Zn(OTf)2-Catalyzed Synthesis of 2-Alkynylazetidines and their Ring Expansion to Functionalized 1,4,5,6-Tetrahydropyridines

Article abstract of DOI:10.1002/adsc.201800932

A zinc(II) triflate catalyzed reaction of β-chloro aldimines with terminal alkynes leading to a rapid and efficient formation of 2-alkynylazetidines in good to excellent yield has been described. The catalytic hydrogenation of the 2-alkynylazetidines resulted in acyclic secondary amines by reductive cleavage of the 2-alkynylazetidine. Further, these non-activated 2-alkynylazetidines were ring expanded in a reaction with dimethyl acetylenedicarboxylate in the presence of zinc(II) triflate to give 4-alkynyltetrahydropyridines. Catalytic reduction of these 4-alkynyltetrahydropyridines led to an efficient conversion to 4-alkyltetrahydropyridine carboxylates. (Figure presented.).

Full text of DOI:10.1002/adsc.201800932

Title: Zn(OTf)₂-Catalyzed Synthesis of 2-
Alkynylazetidines and their Ring Expansion to Functionalized
1,4,5,6-Tetrahydropyridines
Authors: Syeda Aaliya Shehzadi, Khushbu Kushwaha, Hans Sterckx,
and Kourosch Abbaspour Tehrani
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.201800932
Link to VoR: http://dx.doi.org/10.1002/adsc.201800932

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Zn(OTf)2-Catalyzed Synthesis of 2-Alkynylazetidines and their
Ring Expansion to Functionalized 1,4,5,6-Tetrahydropyridines
Syeda Aaliya Shehzadi, Khushbu Kushwaha, Hans Sterckx, and Kourosch Abbaspour
Tehrani*
Organic Synthesis, Department of Chemistry, University of Antwerp, Groenenborgerlaan 171, B-2020, Antwerp, Belgium
Fax: (+32)32653233; phone: (+32)32653226; e-mail: *kourosch.abbaspourtehrani@uantwerpen.be
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. A zinc(II) triflate catalyzed reaction of β-chloro Catalytic reduction of these 4-alkynyltetrahydropyridines
aldimines with terminal alkynes leading to a rapid and led to an efficient conversion to 4-alkyltetrahydropyridine
efficient formation of 2-alkynylazetidines in good to carboxylates.
excellent yield has been described. The catalytic
hydrogenation of the 2-alkynylazetidines resulted in acyclic Keywords: Alkyne; 2-Alkynylazetidine; β-Chloro imine;
secondary amines by reductive cleavage of the 2- Cycloaddition; Ring expansion; Schiff base;
alkynylazetidine.
Further,
these
non-activated
2- Tetrahydropyridine; Zinc
alkynylazetidines were ring expanded in a reaction with
dimethyl acetylenedicarboxylate in the presence of zinc(II)
triflate to give 4-alkynyltetrahydropyridines.
exceptions compared to aziridines.[12b,c] Simple
methods for the synthesis of 1,2,3-trisubstituted
azetidines are rare. The synthesis of 2-
alkynylazetidines remains a new and challenging task
as these compounds bear the potential to be ring
expanded to other azaheterocyclic compounds. So far,
2-alkynylazetidines have been reported in literature
as intermediates or as side products.[13] For instance
Ohno and Tanaka et al.[14] isolated 2-
alkynylazetidines as side products from the Pd(0)
catalyzed cyclization of aminoalkyl substituted
bromoallenes. The synthesis of 1-tosyl-2-
alkynylazetidine in 38% yield by the intramolecular
Mitsunobu reaction of a 1,3-aminoalcohol can be
mentioned as a single example.[15]
The most general approach to 2-alkynylazetidines is
the controlled reduction of the corresponding
azetidin-2-ones with chloroalane.[13b,16] Alkynyl
substituted azetidinones may be prepared via
Mitsunobu cyclization of N-(3-hydroxypropanoyl)
amides (Scheme 1, route a)[16], ketene-alkynylimine
cycloaddition (Scheme 1, route b)[16,17] or ester
enolate-alkynyl imine cycloaddition (Scheme 1, route
c)[16]. Finally, Yanpeng et al. described the Cu(I)
iodide catalyzed synthesis of highly functionalized 4-
alkynyl-2-imino azetidines starting from terminal
alkynes, sulfonyl azides and imidoyl chlorides while
by using triethylamine as base for this reaction.[18] No
reduction of these compounds to the corresponding
azetidines has been reported.
