Reaction of N-methylazomethine ylide with aroyl azides: Synthesis of imidazolidin-4-ones

Article abstract of DOI:10.1016/j.tetlet.2019.02.010

Aryl isocyanates generated in situ from aroyl azides react with N-methylazomethine ylide generated in situ from N-methylspiroanthraceneoxazolidine at 210 °C in a microwave reactor to form 3-arylimidazolidin-4-ones in 30–81% yield.

Full text of DOI:10.1016/j.tetlet.2019.02.010

Accepted Manuscript
Reaction of N-methylazomethine ylide with aroyl azides: synthesis of imidazo-
lidin-4-ones
Evgeny M. Buev, Vladimir S. Moshkin, Vyacheslav Y. Sosnovskikh
PII:
S0040-4039(19)30119-4
https://doi.org/10.1016/j.tetlet.2019.02.010
TETL 50604
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
28 December 2018
25 January 2019
4 February 2019
Please cite this article as: Buev, E.M., Moshkin, V.S., Sosnovskikh, V.Y., Reaction of N-methylazomethine ylide
with aroyl azides: synthesis of imidazolidin-4-ones, Tetrahedron Letters (2019), doi: https://doi.org/10.1016/j.tetlet.
2019.02.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Graphical Abstract
Reaction of N-methylazomethine ylide with
Leave this area blank for abstract info.
aroyl azides: synthesis of imidazolidin-4-ones
Evgeny M. Buev, Vladimir S. Moshkin*, Vyacheslav Y. Sosnovskikh
* Corresponding author. Tel: +7 3433899597; e-mail: mvslc@mail.ru (V. S. Moshkin).

Reaction of N-methylazomethine ylide with aroyl azides: synthesis of
imidazolidin-4-ones
Evgeny M. Buev, Vladimir S. Moshkin*, Vyacheslav Y. Sosnovskikh
Institute of Natural Sciences and Mathematics, Ural Federal University, 620000 Ekaterinburg, Russian Federation
* Corresponding author. Tel: +7 3433899597; e-mail: mvslc@mail.ru
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
Aryl isocyanates generated in situ from aroyl azides react with N-methylazomethine ylide
generated in situ from N-methylspiroanthraceneoxazolidine at 210 °C in a microwave reactor to
form 3-arylimidazolidin-4-ones in 30–81% yield.
2018 Elsevier Ltd. All rights reserved.
Available online
Keywords:
Nonstabilized azomethine ylide
[3+2]-Cycloaddition
Aroyl azides
Isocyanates
The [3+2]-cycloaddition of azomethine ylides is an attractive
and useful tool for the construction of various azaheterocycles.1
In particular, nonstabilized azomethine ylides are electron-rich
1,3-dipoles widely used in the synthesis of pyrrolidines from
electron-deficient alkenes.2 Also, they smoothly react with
carbonyl dipolarophiles to give oxazolidine rings which are
versatile intermediates in the synthesis of a wide range of
heterocyclic compounds.3
Scheme 1. Reactions of benzoyl azide (1a).
We commenced our study by examining the model reaction of
benzoyl azide (1a) with N-benzylazomethine ylide derived from
N-(methoxymethyl)-N-(trimethylsilylmethyl)benzylamine8 in the
presence of catalytic TFA at room temperature. Mild conditions
for ylide generation were chosen to reduce the rate of the Curtius
rearrangement. Unfortunately, we found that this reaction
resulted in a complex mixture of products probably due to the
instability of 5-azidooxazolidine 3 (Table 1, entry 1). As
expected, utilization of another classic method for the generation
of ylide 7, condensation of sarcosine and paraformaldehyde in
benzene at reflux (80 C), should occur via the formation of
phenyl isocyanate (6a) and subsequent [3+2]-cycloaddition
(Scheme 1; Table 1, entry 2). Nevertheless, heating the reactants
in benzene with a Dean-Stark trap for 3.5 h surprisingly gave
fenuron (8a) (1,1-dimethyl-3-phenylurea) in 87% yield.
