Cyclisation at very high temperature. Thermal transformations of N-alkyl and N,N-dialkyl cinnamic amides into pyrrolidin-2-ones under FVT conditions

Article abstract of DOI:10.1016/j.tetlet.2005.03.004

Novel cyclisations of N-alkyl and N,N-dialkyl cinnamic amides to the corresponding pyrrolidin-2-ones under the conditions of flash vacuum thermolysis, are described. It was found that this reaction proceeds at 950-1000°C affording a mixture of isomeric pyrrolidin-2-ones in various yields. Two possible mechanisms are proposed for the process.

Full text of DOI:10.1016/j.tetlet.2005.03.004

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 3093–3095
Cyclisation at very high temperature. Thermal transformations
of N-alkyl and N,N-dialkyl cinnamic amides into
pyrrolidin-2-ones under FVT conditions
*
Stanisław Lesniak and Beata Pasternak
´
´ ´ ´ ´
Department of Organic and Applied Chemistry, University of Łodz, Narutowicza 68, 90-136 Łodz, Poland
Received 2 December 2004; revised 22 February 2005; accepted 1 March 2005
Available online 16 March 2005
Abstract—Novel cyclisations of N-alkyl and N,N-dialkyl cinnamic amides to the corresponding pyrrolidin-2-ones under the condi-
tions of flash vacuum thermolysis, are described. It was found that this reaction proceeds at 950–1000 ꢀC affording a mixture of
isomeric pyrrolidin-2-ones in various yields. Two possible mechanisms are proposed for the process.
ꢁ 2005 Elsevier Ltd. All rights reserved.
Flash vacuum thermolysis (FVT) has been widely ap-
plied in retro-ene reactions.1 It appears from the numer-
ous relevant contributions, that there is no common
mechanism for all the reactions of this kind. In any indi-
vidual case, the pathway can be purely concerted or radi-
cal, however, the former is in most cases preferred. It is
also well known that retro-ene reactions under FVT
conditions are methods for accessing unsaturated (and
often very reactive) species. Indeed, simultaneous for-
mation of vinylketenes and imines in the retro-ene reac-
tions of allenic N,N-dialkylamides was observed by
Wentrup and co-workers,2 however, related lactams,
as expected products of the Staudinger reaction between
these reactive systems, were not detected.
O
Me
Me
p-RC6H4
950-1000 oC
1.5x10-3 Torr
H
N
O
N
Me
p-RC6H4
H
1a R = H
1b R = OMe
2a R = H
2b R = OH
Scheme 1.
was not present in the post-reaction mixture. Similarly,
N,N-dimethyl 3-(4-methoxyphenyl)-acrylamide 1b under-
went cyclisation to afford 1-methyl-4-(4-hydroxyphenyl)-
pyrrolidin-2-one5 2b (43%) (Scheme 1). The reaction
proceeded with simultaneous dealkylation of the methoxy
group.
We report here that cinnamic amides bearing one or two
alkyl groups at the nitrogen atom undergo cyclisation
under FVT conditions. According to WentrupÕs results,
flash vacuum thermolysis of a N,N-dimethyl cinnamic
To determine the limitations of such an approach, vari-
ous N-mono- and N,N-disubstituted cinnamic amides
were investigated. Thus, when amides 1c–f were sub-
jected to FVT in a conventional flow system (950–
1000 ꢀC, 1.5 · 10À3 Torr) the expected pyrrolidin-2-ones
(c-lactams) 2c–f were produced in variable yields as dia-
stereoisomeric mixtures. Our results are summarised in
Scheme 2 and in the Table 1.
amide should theoretically afford
a benzylketene
together with methyl-methylene-amine. However, we
found that thermolysis3 of N,N-dimethyl 3-phenyl-
acrylamide 1a, at 950–1000 ꢀC and under a pressure of
1.5 · 10À3 Torr, gave an unexpected result. 1-Methyl-4-
phenyl-pyrrolidin-2-one4 2a was found as the exclusive
product in 88% yield (Scheme 1). The temperature range
was the lowest possible at which the starting material
Ph
O
R2
950-1000 oC
1.5x10-3 Torr
H
N
R2
O
R1
N
R1
Ph
H
Keywords: Amides; Cyclisation; Flash vacuum thermolysis; c-lactams.