Introduction
Azetidines represent an important class of strained
cyclic compounds from both biological[1] and
synthetic point of view.[2] The azetidine moiety is a
valuable constituent of many naturally occurring
products
e.g.
azetidine-2-carboxylic
acid,
penarisidine A and B,[3,4] polyoxin and mugineic acid
and others.[5-7] Recently, substituted azetidines have
attracted considerable attention because of the diverse
biological activities associated to this type of
compounds. For instance, 3-alkoxy- and 3-
aryloxyazetidines have been described as G-protein
coupled receptor agonists,[8] inhibitors of stearoyl-
coenzyme A -9 desaturase,[9] and as antibacterial
agents.[10] Azetidines also act as building blocks for
some drugs such as Azelnidipine, a calcium channel
blocker which is used as antihypertensive agent[11a]
and in lincosamides, which are antibacterial protein
synthesis inhibitors[11b]
.
The chemistry of azetidines has not been explored as
intensively as that of their lower homologue,
aziridines, which already have demonstrated their
broad applicability as versatile building blocks
through a variety of ring-opening and ring-
transformation
reactions.[12a]
Although
some
syntheses of highly functionalized azetidines have
been described with the purpose of converting them
by ring expansion methods to biologically active
heterocyclic moieties, these examples are still
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
For the initial optimization, N-(3-chloro-2,2-
dimethylpropylidene)pentan-1-amine
(1g)
and
phenylacetylene (2a) were used as model substrates.
β-Chloro aldimine 1g was synthesized by
condensation of 3-chloro-2,2-dimethylpropanal and
n-pentylamine in the presence of MgSO4 as
dehydrating agent.
In a first attempt, one equivalent of phenylacetylene
(2a) was reacted with one equivalent of β-chloro
aldimine 1g using 25 mol % AgOTf at 50 °C in
dichloromethane. The analysis of the reaction mixture
showed highly unsatisfactory results with only 3%
conversion to azetidine 5ga and 3% of imine-alkyne
coupling to give 3ga (ring open product) namely 5-
chloro-4,4-dimethyl-N-pentyl-1-phenylpent-1-yn-3-
amine (Table 1, entry 1). The use of Cu-salts under
similar conditions mainly gave homocoupling of
phenylacetylene with no trace of azetidine 5ga (Table
1, entries 2-4), while ferric chloride did not provoke
any reaction and InCl3 gave only 9% yield of
azetidine (Table 1, entries 5-6). With 25 mol % of
Cu(OTf)2 in toluene at elevated temperature (80 °C)
under argon, the imine-alkyne coupling and ring
closure could be improved with less formation of the
homo-coupled product 2aa (Table 1, entry 7).
Since combination of two catalysts sometimes
enhances the overall catalytic activity, a combination
of 25 mol % In(OTf)3 and 10 mol % CuBr was
employed. Both reactions in dichloromethane and
toluene gave unsatisfactory results (Table 1, entries
8-9).
In view of previous observations by our group where
In(OTf)3 was found to be the best catalyst for the
alkynylation of chlorinated aldimines,19 its
application provided promising yields regarding the
alkynylation of 1g (formation of 3ga) despite the fact
that in most cases, a mixture of azetidine 5ga and
non-cyclized product 3ga was obtained
A major improvement in the yield of 5ga was
observed when reaction was performed with 50
mol % ZnCl2 in toluene at 100 °C. More importantly,
besides 80% yield of the desired azetidine 5ga, no
ring opened product 3ga was observed (Table 1, entry
19).
Scheme 1. Previous and current approaches to synthesize
2-alkynyl azetidines.
From the literature overview, it can be concluded that
there is still room for new syntheses of 2-
alkynylazetidines, especially because the reactivity of
these bifunctional compounds has not been
investigated in detail.
Herein, we describe the efficient synthesis of 2-
alkynyl azetidines 5 starting from β-chloro aldimines
1 and terminal alkynes 2 in good to excellent yields
using Zn(OTf)2 as catalyst. The method includes
initial alkynylation of β-chloro aldimines 1 to give
propargyl amines 3, followed by ring closure to
deliver 1,2,3-trisubstituted azetidine 5 in one step
through the formation of azetidinium salt 4 (Scheme
2).
Table 1. Preliminary screening of the reaction conditions
for the synthesis of 5ga.
Entrya Catalyst/
mol%
T/t
(°C/h)
Solvent
Yield (%)b
5ga/3ga/2aa
3/3/0
Scheme 2. Proposed synthesis of 2-alkynylazetidines.