Apparently, the formed phenyl isocyanate (6a) was not reactive
in the 1,3-dipolar cycloaddition under these conditions, and the
generated N-methylazomethine ylide (7) reacted with trace water
to give dimethylamine. The latter readily reacted with isocyanate
to form urea 8a.
Taking into account previous results, the most favourable
precursor of azomethine ylide 7 should allow the reaction to be
carried out without the formation of water and at higher
temperature. These requirements are in line with conditions
recently proposed by us for spiro[anthracene–oxazolidine]
system 9, which undergoes cycloreversion to azomethine ylide
upon heating in an anhydrous solvent.9 Indeed, heating N-
methylspiroanthraceneoxazolidine 9 at reflux with excess
benzoyl azide (1a) in dry o-xylene for 4 h resulted in the
Figure 1. Possible directions for the reaction of aroyl azides and
nonstabilized azomethine ylides.
In the course of our studies regarding azomethine ylide
chemistry,4 we were interested in how they can react with aroyl
azides 1. The latter are readily available and valuable reagents for
the synthesis of amines, amides, ureas and isocyanates.5 Due to
the high reactivity of these compounds several directions could
be assumed for their reaction with dipoles 2 (Fig. 1). Since
nonstabilized azomethine ylides react with the carbonyl group of
isatoic and phthalic anhydrides,3 first, the possibility of
cycloaddition to the C=O bond in acyl azides 1 and formation of
an oxazolidine ring 3 could be expected at low temperatures. The
second direction is a domino-sequence of Curtius rearrangement6
and subsequent cycloaddition of the azomethine ylide to the
formed aryl isocyanate. The chemoselectivity of the latter
reaction is not clear and both oxazolidine 4 or imidazolidine 5
ring formation could be expected. To the best of our knowledge,
there are only a few specific examples of the [3+2]-cycloaddition
of stabilized azomethine ylides derived from aziridines with
isocyanates in the literature.7 Thus, an investigation of the
reactivity of nonstabilized azomethine ylides with aroyl azides 1
is of interest.

formation of imidazolidin-4-one 5a contaminated by an
admixture of starting oxazolidine 9 and traces of fenuron (8a)
(Table 1, entry 3). At the same time, increasing the reaction
temperature to 210 C and heating the reagents in dry o-xylene in
a microwave reactor under an argon atmosphere for 15 min
increased the conversion of oxazolidine 9 into imidazolidine 5ª
(Table 1, entry 4).
Table 1. Optimisation of the reaction conditions.
Yield 5aa
Entry
Conditions
(%)
1a (1 equiv.), N-(MeOCH2)-N-
(Me3SiCH2)benzylamine (1.05 equiv.), TFA
(0.35 equiv.), CH2Cl2, 0–25 C, 24 h
Complex
mixture
1
1a (1 equiv.), sarcosine (1.2 equiv.), CH2O (1.8
equiv.), PhH, reflux, 3.5 h
2
3
4
5
6
7
8a (87%)
1a (1.4 equiv.), 9 (1.0 equiv.), o-xylene, reflux,
4 h
b
–
1a (1 equiv.), 9 (1.1 equiv.), o-xylene, MW, 210
C, 15 min
c
–
1a (1.3 equiv.), 9 (1 equiv.), o-xylene, MW, 210
C, 15 min
d
–
1a (1.5 equiv.), 9 (1 equiv.), o-xylene, MW,
210 C, 15 min
68e
1a (1.5 equiv.), 9 (1 equiv.), MW, 210 C, 15
min
42f
a Yield after the acidic extraction of bases from the crude mixture with 1.5M
HCl, subsequent neutralization with NaHCO3, and extraction with PhMe. b
Mixture of 5a:9:8a (NMR ratio 8:3:1). с Mixture of 5a:9:8a (NMR ratio
13:7:1). d Product was contaminated by starting oxazolidine 9. e Product was
purified from traces of fenuron 8a by dissolution in a mixture of
Et2O/hexane. f Product was purified by column chromatography.
