
Chem. 1979, 16, 383–384; b) W. Fliege, R. Grashey
and R. Huisgen, Eur. J. Inorg. Chem., 1984, 117,
1194–1214; c) G. Molteni, M. Orlandi, G. Broggini, J.
Chem. Soc., Perkin Trans. 1 2000, 3742–3745.
Acknowledgements
This research was financially supported by Imperial College
London and the EPSRC (DTP studentship to F. S. and
EP/K030760). L. C. gratefully acknowledges financial support
from the Spanish government through the 2017 Salvador de
Madariaga program
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
[12] For attempts to render this transformation greener,
see: a) H. Kaddar, J. Hamelin, H. Benhaoua, J. Chem.
Res., Synopses 1999, 718–719; b) R. Atir, S. Mallouk,
K. Bougrin, M. Soufiaoui, A. Laghzizil, Synth.
Commun. 2006, 36, 111–120.
6487−6497; (c) G. Alagona, C. Ghio. P. I. Nagy. Phys.
Chem. Chem. Phys. 2010, 12, 10173–10188; (d) A.
Kyrychenko, J. Waluk, J. Phys. Chem. A 2006, 110,
11958–11967.
[22] S. A. Patil, P. A. Medina, D. Gonzalez-Flores, J. K.
Vohs, S. Dever, L. W. Pineda, M. L. Montero, B. D.
Fahlman, Synth. Commun. 2013, 43, 2349–2364.
[13] F. Sebest, L. Casarrubios, H. S. Rzepa, A. J. P. White,
S. Díez-González, Green. Chem. 2018, 20, 4023–4035.
[14] a) D. A. Alonso, A. Baeza, R. Chinchilla, G. Guillena,
I. M. Partor, D. J. Ramón, Eur. J. Org. Chem., 2016,
612–632; b) J. García-Álvarez, Eur. J. Inorg. Chem.
2015, 5147–5157; c) Q. Zhang, K. De Oliveira Vigier,
S. Royer, F. Jérôme, Chem. Soc. Rev. 2012, 41,
7108‒7146.
[23] S. A. Patil, P. A. Medina, S. Dever, Jo. W. Ziller,
B. D. Fahlman, J. Chem. Crystallogr. 2014, 44, 82–88.
[24] Compared to triazoline 3aa reported in reference
Fehler! Textmarke nicht definiert., N(1)–C(5) bond
was 0.027(7) Å longer and the N(1)–C(5)–C(4) angle
smaller by 2.2(4)°.
[15] Isoxazoles: J. M. Pérez, D. J. Ramón, ACS
Sustainable Chem. Eng. 2015, 3, 2343–2349.
Pyrrolidines: R. Singh, A. Singh, J. Iran Chem. Soc.
2017, 14, 1119–1129.
[25] D. Subba Reddy, W. R. Judd, J. Aubé, Org. Lett.
2003, 5, 3899–3902.
[26] B. C. Hauback, F. Mo, Z. Kristallogr. 1990, 191, 195–
207.
[16] Cyclohexenes: A. P. Abbott, G. Capper, D. L. Davies,
R. K. Rasheed, V. Tambyrajah, Green Chem. 2002, 4,
24–26.
[27] a) D. J. Cram, W. D. Nielsen, B. Rickborn, J. Am.
Chem. Soc. 1960, 82, 6415–6416; b) E. J. Corey and E.
T. Kaiser, J. Am. Chem. Soc. 1961, 83, 490–491; c) S.
Wolfe, A. Rauk, I. G. Csizmadia, J. Am. Chem. Soc.
1969, 91, 1567–1569.
[17] Triazoles: a) F. Ilgen, B. König, Green Chem., 2009,
11, 848–854; b) M. A. P. Martins, G. C. Paveglio, L. V.
Rodrigues, C. P. Frizzo, N. Zantta, H. G. Bonacorso,
New J. Chem. 2016, 40, 5989–5992. Tetrazoles: S. A.
Padvi, D. S. Dalal, Synth. Commun. 2017, 47, 779–797.
[28] For a similar 1,4,4-trisubstituted triazoline from
acrylonitrile, see reference Fehler! Textmarke nicht
definiert.d.
[18] P. K. Kadaba, Synthesis 1973, 71–84.
[29] a) O. Achmatowicz and J. Michalski, Rocziki Chemii,
1956, 30, 243–251; b) J. J. Krutak, R. D. Burpitt, W. H.
Moore, J. A. Hyatt, J. Org. Chem. 1979, 44, 3447–
3858.
[19] S. P. Simeonov, C. A. M. Afonso, RSC Adv. 2016, 6,
5485–5490.
[20]B3LYP+GD3BJ/6-311++G(d,p)
correction for a highly polar solvent, specified as
solvent=N-MethylFormamide-mixture in the
with
solvent
[30] A. P. Abbott, G. Capper, S. Gray, ChemPhysChem
2006, 7, 803–806.
Gaussian16 program system. Full details, including
input and output files, can be inspected at DOI:
10.14469/hpc/4842
[31] SHELXTL v5.1, Bruker AXS, Madison, WI, 1998.
[32] SHELX-2013, G.M. Sheldrick, Acta Cryst. 2015,
C71, 3–8.
[21] For selected examples of solvent assisted proton
transfer reactions, see: (a) H. S. Rzepa, Guanidine as
proton
transfer
catalyst,
Blog, 2013,
DOI:
10.14469/hpc/6314; (b) P. Lima da Silva, L.
Guimaraꢁs, J. R. Pliego, Jr. J. Phys. Chem. B 2013, 117,
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901614
Advanced Synthesis & Catalysis
FULL PAPER
Cycloaddition Reactions of Azides and Electron-
Deficient Alkenes in Deep Eutectic Solvents:
Pyrazolines, Aziridines and Other Surprises
Adv. Synth. Catal. Year, Volume, Page – Page
Filip Sebest, Kushal Lachhani, Luis Casarrubios,
Andrew J. P. White, Henry S. Rzepa, and Silvia
Díez-González *
10
This article is protected by copyright. All rights reserved.