1
2
3
AgOTf/25
CuBr/25
Cu(OTf)2/25
50/18
50/18
50/18
CH2Cl2
CH2Cl2
CH2Cl2
Results and Discussion
0/4/12
0/7/25
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
4
5
6
7
8
CuBr2/25
FeCl3/25
InCl3/25
Cu(OTf)2/25
In(OTf)3/
50/18
50/18
50/18
80/18
50/18
CH2Cl2
CH2Cl2
CH2Cl2
PhMe
0/2/10
NRc
9/3/0
21/12/8d
0/25/5
CH2Cl2
CuBr (25/10)
In(OTf)3/
9
100/18
PhMe
0/30/13
Entrya Catalyst/
mol%
T/t
(°C/h)
Solvent
Yield 5aa
(%)b
CuBr (25/10)
In(OTf)3/25
In(OTf)3/50
In(OTf)3/25
In(OTf)3/25
In(OTf)3/50
In(OTf)3/50
In(OTf)3/50
10
11
12
13
14
15
16
17
18
50/18
50/24
100/18 DMSO
100/24
70/24
70/24
CH2Cl2
CH2Cl2
0/25/0
0/39/0
complex
10/33/0
SMe
SMe
25/50/0
SMe
1
ZnCl2/50
100/24 PhMe
89
29
6
2
ZnCl2/50
100/6
100/6
PhMe
PhMe
3
ZnCl2/25
PhMe
DCE
THF
PhMe
CH2Cl2
CH2Cl2
PhMe
4
ZnCl2/25
100/18 PhMe
50/18 DCM
40
16
35c
55
13
97
NRd
87
72
93
90
76
49
5
ZnCl2/25
6
ZnCl2/50
100/0.5 PhMe
100/24 PhMe
100/24 PhMe
100/24 PhMe
100/24 PhMe
100/24 PhMe
100/12 PhMe
100/24 PhMee
100/24 PhMee
100/24 PhMee
100/24 PhMee
120/24
7
ZnBr2/50
In(OTf)3/100 55/24
In(OTf)3/50
ZnCl2/50
55/24
29/0/0
80/0/0
8
ZnI2
19
a)
100/18
9
Zn(OTf)2/50
Zn(OAc)2/50
Zn(OTf)2/25
Zn(OTf)2/50
Zn(OTf)2/25
Zn(OTf)2/20
Zn(OTf)2/15
Zn(OTf)2/10
Reactions were performed at 0.5 mmol scale of 2a and
1g in 2 mL of solvent. b) Yields were calculated from the
1H NMR from the reaction mixture after basic work up
using 1,3,5-trimethoxybenzene as internal standard. c) No
10
11
12
13
14
15
16
d)
e)
Reaction.
Reaction was performed under argon.
Starting Material.
In order to simplify the NMR spectral analysis further
optimization was performed using N-(3-chloro-2,2-
a) Reactions were performed at 0.5 mmol scale of 1a and
2a in 2 mL of solvent. b) 1H NMR yields after washing with
0.5 N NaOH. c) MW: microwave conditions: 200 W, 100
dimethylpropylidene)propan-1-amine
(1a).
By
applying the conditions from entry 19 (Table 1) to
compound 1a, 89 % of the corresponding azetidine
was obtained (Table 2, entry 1). By using lower
amounts of ZnCl2, or at lower temperature or for
shorter times, lower yields of 5aa were obtained
(Table 2, entries 2-5). Under microwave conditions at
100 °C for 30 min, only 35% yield of 2-alkynyl
azetidine 5aa was obtained (Table 1, entry 6).
d)
psi. No reaction. e ) Toluene was dried over molecular
sieves.
Finally, using the optimized conditions for the alkyne
coupling and azetidine formation, a one pot reaction
starting from β-chloropropanal, n-propylamine and
phenylacetylene (2a) was investigated. This approach
was abandoned as it did not provide the expected
azetidine 5aa and only starting materials were
recovered. Based on the results of the catalyst
screening (Table 1) and optimization experiments
with Zn(OTf)2 (Table 2), the best reaction condition
was established to be the reaction of one equivalent
of alkyne (2) and aldimine (1) in the presence of 25
mol% of Zn(OTf)2 as catalyst in dry toluene at
100 °C for 24 h, in a sealed vial (Table 2, entry 13).
With these optimized conditions in hand, the scope of
In the next step, the effect of other Zn(II) salts, like
ZnBr2, ZnI2, Zn(OAc)2 and Zn(OTf)2 was
investigated (Table 2, entries 8-10). The use of 50
mol % of the stronger Lewis acid Zn(OTf)2 gave an
optimum yield of 97% of the desired azetidine 5aa
(Table 2, entry 9). As there is a great difference from
a sustainability point of view, the catalyst loading
was reduced to 25 mol % of Zn(OTf)2 giving a little
lower but still acceptable yield of 87% compared to
entry 9 (Table 2, entry 11). The same reaction using
dry toluene gave a 93% yield of the respective
azetidine 5aa (Table 2, entry 13). As both the
aldimine and Zn(OTf)2 are moisture sensitive
compounds the following experiments were
performed in dry toluene.
Zn(OTf)2-catalyzed
alkynylation
of
β-chloro
aldimines, containing various nitrogen substituents
was explored. The β-chloro aldimines 1 were
prepared in good yield by the condensation of 3-
chloro-2,2-dimethylpropanal with different primary
amines in dichloromethane in the presence of
anhydrous MgSO4 (see SI). For instance, under
optimized conditions, β-chloro aldimines with N-i-
propyl (1b), N-allyl (1c), N-cyclopentyl (1d) and N-
cyclohexyl (1e) groups reacted with phenylacetylene
(2a) to deliver the expected 2-alkynylazetidines in
good to excellent yields (Table 3, 5aa-ea). Various
Table 2. Optimization of imine-alkyne coupling with
different zinc(II) salts.