Unfortunately, the product was also contaminated by traces of
spiroanthraceneoxazolidine 9. To overcome this circumstance a
simple modification, consisted of using of 1.5 equiv. of benzoyl
azide (1a), was applied (Table 1, entry 6). Gratifyingly, 1-
methyl-3-phenylimidazolidin-4-one (5a) was isolated in 68%
yield.10,11 It should be noted that imidazolidin-4-one 5a could be
readily isolated from the crude mixture by converting it into the
corresponding water-soluble hydrochloride with subsequent
extraction of non-basic admixtures by PhMe and basification
with NaHCO3. As a side note, we also managed to carry out the
reaction of benzoyl azide (1a) and spirooxazolidine 9 under
solvent-free conditions and product 5a was obtained in 42% yield
after chromatographic purification (Table 1, entry 7). It should be
noted that such 3-arylimidazolidin-4-ones 5 were previously
synthesized by the cyclization of glycinamides and were reported
as muscular relaxants and antirheumatic agents.12
With the optimized conditions in hand, we investigated the
influence of substituents on the aromatic ring (Table 2).
Gratifyingly, aroyl azides 1 bearing chlorine substituents in the
meta- and para-positions reacted with spiro[anthracene–
oxazolidine] 9 to give 3-(chlorophenyl)-1-methylimidazolidin-4-
ones 5b and 5c in 54% and 81% yield, respectively.11 Insertion of
an additional electron-withdrawing NO2 group in the meta-
position did not affect the reaction and imidazolidine 5d was
isolated in 71% yield. Electron-donating methoxy groups were
tolerated and cycloadducts 5e and 5f were obtained in 64% and
69% yield, respectively. In contrast, increasing the number of
electron-donating groups in the aromatic moiety of aroyl azide 1
decreased the yield of 3-arylimidazolidines 5g and 5h (59% and
36%, respectively). Nicotinoyl azide obtained from nicotinic acid
also reacted with N-methylspiroanthraceneoxazolidine 9 in
moderate yield to give 5i (45%).11
a Reagents and conditions: aroyl azide 1 (1.5 equiv.),
spiroanthraceneoxazolidine 9 (1 equiv.), dry o-xylene, MW, 210 C, 15 min. b
Isolated yields of chromatographically purified compounds 5f−j are specified.
Interestingly, cinnamoyl azide, which possesses two reactive
groups, namely, an acyl azide and activated double carbon-
carbon bond, reacted chemoselectively to give enamide 5j in
30% yield. Signals of the pyrrolidine ring were not found in the
1H NMR spectrum of the crude product. Obviously, this is due to
the fact that Curtius rearrangement starts at a lower temperature
than the elimination of azomethine ylide from
spiroanthraceneoxazolidine 9.
The reaction of bulky pivaloyl azide did not result in the
cycloaddition product, probably, due to the inactivity of
azomethine ylide to the sterically hindered C=N moiety of the
formed isocyanate. Notably the reaction also did not occur with
more sterically available C=O group of the isocyanate.
In view of the fact that the probable intermediate of this
reaction is an aryl isocyanate, we performed the reaction of
phenyl isocyanate 6a with spiroanthraceneoxazolidine 9 (phenyl
isocyanate 6a (1.5 equiv.), 9 (1 equiv.), o-xylene, MW, 210 C,
15 min). As expected, the reaction proceeded via [3+2]-
cycloaddition of N-methylazomethine ylide at the C=N group and
gave imidazolidin-4-one 5a in 83% yield (Scheme 2).
Table 2. Synthesis of 1-methyl-3-arylimidazolidin-4-ones 5a–j.a

intermediates – aryl isocyanates and nonstabilized N-
methylazomethine ylide. To the best of our knowledge, the
observed process is a unique example of the conversion of an
oxazolidine ring to an imidazolidine core. The results obtained
reveal new aspects of the diverse azomethine ylides reactivity
and further investigation of the reactions of these ylides is
underway in our laboratory and will be reported in due course.
