Reaction of diphenyldiazomethane with N-methyloxy- and N- ethyloxycarbonyl-N-(2,2,2-trichloroethylidene)amines

Article abstract of DOI:10.1002/(SICI)1099-0690(199907)1999:7<1541::AID-EJOC1541>3.0.CO;2-A

Reaction of the title imines with diphenyldiazomethane gives a Δ³- 1,3,4-triazoline, which leads, after loss of dinitrogen, to a transient azomethine ylide. Subsequent elimination of ethyl or methyl chloroformate gives the unexpected 1,1-diphenyl-4,4-dichloro-2-aza-1,3-butadiene.

Full text of DOI:10.1002/(SICI)1099-0690(199907)1999:7<1541::AID-EJOC1541>3.0.CO;2-A

FULL PAPER
Reaction of Diphenyldiazomethane with N-Methyloxy- and
N-Ethyloxycarbonyl-N-(2,2,2-trichloroethylidene)amines
[a]
[a]
[a]
[a]
´
Sandrine Jacquot, Abdelhak Belaissaoui, Gerard Schmitt,* Bernard Laude,
Marek M. Kubicki,[b] and Olivier Blacque[b]
Keywords: 1,3-Dipolar cycloadditions / Diphenyldiazomethane / N-Alkyloxycarbonyl-N-(2,2,2-trichloroethylidene)-
amines / ∆3-1,3,4-Triazoline / Aziridines / Azomethine ylides / 2-Azadiene
Reaction of the title imines with diphenyldiazomethane gives
a ∆3-1,3,4-triazoline, which leads, after loss of dinitrogen, to
a transient azomethine ylide. Subsequent elimination of
ethyl or methyl chloroformate gives the unexpected 1,1-
diphenyl-4,4-dichloro-2-aza-1,3-butadiene.
commonly found as a secondary product of the expected
triazoline.[1]
Introduction
The known reactivity of the CϭN double bond in vari-
ous azomethines towards diazoalkanes leads to the regiose-
lective formation of ∆2-1,2,3-triazolines.[1Ϫ4] However, some
examples of the formation of ∆3-1,3,4-triazolines have also
been reported.[1][4] It has often been observed that ther-
molysis of the triazolines gives the corresponding aziridines
or other products resulting from the loss of dinitrogen.
We present here the reactions of diphenyldiazomethane 2
with N-ethyloxy- and N-methyloxycarbonyl-N-(2,2,2-trich-
loroethylidene)amines 1a and 1b. The 1,3-dipolarophilic re-
activity of the electron-deficient imine 1a has recently been
assessed using 1,3-dipoles such as nitrones,[5] nitrilimines
and nitriloxides,[6] as well as ethyl diazoacetate.[7] In this
last case, the reaction involves nucleophilic addition of the
ethyl diazoacetate with formation of a new 1,3-dipole.[7]
This addition should not occur with diphenyldiazomethane
since this 1,3-dipole has no mobile proton.
Scheme 1. Reaction of 1a,b with 2 at 65°C
Results and Discussion
Reactions
After stirring a toluene solution of the imine 1a or 1b
with diphenyldiazomethane 2 for 24 h at 65°C, a major
product 3 (70%) was formed, together with a by-product 4
(7%). Elemental analysis, IR, and NMR data of 3 were not
consistent with the structure of the expected triazoline. The
structure of this product was later unambiguously estab-
lished by an X-ray diffraction study (Figure 1). Elemental
analysis and spectroscopic data showed 4 to be an aziridine
[a]
´
´
Laboratoire de Chimie et Electrochimie Moleculaire, Universite
´
de Franche-Comte,
Figure 1. ORTEP plot (50% probability level) of the molecular
structure of 3
16 Route de Gray, La Bouloie, F-25030 Besanc¸on, France
Fax: (internat.) ϩ33 (0)3 81 66 62 30
E-mail: sandrine.jacquot@univ-fcomte.fr
Formally, compound 3 is a secondary product formed by
loss of dinitrogen and alkyl chloroformate from a triazoline
1,3-cycloadduct of 1 with 2. Indeed, a quantitative gas-
[b]
`
`
´
Laboratoire de Synthese et dЈElectrosynthese Organometalli-
´
ques (UMR 5632), Universite de Bourgogne,
6 Boulevard Gabriel, F-21000 Dijon, France
Eur. J. Org. Chem. 1999, 1541Ϫ1544
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ193X/99/0707Ϫ1541 $ 17.50ϩ.50/0
1541