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
other N-substituents such as, methylcyclopropyl (1f),
n-pentyl (1g), i-amyl (1h), t-Bu (1i), benzyl (1j), 4-
methoxybenzyl (1k) and 2,4-dimethoxybenzyl (1l)
gave corresponding azetidines in moderate to good
yields (Table 3, 5fa-la). In case of N-
cyclopropylaldimines (1m), instead of the expected
azetidine 5ma, an unidentified product was obtained.
Inductively electron-withdrawing groups such as 4-
chloro, 4-bromo and 2-methyl-4-fluoro groups (2f-h)
were also well tolerated and afforded the
corresponding products in moderate to good yield
(Table 4, 5af-ah). Strong electron withdrawing
groups like in 4-(trifluoromethyl)phenylacetylene (2i)
gave the lowest yield of 52% (Table 4, 5ai).
Besides arylacetylenes, a number of acyclic and
cyclic aliphatic acetylenes 2j-l was also evaluated.
With cyclohexylacetylene (2j), a good yield (77%)
was obtained (Table 4, 5aj). With n-pentyne (2k) and
n-hexyne (2l) moderate yields (68% & 63%) of
respective azetidines were obtained after simple
aqueous basic workup (Table 4, 5ak-al).
Table 3. Scope of the azetidine synthesis: variation of
amine component.a
Table 4. Scope of the azetidine synthesis: variation of the
alkyne component.a
a)
All reactions were performed using 0.5 mmol of 1a-m
and 2a in 2 mL of toluene in a sealed vial under air for 24
h. b) 1H NMR yield after basic workup. c) Isolated yields
after chromatography using a short column of silica gel.
a)
All reactions were performed using 0.5 mmol 1a or 1g
with 1 equiv. 2b-l in toluene (2 mL) in a sealed vial under
air for 24 h. b) 1H NMR yield after basic workup. c) Isolated
yield after chromatography using a short column of silica
gel.
Further, this reaction was extended to aryl and alkyl
acetylenes (Table 4). The reaction of N-
propylaldimine 1a with various aromatic alkynes
furnished good to excellent yields of azetidines. A
variety of mild electron-donating substituents such as
o-, m-, p-methyl and p-ethyl (2a-e) groups were well
tolerated under the optimized reaction conditions and
the corresponding 2-alkynylazetidines were obtained
in good yields (92-84%) (Table 4, 5ab-ae).
Surprisingly, a different reaction product 6a was
isolated in 68 % yield in the reaction of 4-
methoxyphenylacetylene (2m) and β-chloro aldimine
1a in the presence of Zn(OTf)2 (Scheme 3).
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
Scheme 3. Reaction of 4-methoxyphenylacetylene with N-
propyl-β-chloro aldimine.
In order to confirm that the n-propyl group is not
playing any role in this transformation, a reaction
involving 4-methoxyphenylacetylene (2m) and N-
isopropyl β-chloro aldimine (1b) was performed.
Again, in this case 1-methoxy-4-(4-methylpent-3-en-
1-yn-1-yl)benzene 6a, was obtained after flash
column chromatography.
In order to gain some understanding of this reaction it
was extended to a series of differently substituted
methoxyarylacetylenes, e.g. 2-methoxy (2n), 3-
methoxy
(2o),
2-methyl-4-methoxy
(2p)
phenylacetylenes and 5-methoxynaphthylacetylene
(2q). Interestingly, the reaction of 1a with 2-
methoxyphenylacetylene
(2n)
or
6-
methoxynaphthylacetylene (2q) furnished both the
corresponding 2-alkynylazetidines 5an and 5aq and
the 1,3-enyne product 6b and 6d respectively (Table
5, entry 2 and 5).
a) Reactions were performed at 0.5 mmol scale of 1a and 2
in 2 mL of toluene. b) Yields after column chromatography
on silica gel. c) Besides 6c also an imine 6c' was isolated in
18% yield.
The reaction with an electron withdrawing 3-methoxy
substituent on phenylacetylene (2o) gave only the
corresponding 2-alkynylazetidine 5ao with a low
yield (26%) (Table 5, entry 3) and no 1,3-enyne
product was observed. For further understanding the
reaction was performed between 1a and 2-methyl-4-
methoxyphenylacetylene (2p). Besides the formation
of the azetidine fragmentation product, 4-methoxy-2-
methyl-1-(4-methylpent-3-en-1-yn-1-yl)benzene (6c)
in 40% yield, another side product 6c' was obtained
in 18% yield (Table 5, entry 4).