1c-f
2c-f
*
Corresponding author. Tel.: +48 42 635 57 65; fax: +48 42 678 16
09; e-mail: slesniak@chemul.uni.lodz.pl
Scheme 2.
0040-4039/$ - see front matter ꢁ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.03.004

´
S. Lesniak, B. Pasternak / Tetrahedron Letters 46 (2005) 3093–3095
3094
Table 1.
Starting amide
Producta
Yieldb (%)
trans/cis ratioc
Ref.
1c: R1 = Et, R2 = Me
2c: R1 = Et, R2 = Me
2e: R1 = H, R2 = Me
2d: R1 = Ph, R2 = Me
2e: R1 = H, R2 = Me
2f: R1 = H, R2 = Ph
38
12
63
62
50
2/1
2/1
2/1
2/1
2/1
6
7
8
7
9
1d: R1 = Ph, R2 = Me
1e: R1 = H, R2 = Me
1f: R1 = H, R2 = Ph
a In all cases in crude thermolysis mixtures small amounts of styrene were detected, additionally in the case of 1f, traces of bibenzyl were observed
(1H NMR).
b Yields are for pure products isolated by flash chromatography.
c Stereochemical assignments of the trans and cis isomers are based on the 1H NMR spectra.
The formation of the dealkylated product 2e starting
from substrate 1c can be explained as a result of a
secondary retro-ene reaction of 2c, that is, by the b-elimi-
nation reaction of amides. Indeed, when 2c was re-pyr-
olysed at about 1050 ꢀC, lactam 2e was obtained in a
pure state.
H
H
H
R2
FVT
O
R2
O
R2
.
O
N
N
N
R1
R1
R1
.
.
.
A
B
Ph
Ph
1
Ph
The trans and cis isomers of pyrrolidin-2-ones 2 can be
easily distinguished9a from each other on the basis of
the characteristic chemical shifts of protons in the
methyl group at C-5 (for 2c–e) or the methine proton
at C-5 for 2f. For trans isomers, the protons of the
methyl groups at C-5 appear at d ꢀ 1.1 ppm, whereas
related protons in cis isomers resonate at d ꢀ 0.8 ppm.
Similarly, for the trans isomer of 2f the C-5 proton
appears at d = 4.68 ppm, (d, J = 6.6 Hz), whereas that
in the cis isomer is at d = 5.00 ppm (d, J = 7.6 Hz). Thus,
the trans/cis ratios were determined by integration of the
aforementioned signals in the 1H NMR spectra of crude
diastereoisomeric mixtures.
Ph
Ph
R2
HO
R1
N
R2
R2
O
OH
OH
N
N
.
.
R1
R1
C
Ph
2
(a)
Ph
R2
Ph
R2
Ph
H
O
R2
O
N
N
N
R1
R1
R1
2
1
(b)
Scheme 3.
Next we examined the reactivity of N,N-dialkylamides
without an aryl substituent at the 3-position of acrylic
moiety. Thus, N,N-dimethyl acrylamide and trans-but-
2-enoic acid N,N-dimethyl amide were found to be
stable in the whole accessible range of temperatures
(up to 1200 ꢀC). On the contrary, 3-methylbut-2-enoic
acid N,N-dimethylamide underwent the retro-ene reac-
tion easily at 1000 ꢀC, with the quantitative formation
of 3-methylbut-3-enoic acid N,N-dimethylamide. These
results showed clearly that a phenyl group incorporated
into a,b-unsaturated N-alkylamides is essential for
effecting cyclisation to pyrrolidin-2-ones 2.