Scheme 2. Reaction of phenyl isocyanate 6a with spiroanthraceneoxazolidine
9.
Despite the considerable amount of examples of the
cycloadditions of nonstabilized azomethine ylides to carbonyl
compounds in the literature,3 the number of cycloadditions to
imines are limited.13 Considering the obtained results for the
reactivity of isocyanates, which possess both C=O and C=N
groups, we were interested in the comparing the reactions of
isocyanates with other dipolarophiles towards N-
Acknowledgments
The reported study was funded by RFBR according to the
research project № 18-33-00042. We also thank Sergey Usachev
for his remarks.
References and notes
methylazomethine ylide (7).
1. (a) O. Tsuge, S. Kanemasa, Adv. Heterocycl. Chem. 45 (1989) 231;
(b) L.M. Harwood, R.J. Vickers, in: Synthetic Applications of 1,3-
Dipolar Cycloaddition Chemistry Toward Heterocycles and Natural
Products, Chemistry of Heterocyclic Compounds, Vol. 59 (Eds.: A.
Padwa and W.H. Pearson), Wiley: Chichester, 2002, p. 169.
(a) R. Grigg, S. Thianpatanagul, J. Chem. Soc., Chem. Commun.
(1984) 180; (b) C. Nájera, J.M. Sansano, Curr. Org. Chem. 7 (2003)
1105; (c) I. Coldham, R. Hufton, Chem. Rev. 105 (2005) 2765; (d)
V.S. Moshkin, V.Y. Sosnovskikh, P.A. Slepukhin, G.-V.
Röschenthaler, Mendeleev Commun. 22 (2012) 29;(e) J.H. Ryan,
Arkivoc part i (2015) 160; (f) H.A. Döndas, M. de Gracia
Retamosa, J.M. Sansano, Synthesis 49 (2017) 2819.
2. A.G. Meyer, J.H. Ryan, Molecules 21 (2016) 935.
(a) V.Y. Sosnovskikh, M.Y. Kornev, V.S. Moshkin, E.M. Buev,
Tetrahedron 70 (2014) 9253; (b) E.M. Buev, V.S. Moshkin, V.Y.
Sosnovskikh, Tetrahedron Lett. 56 (2015) 6590; (c) A.V.
Pavlushin, V.S. Moshkin, V.Y. Sosnovskikh, Mendeleev Commun.
27 (2017) 628.
5. (a) W. Lwowski, in: The Azido Group, (Ed.: S. Patai), John Wiley
& Sons Ltd.: Chichester, 1971, p. 503; (b) E.F.V. Scriven, K.
Turnbull, Chem. Rev. 88 (1988) 297; (c) S. Bräse, C. Gil, K.
Knepper, V. Zimmermann, Angew. Chem. Int. Ed. 44 (2005) 5188;
(d) Organic Azides: Syntheses and Applications, (Eds.: S. Bräse, K.
Banert), John Wiley & Sons Ltd.: Chichester, 2010.
Scheme 3. Comparison of the reactivity of phenyl isocyanate and different
dipolarophiles.
6. (a) G. L’abbé, Chem. Rev. 69 (1969) 345; (b) D.V. Banthorpe, in:
The Azido Group, (Ed.: S. Patai), John Wiley & Sons Ltd.:
Chichester, 1971, p. 397.