To confirm whether the enyne 6 is formed directly
from 3a and 4 or via the alkynylazetidine 5, 2-((2-
methoxyphenyl)ethynyl)-3,3-dimethyl-1-
propylazetidine (5an) was dissolved in toluene and
heated in the absence and in the presence of 25
mol % of Zn(OTf)2 (Scheme 4). The reaction mixture
containing Zn(OTf)2 afforded both starting azetidine
5an and 1-3-enyne type product 6b in 2:3 ratio
Table 5. Effect of methoxy substituted aryl acetylenes on
2-alkynylazetidine synthesis.
1
(shown by H NMR spectrum) (Scheme 4, route b),
while the mixture without Zn(OTf)2 showed not even
trace of 1,3-enyne product 6b, only starting material
with few extra peaks were observed (Scheme 4, route
a). Based on this result it may be concluded that
Zn(OTf)2 is involved in the cleavage of azetidine ring.
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
with nucleophiles occur very well in coordinating and
polar solvents the screening was started in DMF and
DMSO, using ZnCl2 as catalyst. The first reaction
was performed using equimolar amounts of azetidine
5aa and DMAD 9a in the presence of 20 mol %
ZnCl2 in DMF at 100 °C for 24 h. To our delight a
44% yield of [4+2]-cycloaddition type product 10a
namely dimethyl 5,5-dimethyl-4-(phenylethynyl)-1-
propyl-1,4,5,6-tetrahydropyridine-2,3-dicarboxylate
was obtained. In DMSO the yield was reduced to
38% (Table 6, entry 1-2), while in a non-polar solvent
like toluene, a similar yield (41%) was obtained
(Table 6, entry 3).
When the reaction was performed at 90 °C for 24 h
using 20 mol % of ZnCl2 in the presence of 2-
methyltetrahydrofuran, a very promising 85% yield
was obtained, while on changing the solvent to THF
and MTBE, yields were reduced to 54% and 36%,
Scheme 4. Formation of 1,3-enyne product 6b through Zn-
catalyzed azetidine ring cleavage.
A plausible mechanism was proposed for the
azetidine ring cleavage and for the formation of 1,3-
en-yne product (see SI).
respectively
(Table
6,
entries
4-6).
In
dichloromethane no product was obtained at all
(Table 6, entry 7).
As a synthetic application of the developed azetidines, Having 2-methyltetrahydrofuran as the preferred
the catalytic hydrogenation of N-propyl-3,3-dimethyl-
2-phenylethynylazetidine (5aa) was attempted. After
stirring 5aa for 12 h at room temperature in the
presence of Pd/C in MeOH and under 2-3 bar H2
pressure, the ring opened product 8 was obtained in
98 % yield. Similar reductive cleavage of azetidines
have been observed in literature with 2-
arylazetidines.[13b]
solvent, other Zn-salts were evaluated. Between
ZnBr2, ZnI2 and Zn(OTf)2, the latter was found to be
the best catalyst giving a 96% yield (Table 6, entries
8-10). Then, the catalyst loading was lowered and
even at 2 mol %, a very high yield of 95% could be
obtained (Table 6, entry 12). The reduction of both,
reaction temperature and time, gave lower yields or
incomplete conversions (Table 6, entries 15-16).
Table 6. Optimization of the ring expansion of 2-
alkynylazetidines with dimethyl-2-butynedioate.
Entry
Catalyst/
mol%
T/t
(°C/h)
Solvent
Yield
10a
(%)a,b
44
38
41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17e
ZnCl2/20
ZnCl2/20
ZnCl2/20
ZnCl2/20
ZnCl2/20
ZnCl2/20
ZnCl2/20
ZnBr2/10
ZnI2/10
Zn(OTf)2/10
Zn(OTf)2/5
Zn(OTf)2/2
Zn(OTf)2/2
Zn(OTf)2/2
Zn(OTf)2/2
No catalyst
Zn(OTf)2/2
100/24
100/24
100/24
90/24
66/16
60/24
50/16
50/16
90/24
90/24
90/24
90/24
50/24
90/15
90/8
DMF
DMSO
PhMe
2-MeTHF
THF
Scheme 5. Reductive cleavage of 2-alkynylazetidine 5aa.
85
54
Azetidine ring expansion reactions are far less studied
than aziridines and the ones which have been
reported start from N-protected or activated
azetidines.[15-18] The ring expansion of unactivated or
N-alkyl azetidines is rarely reported. Bearing in mind
our recent work on the synthesis of 1,6-
dihydropyridines via a Zn(OTf)2-catalyzed three-
component cascade reaction of in situ formed
MTBE
CH2Cl2
36
NRc
45
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
2-MeTHF
39
96
96
95
67+SMd
94
67
NR
SMd
propargylamines
and
dimethylacetylene
dicarboxylate (DMAD)[20] we aimed to explore the
ring expansion study of the synthesized 2-
alkynylazetidines using DMAD (9a) as a strong
electrophile. Since most of the reactions of DMAD
90/24
90/15
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
18f
Zn(OTf)2/25 100/24
All reactions were performed at 0.5 mmol scale of
azetidine 5aa with 1 equiv. of DMAD in 2 mL of solvent
PhMe
60
a)
b)
in a sealed vial.