FVT conditions from the alkylamide group to phenoxyl
or thiophenoxyl radicals via a six- or seven-membered
transition state has been described by McNab.10
Another possible mechanism can be postulated as a pro-
cess in which the key intermediate is a dipolar species
formed via [1,4]-hydrogen atom transfer, and stabilised
by the phenyl group in an analogous fashion to a diradi-
cal (Scheme 3b).
This can be explained in terms of a significant stabilisa-
tion of the biradical A formed by a reversible cleavage
of the C@C double bond, in the first step of the reaction
(Scheme 3a). In our opinion, the cyclisation can proceed
via a biradical process. The biradical A is stabilised by an
adjacent phenyl group, and by resonance between the
canonical forms A and B, in which an unpaired electron
is localised on an oxygen atom. Subsequent intramolecu-
lar [1,4]-hydrogen atom transfer via a five-membered
transition state, from the N-alkyl substituent to an oxyl
group (form B) results in the formation of a new biradical
C which cyclises to an unobserved enol, after a low-en-
ergy rotation around the C–N bond. Alternatively, an
enol biradical C could tautomerise to its keto form before
closure to give directly the c-lactam 2 (path not shown in
Scheme 3). An analogous hydrogen atom transfer under
It was in principle possible that amide 1 could also un-
dergo a retro-ene reaction in line with WentrupÕs results,
with the formation of a corresponding benzylketene and
an imine. An interaction between such highly reactive
species would be expected to lead to zwitterionic inter-
mediates which subsequently cyclise to b- or d-lactams,
as has been found in many related reactions.11 However,
none of these products was detected in our experiments.
In order to exclude the formation of c-lactams as a re-
sult of interaction between the species mentioned above,
we examined the reaction of benzylketene with benzyl-
idene aniline using an Ôacid chloride–imine approachÕ.12
This procedure afforded trans -3-benzyl-1,4-diphenyl-
azetidin-2-one13 in moderate yield, but 4,5-diphenyl-
pyrrolidin-2-one was not detected in the post-reaction

´
S. Lesniak, B. Pasternak / Tetrahedron Letters 46 (2005) 3093–3095
3095
6. Selected data for compounds 2c. The isomers were
separated chromatographically (silica gel, hexane:AcOEt):
cis-1-ethyl-5-methyl-4-phenyl-pyrrolidin-2-one. Mp 106–
108 ꢀC (hexane/dichloromethane). IR (KBr) mC@O 1680
mixture. Therefore, the alternative mechanism involving
a retro-ene reaction of amide 1 was disproved.
Support for a biradical mechanism came from the
thermolysis of 2,N,N-trimethyl-3-phenylacrylamide 1g.
FVT performed under the aforementioned conditions
afforded trans and cis 1,3-dimethyl-4-phenyl-pyrrol-
idin-2-one14 (2:1) in 32% yield, together with a 68%
recovery of the starting amide, which was found as an
equimolar mixture of trans- and cis-isomers. The lower
yield of c-lactam and formation of E- and Z-isomers of
the substrate 1g can be explained by a different stabilisa-
tion of the initial biradical. This reversible process pro-
duces a benzylic and tertiary carbon-centred biradical,
which reacts far less rapidly than the structurally related
system with a secondary carbon-centred radical. As a
consequence of the greater significance of resonance
structure A (compared with B) easier isomerisation of
the starting amide and, at the same time, a lower yield
of the cyclisation would be expected, as found.
cmÀ1 1H NMR (CDCl3, 200 MHz, TMS) d = 0.78 (d,
.