As a model system we chose the reaction of equimolar
amounts of phenyl isocyanate (6a), spiro[anthracene–
oxazolidine] 9 and another dipolarophile using the optimised
conditions for the [3+2]-cycloaddition to isocyanates (o-xylene,
MW, 210 C, 15 min). Thus, benzaldehyde (10) is a well-known
C=O dipolarophile that reacts with the azomethine ylide derived
from precursor 9 to give 5-phenyloxazolidine 11 in 73% isolated
yield.9 Heating the mixture of isocyanate 6a and benzaldehyde 10
with the source of nonstabilized azomethine ylide 9 resulted in a
mixture of imidazolidine 5a and oxazolidine 11 in NMR ratio 1 :
5, respectively (Scheme 3). Under the same conditions, heating
diethyl 2-benzylidenemalonate 12 as a classical C=C
dipolarophile with isocyanate 6a gave a mixture of products with
the predominance of diethyl 1-methyl-4-phenylpyrrolidine-3,3-
dicarboxylate 13 (NMR ratio 5a : 13 1 : 10). Remarkably,
azomethine ylide 7 possesses diverse reactivity and it is able to
act not only as a dipole but also as a base and a source of
iminium cation.14 Thus, we performed a more complicated
comparison consisting of heating N=C=O dipolarophile 6a and
CH-acidic diethyl malonate 14. This experiment resulted in an
almost equal mixture of imidazolidine 5a and diethyl pyrrolidine-
3,3-dicarboxylate 15 (NMR ratio 1 : 1.3). The latter result shows
that the reactivity of ylide 7 as a base and as a dipole toward a
low reactivity isocyanate is comparable. At the same time, it
showed better reactivity with C=O and C=C dipolarophiles than
with N=C=O.
7. (a) H. Benhaoua, F. Texier, R. Carrié, Tetrahedron 42 (1986) 2283;
(b) H. Benhaoua, F. Texier, L. Toupet, R. Carrié, Tetrahedron 44
(1988) 1117; (c) O. Mamoun, H. Benhaoua, Bull. Soc. Chim. Belg.
103 (1994) 753; (d) K. Zhang, P.R. Chopade, J. Louie, Tetrahedron
Lett. 49 (2008) 4306.
8. Y. Terao, H. Kotaki, N. Imai, K. Achiwa, Chem. Pharm. Bull. 33
(1985) 896.
9. E.M. Buev, V.S. Moshkin, V.Y. Sosnovskikh, Org. Lett. 18 (2016)
1764.
10. General procedure for the preparation of 1-methyl-3-
arylimidazolidin-4-ones 10a–i. A 10 mL microwave reaction tube
was charged with a stirrer bar, the corresponding aroyl azide 1 (1.5
mmol), 3'-methyl-10H-spiro[anthracene-9,5'-oxazolidin]-10-one (9)
(1.0 mmol) and dry o-xylene (2 mL). The vial was purged with an
argon atmosphere and sealed with a cap. After pre-stirring for 3
min, the mixture was heated in a microwave reactor at 210 °C for
15 min with stirring. (Caution! Rapid evaluation of N2 starts at
100–120 °C. High pressure in the vial.). After cooling with a
compressed air flow, the resulting mixture was diluted with PhMe
(5 mL). The precipitated anthraquinone was filtered off. The
solution was extracted with cold 1.5 M HCl (10 mL), and the
aqueous phase was washed with PhMe (2 × 5 mL). The aqueous
layer was basified with NaHCO3 to pH 8, and extracted with PhMe
(2 × 5 mL). The organic phase was dried over Na2SO4 and
evaporated under reduced pressure to give the desired product.
Imidazolidinones 5f–j was purified by column chromatography
(eluent: chloroform/ethanol). Product 5h was additionally purified
from trace amounts of urea by dissolution in warm Et2O, urea was
crystallized by addition of hexane, filtered off, washed with hexane,
and solvents were evaporated in vacuo to give 3-arylimidazolidin-
4-one 5h.