Yield of crude reaction mixture
determined by 1H NMR.c) No Reaction. d) Starting Material.
e)
One pot reaction of β-chloro aldimine 1a,
phenylacetylene 2a and DMAD. f) β-chloro aldimine 1a is
first reacted with phenylacetylene 2a in the presence of
Zn(OTf)2 in toluene for 24 h at 100 °C, then DMAD was
added and reacted for another 24 h.
The one pot reaction starting from β-chloro aldimine
1a, terminal alkyne 2a and DMAD was also
investigated, but only starting material was recovered
(Table 6, entry 17). When the reaction was performed
in a sequential way, where first β-chloro aldimine 1a
was reacted with phenylacetylene 2a, to deliver
azetidine 5aa in situ, which later reacted with DMAD,
the final product 10a was isolated in 60% yield
(Table 6, entry 18).
With the optimized conditions in hand, the substrate
scope of different azetidines in reaction with DMAD
was explored. First the effect of the N-substituents on
the azetidine was explored and it was observed that
azetidines with primary and secondary (n-propyl 5aa
and iso-propyl 5ba) groups reacted very well along
with other azetidines having primary substituents
such as allyl 5ca, iso-amyl 5da and benzyl 5ja groups
(Table 7, 10a-10e), while the azetidine with the bulky
tert-butyl group 5ia gave a complex reaction mixture
from which, the desired product could not be isolated.
The generality of developed conditions was then
extended to variously alkyne-substituted azetidines.
Azetidines having alkynes with moderately electron
donating groups such as 3-methyl 5ac, 4-methyl 5ad
and 4-ethyl 5ae groups, gave good to moderate yields
(Table 7, 10g-10i), while moderate electron
withdrawing groups such 4-chloro 5af and 2-methyl-
4-fluoro 5ah had a decreasing effect on the yield
(Table 7, 10j-10k). With strong electron withdrawing
group such as 4-CF3 5al, no product was obtained.
a)
All reactions were performed on 0.50 mmol scale of
azetidines 5 with 1 equiv. of DMAD (9a) in 1 mL of 2-
b)
MeTHF in sealed vials.
Yields after column
chromatography. c) Reaction time is 20 h.
Table 7. [4+2]-Cycloaddition of 2-alkynylazetidines 5
with DMAD for the synthesis of 10a-l.a,b
In order to show the robustness and ease of use of the
developed method a gram-scale experiment was
conducted. On a 1 g scale of 5aa with DMAD, the
product 10a could be isolated in 90% yield. Next,
the regiospecificity of the ring expansion was
evaluated by the reaction of azetidine 5aa with
methyl propiolate (9b). We were delighted to see that
2-alkynylazetidine 5aa reacted very well with one
equivalent of methyl propiolate (9b) under the
optimized conditions to give only one product 11a,
namely methyl 5,5-dimethyl-4-(phenylethynyl)-1-
propyl-1,4,5,6-tetrahydropyridine-3-carboxylate
(Scheme 6). This suggested that the reaction is
regiospecific due to the nucleophilic attack of the
azetidine on the more electrophilic alkyne carbon of
9b.
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
The reaction of N-benzyl 4-alkynyltetrahydopyridine-2,3-
dicarboxylate (10d) with hydrogen gas, led to
debenzylation and reduction of the triple bond leading to
N-deprotected enamine 12b in 69% yield (Scheme 9).
Scheme 6. Formation of 4-alkynyltetrahydropyridine-3-
carboxylate 11a from 5aa and methyl propiolate (9b).
A plausible mechanism, was proposed for this
transformation which could be described as 1,4-
dipolar cycloaddition on an alkyne (Scheme 7).
However, no experimental proof for this was
obtained.
Scheme 9. Catalytic debenzylation/reduction of N-benzyl-
4-alkynyltetrahydropyridine-2,3-dicarboxylate 10d.
Conclusion
In summary, a highly efficient and operationally
simple
Zn(II)-catalyzed
synthesis
of
2-
alkynylazetidines 5 starting from imines and
acetylenes has been developed. The reaction proceeds
via a zinc(II)-catalyzed addition of an acetylide to a
β-chloro aldimine, followed by a cyclization of the in
situ formed -chloro propargylic amine. The
generality of the reaction was illustrated with various
substituents on the imino nitrogen atom and the
alkynes. The use of electron rich acetylenes like
methoxy substituted phenylacetylenes led to
fragmentation of the initially formed 2-
alkynylazetidine to give 1,3-enyne type products 6.