J = 7.6 Hz, 3H), 1.11 (t, J = 7 Hz, 3H), 2.50–3.25 (m, 3H),
3.45–4.15 (m, 3H), 7.10–7.50 (m, 5H). 13C NMR (CDCl3,
50 MHz, TMS) d = 13.04 (CH3), 14.90 (CH3), 35.48
(2 · CH2), 42.72 (CH), 57.19 (CH), 127.48, 128.39, 128.91
(Car), 139.33 (Cq), 174.17 (C@O). MS (70 eV); m/z (%):
203 (4), 175 (45, M+ÀC2H4), 117 (7), 105 (10), 104 (100),
103 (23), 77 (11). HRMS m/z calcd for C13H17NO
203.13101, found 203.13093. Anal. Calcd for C13H17NO
(203.29): C, 76.81; H, 8.43; N, 6.89. Found: C, 76.88; H,
8.56; N, 6.76. trans-1-Ethyl-5-methyl-4-phenyl-pyrrolidin-
2-one. Mp 115–117 ꢀC (hexane/dichloromethane). IR
(KBr) mC@O 1680 cmÀ1 1H NMR (CDCl3, 200 MHz,
.
TMS) d = 1.11 (d, J = 6.6 Hz, 3H), 1.24 (t, J = 7 Hz, 3H),
2.40–3.90 (m, 6H), 7.10–7.50 (m, 5H). 13C NMR (CDCl3,
50 MHz, TMS) d = 17.66 (CH3), 20.55 (CH3), 35.51
(CH2), 39.46 (CH2), 49.86 (CH), 57.83 (CH), 127.48,
127.57, 129.21 (Car), 141.40 (Cq), 176.97 (C@O). MS
(70 eV); m/z (%): 203 (4), 175 (49, M+ÀC2H4), 117 (9), 105
(12), 104 (100), 103 (23), 77 (11). HRMS m/z calcd for
C13H17NO 203.13101, found 203.13089. Anal. Calcd for
C13H17NO (203.29): C, 76.81; H, 8.43; N, 6.89. Found: C,
76.84; H, 8.31; N, 6.72.
In summary, a thermal transformation of cinnamic
amides into pyrrolidin-2-ones, followed by intramolecu-
lar cyclisation, has been demonstrated. Although the
products obtained are fairly simple and can be prepared
by other methods, it should be noted that this type of
cyclisation is reported here for the first time and further
applications of this reaction are now being examined.
Our results show that flash vacuum thermolysis can
provide unexpected results.
7. Langlois, M. Bull. Soc. Chim. Fr. 1971, 2976–2980.
8. Selected data for compound 2d. Purified by chromato-
graphy (silica gel, hexane:AcOEt 1:1). As a diastereoisomeric
mixture. IR (neat): mC@O 1693 cmÀ1 1H NMR (CDCl3,
.
200 MHz, TMS) d = 0.84 (d, J = 7.6 Hz, 3H, cis), 1.21 (d,
J = 6.5 Hz, 3H, trans), 2.65–3.30 (m, 2H), 3.70–4.05 (m,
1H, ), 4.07–4.30 (m, 1H, trans) 4.35–4.72 (m, 1H, cis) 7.15–
7.60 (m, 10H). 13C NMR (CDCl3, 50 MHz, TMS)
d = 14.11(CH3, cis), 19.09 (CH3, trans), 36.03 (CH2, cis),
39.64 (CH2, trans) 42.32 (CH, cis), 46.76 (CH, trans), 59.89
(CH, cis), 63.21 (CH, trans), 126.19, 126.48, 127.57, 127.69
128.30, 129.00, 129.33 (Car), 137.81, 138.30, 138.76, 141.95
(4 · Cq), 173.23 (C@O, trans), 173.60 (C@O, cis). Anal.
Calcd for C17H17NO (251.33): C, 81.24; H, 6.82; N, 5.57.
Found: C, 81.01; H, 7.06; N, 5.29.
Acknowledgements
This research project was supported by grant No 505/
´ ´
681/2003 from the University of Łodz.
References and notes
1. Ripoll, J.-L.; Vallee, Y. Synthesis 1993, 659–677.
2. Bibas, H.; Koch, R.; Wentrup, C. J. Org. Chem. 1998, 63,
2619–2626.