S. Jacquot, A. Belaissaoui, G. Schmitt, B. Laude, M. M. Kubicki, O. Blacque
FULL PAPER
chromatographic analysis of the reaction mixture of 1a with at δ > 7.50 is in good agreement with structure 6, where the
2 led to the detection of >80% of the estimated stoichio- CH group resides between two deshielding nitrogen atoms.
metric amount of ethyl chloroformate.
In the case of structure 6Ј, this signal would be expected to
The 2-azadiene 3 may also be formed from the aziridine appear below δ ϭ 6.00.[8] The signals of the alkyl ester
4 following loss of alkyl chloroformate. According to litera- groups are broadened (see Figure 2 for the case of 6a),
ture data,[1] heating to a temperature >100°C allows cleav- probably due to restricted rotation because of the steric hin-
age of the CϪC bond of an aziridine ring, leading to the drance imposed by the 2,2-diphenyl groups and the 5-CCl3
corresponding azomethine ylide. Indeed, while 4a is ther- moiety. Such hindrance should not be present in the triazol-
1
mally stable at 65°C, refluxing a toluene solution of this inic structure 6Ј. Moreover, H-NMR examination of the
compound results in quantitative conversion to the 2-azadi- mother liquor of 6 did not reveal the expected signal of the
ene 3 and ethyl chloroformate, this process certainly pro- 5-H proton (δ ഠ 5.50Ϫ6.00) of the regioisomer 6Ј.
ceeding via the azomethine ylide 5a (Scheme 2).
We attempted to rationalize the direction of the addition
by carrying out some semi-empirical MO calculations (PM3
level, optimized geometries) on the starting reactants 1b
and 2. The results of these calculations show that the cyclo-
addition occurs by interaction between the LUMO of 1b
and the HOMO of 2. However, it is not possible to justify
the observed direction of addition since the orbital coef-
ficients of the C and N atoms of the LUMO of 1b have
equal weightings.
Study of the Evolution of Cycloadduct 6a
We observed that heating a solution of cycloadduct 6a in
toluene at 65°C furnished 2-aza-1,3-butadiene 3 in high
yield (> 90%). Thus, we tried to follow this evolution in an
NMR tube in CDCl3 at 50°C over a period of 10 h (Fig-
ure 2).
Scheme 2. Evolution of 4c at 110°C
In order to identify some of the transient species leading
to the 2-azadiene 3, which should allow elucidation of the
mechanism of the reaction, we carried out a study at room
temperature. After 30 h at this temperature, a mixture of
the imine 1a or 1b and diphenyldiazomethane 2 in toluene
solution gave only one isolable product (yield: 70%), el-
emental analysis and spectrometric data of which showed it
to be one of the two regioisomeric triazolines 6 or 6Ј formed
by 1,3-dipolar cycloaddition (Scheme 3).
Figure 2. Evolution of the triazoline 6a at 323 K
This study revealed that 6a is not transformed directly to
the final 2-azadiene 3. In fact, a second intermediate 5 is
involved. The H-NMR spectrum of this transient species
Scheme 3. Reaction of 1a,b with 2 at 20°C
1
With the aim of determining the direction of addition 5 features well-resolved signals for the ethyl group of the
of this cycloadduct, we tried to undertake an X-ray crystal ester function. The absence of steric hindrance is probably
structure analysis. However, all attempts to obtain suitable due to a ring-opening of 6a, leading to two possible transi-
single crystals were unsuccessful. Thus, the structure of the ent species 5a or 5Јa. The azomethine ylide structure 5a
1
∆3-1,3,4-triazoline 6 was established on the basis of its H- would be formed from 6a by 1,3-dipolar cycloreversion and
NMR data: The chemical shift of the 5-H proton (Cl3CCH) loss of dinitrogen, while the N-(2,2-dichlorovinyl)iminium
1542
Eur. J. Org. Chem. 1999, 1541Ϫ1544