The limitation of the substrates to gem-dimethyl
substituted β-chloro aldimines can be rationalized by
the fact that after addition of the acetylide to the C=N
bond, the cyclization of the resulting propargylamine
to the corresponding azetidines, will be accelerated
by the Thorpe-Ingold effect.[21] Besides this kinetic
effect, ,-dimethyl substitution of β-chloro
aldimines also avoids the formation of less reactive
enamines, which do not easily react with acetylides.
The presence of a triple bond in combination with a
strained azetidine makes compounds 5 interesting
building blocks for further elaboration. The N-alkyl-
2-alkynylazetidines behaved, in combination with
electrophilic alkynes 9a-b like 1,4-dipoles and led to
ring expanded products, 4-alkynyltetrahydropyridine
carboxylates 10. The catalytic hydrogenation of 2-
alkynylazetidines not only reduced the triple bond but
also opened the ring to give amino alkanes with the
desired substituents. The reactivity of 2-
alkynylazetidines towards strong electrophiles such
as phenylisothiocyanates is underway.
Scheme 7. Proposed mechanism for the synthesis of 4-
alkynyltetrahydropyridines 10-11.
The synthesized tetrahydropyridine-2,3-dicarboxylate
10a could be hydrogenated in the presence of Pd/C in
methanol. The triple bond was reduced completely
within 3 h under one atmosphere of H2, while the
double bond remained intact to deliver 12a as final
product in good yield (Scheme 8).
Experimental Section
Scheme 8. Catalytic reduction of 4-alkynyltetrahydro-
pyridine-2,3-dicarboxylate 10a.
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
General experimental procedure for the synthesis of 2-
alkynylazetidines
[3] J. F. Ma, K. Nomoto, Plant Physiol. 1993, 102, 373–
378.
[4] T. Takemoto, K. Nomoto, S. Fushiya, R. Ouchi, G.
Kusano, H. Hikino, S. I. Takagi, Y. Matsuura, M.
Kakudo, Proc. Jpn. Acad. Ser. B, 1978, 54, 469-473.
In an oven dried 10 mL vial containing of dry toluene (2
mL) and a stirring bar, the aldimines (1a-k) (0.50 mmol),
acetylenes (2a-p) (0.50 mmol, 1 equiv.) and Zn(OTf)2
(0.046 g, 0.25 mmol) were added successively and the vial
was sealed under air. The reaction mixture was stirred at
100 °C for 24 h. Afterwards the reaction mixture was
diluted with CH2Cl2 (10 mL) washed and extracted with
0.5 N NaOH (10 mL). The organic phase was dried using
MgSO4 and the solvent was removed under reduced
pressure. The crude reaction mixture was purified by
chromatography over silica gel using n-heptane/ethyl
acetate (80:20) as eluent to afford the pure products.
[5] S. Knapp, Y. H. Dong, Tetrahedron Lett. 1997, 38,
3813-3816.
[6] H. Takikawa, T. Maeda, K. Mori, Tetrahedron Lett.
1995, 36, 7689-7692.
[7] a) H. Yoda, T. Uemura, K. Takabe, Tetrahedron Lett.
2003, 44, 977-979; b) A. Feula, S. S. Dhillon, R.
Byravan, M. Sangha, R. Ebanks, M. A. H. Salih, N.
Spencer, L. Male, I. Magyary, W.-P. Deng, F. Müller, J.
S. Fossey, Org. Biomol. Chem. 2013, 11, 5083-5093.
[8] M. C. T. Fyfe, W. Gattrell, C. M. Rasamison, PCT Int.
Appl. 2007, WO 2007116230 Al; Chem. Abstr. 2007,
147, 469218.
General experimental procedure for the synthesis of 4-
alkynyltetrahydropyridine carboxylates (10-11)
[9] E. Isabel, R. Oballa, D. Powell, J. Robichaud, PCT Int.
Appl. 2007, WO 2007143823 Al; Chem. Abstr. 2007,
148, 78872.
The 2-alkynyl-N-alkylazetidine (5aa-al) (0.50 mmol) was
added in a 10 mL oven dried pressure tube containing a
magnetic bar, followed by addition of activated alkyne (9a-
b) (0.50 mmol) and Zn(OTf)2 (3.6 mg, 10.00 µmol, 2
mol %). Then, 2-methyltetrahydrofuran (1 mL) was added.
The vial was sealed under air and heated at 90 °C for 15 h.
Afterwards the reaction mixture was diluted with CH2Cl2
(10 mL), washed with 0.5 N NaOH (10 mL) and extracted
with CH2Cl2 (3 × 10 mL). The organic phase was dried
using MgSO4, filtered and the solvent was removed in
vacuo. The crude reaction mixture was purified by
chromatography over silica gel using n-heptane/ethyl
acetate (80:20) as eluent to afford the final products (10a-k
and 11a).
[10] V. P. V. N. Josyula, A. R. Renslo, PCT Int. Appl.
2007, WO 2007004049 Al; Chem. Abstr. 2007, 146,
142631.