In conclusion, we developed a new reaction for the synthesis
of 3-arylimidazolidin-4-ones 5 from aroyl azides 1 and
spiro[anthracene-oxazolidine] 9 via in situ generation of the two

11. 1-Methyl-3-phenylimidazolidin-4-one (5a). Yellow solid, 68%
yield, mp 53−55 °C (previously reported as the hydrochloride salt
mp 188 °C; ref. 12). 1H NMR (400 MHz, CDCl3) δ 7.53 (dd, J =
8.6, 0.9 Hz, 2H, 2,6-HPh), 7.38 (t, J = 8.0 Hz, 2H, 3,5-HPh), 7.17
(t, J = 7.4 Hz, 1H, 4-HPh), 4.55 (s, 2H, 2-CH2), 3.45 (s, 2H, 5-
CH2), 2.54 (s, 3H, NCH3); 13C NMR (101 MHz, CDCl3) δ 170.1,
137.5, 129.2, 125.1, 119.4, 72.3, 59.0, 41.7. HRMS (ESI) calcd for
(C10H13N2O)+ [M+H]+: 177.1022, found: 177.1027.
Viscous yellow oil, 45% yield. 1H NMR (500 MHz, CDCl3) δ 8.60
(d, J = 2.5 Hz, 1H, 2-HPy), 8.42 (dd, J = 4.8, 1.2 Hz, 1H, 4-HPy),
8.20 (ddd, J = 8.4, 2.5, 1.2 Hz, 1H, 6-HPy), 7.33 (dd, J = 8.4, 4.8
Hz, 1H, 5-HPy), 4.58 (s, 2H, 2-CH2), 3.46 (s, 2H, 5-CH2), 2.56 (s,
3H, NCH3); 13C NMR (126 MHz, CDCl3) δ 171.0, 145.5, 139.4,
134.7, 126.9, 124.0, 71.6, 58.7, 41.6. HRMS (ESI) calcd for
(C9H12N3O)+ [M+H]+: 178.0975, found: 178.0978.
12. J. Hellerbach, US Patent 3428646, 1969.
3-(4-Chlorophenyl)-1-methylimidazolidin-4-one (5с). Yellow solid,
81% yield, mp 78−79 °C. 1H NMR (400 MHz, CDCl3) δ 7.49 (d, J
= 9.1 Hz, 2H, Ar), 7.33 (d, J = 9.1 Hz, 2H, Ar), 4.52 (t, J = 0.8 Hz,
2H, 2-СH2), 3.44 (t, J = 0.8 Hz, 2H, 5-СH2), 2.53 (s, 3H, NCH3);
13C NMR (101 MHz, CDCl3) δ 170.2, 136.1, 130.1, 129.2, 120.4,
72.2, 58.9, 41.6. HRMS (ESI) calcd for (C10H12ClN2O)+ [M+H]+:
211.0633, found: 211.0638.
1-Methyl-3-(pyridin-3-yl)imidazolidin-4-one (5i). Product 5i was
synthesized according to the general procedure except for the
reaction work-up. After microwave heating and compressed air
flow cooling, the resulting mixture was diluted with a mixture of
PhMe/hexane (2:1, 8 mL) and the precipitated solid anthraquinone
was filtered off. The solvents were evaporated in vacuo and the
crude product was isolated by column chromatography (eluent:
chloroform/ethanol 100/4; Rf (chloroform/ethanol, 100/10) = 0.36).
13. (a) J. Chastanet, G. Roussi, J. Org. Chem. 53 (1988) 3808; (b) C.
Izquierdo, F. Esteban, J.L.G. Ruano, A. Fraile, J. Alemán, Org.
Lett. 18 (2016) 92; (c) J.K. Laha, K.P. Jethava, K.S.S.
Tummalapalli, S. Sharma, Eur. J. Org. Chem. (2017) 4617; (d) M.
Beuvin, M. Manneveau, S. Diab, B. Picard, M. Sanselme, S.R.
Piettre, J. Legros, I. Chataigner, Tetrahedron Lett. 59 (2018) 4487.
14. E.M. Buev, V.S. Moshkin, V.Y. Sosnovskikh, J. Org. Chem. 82
(2017) 12827.
Supplementary data
Supplementary data associated with this article can be found
in the online version.