Reaction of Diphenyldiazomethane with N-Alkoxycarbonyl-N-(2,2,2-trichloroethylidene)amines
FULL PAPER
Scheme 4. Evolution of 6a at 323 K
at 50.3 MHz for 13C. Chemical shifts were measured relative to
TMS in CDCl3 or [D8]THF as solvents. Ϫ Analytical data were
obtained at the Elemental Analysis Center, Dijon. Ϫ Melting
points were determined on an Electrothermal 9200 apparatus and
are uncorrected. Ϫ N-Ethyloxy- and N-methyloxycarbonyl-N-
(2,2,2-trichloroethylidene)amines[10][11] and diphenyldiazome-
thane[12] were synthesized according to literature methods.
salt 5Јa would arise by loss of one chloride ion[9] (Scheme
4).
We favour the azomethine ylide structure 5a for the inter-
1
mediate because in some of the recorded H-NMR spectra
the signals of aziridine 4a appeared after the formation of
the transient species 5. Compound 4a can only be derived
from ring-closure of the azomethine ylide 5a (Scheme 5). It
is obtained at 50Ϫ65°C.
Reactions at 65°C. ؊ General Procedure: A solution of diphenyldia-
zomethane (32 mmol) and N-ethyloxy- or N-methyloxycarbonyl-N-
(2,2,2-trichloroethylidene)amine (32 mmol) in toluene (80 mL) was
heated at 65°C for 24 h. The solvent was then evaporated and the
residue was recrystallized from ethanol.
(a) Synthesis of 4,4-Dichloro-1,1-diphenyl-2-azabuta-1,3-diene (3):
Yield: 70%, pale-yellow needles, m.p. 122°C. Ϫ 1H NMR (CDCl3):
δ ϭ 7.05 (s, 1 H, ethylenic CH), 7.15Ϫ7.80 (m, 10 H, arom. H). Ϫ
13C NMR (CDCl3): δ ϭ 123.4 (s, ϭCCl2), 128.2Ϫ131.0 (5 signals
for 10 arom. CH), 135.2 (d, ethylenic CH), 135.4 (s, arom. C), 138.5
(s, arom. C), 168.3 (s, CϭN). Ϫ C15H11Cl2N (276.18): calcd. C
65.24, H 4.01, N 5.07, Cl 25.68; found C 65.19, H 3.99, N 5.10,
Cl 25.72.
Scheme 5. Ring closure of 5
The structure 5a may be further supported by a transient
orange coloration of the solution in the NMR tube. How-
ever, all attempts to trap this 1,3-dipole 5a by cycloaddition
with N-phenyl- or N-methylmaleimide or with DMAD pro-
ved unsuccessful. This may be attributable to the steric hin-
drance of the three atoms of the dipole 5a rendering this
species unreactive.
X-ray Crystal Structure of 3: The structure was solved by direct
methods using SHELXS-97. Hydrogen atoms were placed in calcu-
lated positions. Refinement was carried out by a full-matrix least-
squares method based on F2 values with 163 variables. Crystallo-
˚
graphic data are collected in Table 1. Selected bond lengths [A] and
angles [°] for 3: C1ϪC2 1.319(3), NϪC2 1.392(3), NϪC3 1.293(2);
C1ϪC2ϪN 121.2(2), C2ϪNϪC3 119.7(2), NϪC3ϪC4 117.5(2),
NϪC3ϪC10 125.2(2), C4ϪC3ϪC10 117.3(2), Cl1ϪC1ϪCl2
114.6(1). For further information see ref.[13]
Conclusion
This study has established the successive steps of the
mechanism of the title reaction. The primary cycloadduct
6a leads, after loss of dinitrogen, to a transient azomethine
ylide 5a. Subsequent elimination of alkyl chloroformate
gives the 1,1-diphenyl-4,4-dichloro-2-aza-1,3-butadiene 3.
To the best of our knowledge, this is the first example of
the formation of a 2-azabutadiene as a secondary product
of an azomethine ylide. Work is currently in progress aimed
at extending this reaction to other disubstituted diazoal-
kanes and other conveniently substituted imines
(b) Synthesis of Aziridines 4a and 4b: The mother liquor from the
above crystallization was concentrated and the newly obtained resi-
due was crystallized from ethanol. Ethyl- and methyl 2,2-diphenyl-
3-(trichloromethyl)aziridine-1-carboxylates 4a and 4b were ob-
tained as pale-yellow needles (each 2.56 mmol, 7%).
Ethyl 2,2-Diphenyl-3-(trichloromethyl)aziridine-1-carboxylate (4a):
Yield: 7%, m.p. 101°C. Ϫ IR (KBr): ν˜ ϭ 1720 cmϪ1 (CϭO). Ϫ 1H
3
NMR (CDCl3): δ ϭ 1.06 (t, J ϭ 7.1 Hz, 3 H, CO2CH2CH3), 4.00
(AB part of ABX3, 3J ϭ 7.1 Hz, 2J ϭ 10.7 Hz, 2 H, CO2CH2CH3),
4.52 (s, 1 H, CHCCl3), 7.20Ϫ7.65 (m, 10 H, arom. H). Ϫ 13C NMR
(CDCl3): δ ϭ 13.85 (q, J ϭ 127 Hz, CO2CH2CH3), 57.9 (d, J ϭ
174 Hz, CHCCl3), 60.5 (s, Cq), 63.0 (t, J ϭ 148 Hz, CO2CH2CH3),
96.2 (s, CCl3), 127.7Ϫ129.0 (6 signals for 10 arom. CH),
135.8Ϫ136.8 (2 signals for 2 arom. Cq), 159.5 (s, CO). Ϫ MS
(70 eV); m/z (%): 383 (3.6) [Mϩ], 348 (73.7) [Mϩ Ϫ Cl], 275 (47.3)
Experimental Section
General: IR spectra (KBr) were recorded on a Bio-Rad FTS-7 spec-
1
trometer. Ϫ H- and 13C-NMR spectra were obtained on a Bruker
Spectrospin AC 200 spectrometer operating at 200 MHz for 1H and
[Mϩ
Ϫ
ClCO2Et], 240 (45.0) [Mϩ
Ϫ
Cl
Ϫ
ClCO2Et].
Ϫ
Eur. J. Org. Chem. 1999, 1541Ϫ1544
1543