[11] a) K. Oizumi, H. Nishino, H. Koike, T. Sado, M.
Miyamoto, T. Kimura, Jpn. J. Pharmacol. 1989, 51,
57-64; b) H. O’Dowd, J. G. Lewis, J. Trias, R. Asano,
J. Blais, S. L. Lopez, Craig K. Park, C. Wu, W. Wang,
M. F. Gordeev, Bioorganic Med. Chem. Lett. 2008,
18, 2645-2648.
[12] a) H. Ohno, Chem. Rev. 2014, 114, 7784-7814; b) D.
Antermite, L. Degennaro, R. Luisi, Org. Biomol.
Chem. 2017, 15, 34-50; c) S. Kenis, J. Org. Chem.
2012, 77, 5982-5992.
Acknowledgements
[13] a) G. H. Hakimelahi, A. A. M. Movahedi, A. A.
Saboury, V. Osetrov, G. A. Khodarahmi, K.-S. Shia,
Europ. J. Med. Chem. 2005, 40, 339-349; b) M.
Ojima, T. Zhao, K. Yamato, Nakahashi, J. Org. Chem.
1991, 56, 5263-5277.
This work was financed by Erasmus mundus and the University of
Antwerp (BOF).
References
[14] a) H. Ohno, K. Ando, H. Hamaguchi, Y. Takeoka, T.
Tanaka, J. Am. Chem. Soc. 2002, 124, 15255-15266;
b) H. Hamaguchi, S. Kosaka, H. Ohno, T. Tanaka,
Angew. Chem. Int. Ed. 2005, 44, 1513-1517; c) H.
Hamaguchi, S. Kosaka, H. Ohno, N. Fujii, Tanaka, T.
Chem. Eur. J. 2007, 13, 1692-1708.
[1] a) N. H. Cromwell, B. Phillips, Chem. Rev. 1979, 79,
331–358; b) J. A. Moore, R. S. Ayers, In Chemistry of
Heterocyclic Compounds-Small Ring Heterocycles;
Hassner, A., Ed.; Wiley: New York, NY, 1983; pp 1–
217. Part 2; c) F. Couty, G. Evano, Org. Prep. Proced.
Int. 2006, 38, 427-465; d) D. Ferraris, S. Belyakov, W.
X. Li, E. Oliver, Y. S. Ko, D. Calvin, S. Lautar, B.
Thomas, C. Rojas, Curr. Top. Med. Chem. 2007, 7,
597–608.
[15] A. Okano, K. Tsukamoto, S. Kosaka, H. Maeda, S.
Oishi, T. Tanaka, N. Fujii, H. Ohno, Chem. Eur. J.
2010, 16, 8410-8418.
[16] N. Kern, M. Hoffmann, A. Blanc, J.-M. Weibel, P.
[2] a) S. Stankovic, M. D'hooghe, K. Abbaspour Tehrani,
N. De Kimpe, Tetrahedron Lett. 2012, 53, 107-110; b)
A. Feula, L. Male, J. S. Fossey, Org. Lett. 2010, 12,
5044–5047; d) M. J. Brown, G. J. Clarkson, G. G.
Inglis, M. Shipman, Org. Lett. 2011, 13, 1686-1689; e)
F. Couty, F. Durrat, D. Prim, Tetrahedron Lett. 2003,
44, 5209–5212.
Pale, Org. Lett. 2013, 15, 836-839.
[17] B. Alcaide, C. Polanco, M. A. Sierra, J. Org. Chem.
1998, 63, 6786-6796.
[18] Y. Xing, H. Zhao, Q. Shang, J. Wang, P. Lu, Y.
Wang, Org. Lett. 2013, 15, 2668-2671.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
[19] a) C. C. Malakar, B. U. W. Maes, K. Abbaspour
Tehrani, Adv. Synth. Catal. 2012, 354, 3461-3467; b)
K. Kushwaha, C. C. Malakar, S. Stas, F. Lemière, K.
Abbaspour Tehrani, RSC Advances 2015, 5, 10139-
10151.
[20] Syeda Aaliya Shehzadi, C. M. L. Vande Velde, A.
Saeed, K. Abbaspour Tehrani, Org. Biomol. Chem.
2018, 16, 3241-3247.
[21] M. E. Jung, G. Piizzi, Chem. Rev. 2005, 105, 1735-
1766.
10
This article is protected by copyright. All rights reserved.

10.1002/adsc.201800932
Advanced Synthesis & Catalysis
FULL PAPER
Zn(OTf)2-Catalyzed Synthesis of 2-
Alkynylazetidines and their Ring Expansion to
Functionalized 1,4,5,6-Tetrahydropyridines
Adv. Synth. Catal. Year, Volume, Page – Page
Syeda Aaliya Shehzadi, Khushbu Kushwaha,
Hans Sterckx, Kourosch Abbaspour Tehrani*
11
This article is protected by copyright. All rights reserved.