9. (a) Kee, N. K. Tetrahedron Lett. 1997, 38, 5091–5094; (b)
Potapov, V. M.; Gracheva, R. A.; Savina, S. A.; Sivova, L.
A. J. Org. Chem. USSR (Engl. Transl.) 1989, 25, 1694–
1699; Zh. Org. Khim. RU 1989, 1876–1881.
3. Flash vacuum thermolysis was carried out in
a
30 · 2.5 cm2 electrically heated horizontal quartz tube
packed with quartz rings at 1.5 · 10À3 Torr. The com-
pounds 1 (2 mmol) were slowly sublimed from a flask held
at 80 ꢀC into the thermolysis tube preheated to 950–
1000 ꢀC. The products were collected in a CO2–acetone
trap. After thermolysis, the system was brought to
atmospheric pressure allowing a slow warming up to
room temperature and the products were dissolved in
CHCl3. The solvent was removed under reduced pressure
and the products were purified chromatographically on
silica gel and by recrystallisation.
10. Black, M.; Cadogan, J. I. G.; McNab, H. J. Chem. Soc.,
Perkin Trans. 1 1994, 155–159.
11. (a) Manhas, M. S.; Ghosh, M.; Bose, A. K. J. Org. Chem.
1990, 55, 575–580; (b) Bose, A. K.; Manhas, M. S.; Ghosh,
M.; Shah, M.; Raju, V. S.; Bari, S. S.; Newaz, S. N.;
Banik, B. K.; Chaudhary, A. G.; Barakat, K. J. J. Org.
Chem. 1991, 56, 6968–6970; (c) Van der Steen, F. H.; Van
Koten, G. Tetrahedron 1991, 47, 7503–7524.
12. This efficient one-pot procedure is based on the treatment
of the imine with corresponding acid chloride in toluene at
room temperature, in the presence of triethylamine. (a)
Govindachari, T. R.; Chinnasamy, P.; Rajewari, S.;
Chandrasekaran, S.; Premila, M. S.; Nagarajan, K.; Pai,
B. R. Heterocycles 1984, 22, 585–655; (b) Sandhu, J. S.;
Sain, B. Heterocycles 1987, 26, 777–818; (c) Alcaide, B.;
Martin-Cantalejo, Y.; Plumet, J.; Rodriguez-Lopez, J.;
Sierra, M. A. Tetrahedron Lett. 1991, 32, 803–806.
13. Kashima, Ch.; Fukusaka, K.; Takahashi, K. J. Hetero-
cycl. Chem. 1997, 34, 1559–1565.
4. Sato, T.; Chono, N.; Ishibashi, H.; Ikeda, M. J. Chem.
Soc., Perkin Trans. 1 1995, 1115–1120.
5. Selected data for compound 2b: Purified by chromato-
graphy (silica gel, hexane:AcOEt 1:1). Mp 112–114 ꢀC (hex-
ane). IR (KBr, cmÀ1): mOH 3385, mC@O 1675. 1H NMR
(200 MHz, CDCl3, TMS): d = 2.47–2.80 (m, 2H), 2.90
(s, 3H), 3.25–3.75 (m, 3H), 6.70–7.10 (m, 4H). 13C NMR
(50 MHz, CDCl3, TMS): d = 29.82 (NCH3), 36.57 (CH),
39.13 (CH2), 57.37 (CH2), 116.23, 127.93 (2 · Car), 133.52,
156.36 (2 · Cq ar), 175.42 (C@O). Anal. Calcd for
C11H13NO2 (191.23): C, 69.09; H, 6.85; N, 7.32. Found:
C, 69.22; H, 6.84; N, 7.11.
14. Rasmussen, C. R.; Gardocki, J. F.; Plampin, J. N.;
Twardzik, B. E.; Reynolds, A. J.; Molinari, A. J.;
Schwartz, N.; Bennetts, W. W.; Price, B. E.; Marakowski,
J. J. Med. Chem. 1978, 21, 1044–1047.