S. Jacquot, A. Belaissaoui, G. Schmitt, B. Laude, M. M. Kubicki, O. Blacque
FULL PAPER
(a) Synthesis of Ethyl 2,2-Diphenyl-5-(trichloromethyl)-∆3-1,3,4-tri-
azoline-1-carboxylate (6a): Yield: 70%, white powder, m.p.(dec)
Table 1. Crystallographic data for 3[12]
1
103°C. Ϫ IR (KBr): ν˜ ϭ 1706 cmϪ1 (CϭO). Ϫ H NMR (CDCl3):
Crystal data
empirical formula
molecular mass
crystal system
space group
C15H11Cl2N
276.15
δ ϭ 0.65 (br. s, 3 H, CO2CH2CH3), 3.40Ϫ4.00 (2 br. s, AB part of
ABX3, 2 H, CO2CH2CH3), 6.60Ϫ7.90 (m, 11 H, 10 arom. CH ϩ
CHCCl3). Ϫ 1H NMR ([D8]THF): δ ϭ 0.65 (br. s, 3 H,
CO2CH2CH3), 3.50Ϫ3.90 (2 br. s, AB part of ABX3, 2 H,
CO2CH2CH3), 6.80Ϫ7.90 (m, 10 H, arom. H), 7.60 (s, 1 H,
CHCCl3). Ϫ 13C NMR (CDCl3): δ ϭ 13.2 (q, J ϭ 127 Hz,
CO2CH2CH3), 62.8 (t, J ϭ 148 Hz, CO2CH2CH3), 96.2 (s, CCl3),
109.6 (d, J ϭ 166 Hz, CHCCl3), 115.8 (s, Cq, C-4), 125.0Ϫ131.0 (8
signals for 10 arom. CH), 135.8 (s, arom. C), 137.6 (s, arom. C),
155.5 (s, CO). Ϫ MS (70 eV); m/z (%): 412 (11.4) [Mϩ ϩ H], 348
(86.2) [Mϩ Ϫ Cl Ϫ N2]. Ϫ C18H16Cl3N3O2 (411.71): calcd. C 52.39,
monoclinic
P21/n
cell dimensions:
˚
a [A]
9.874(1)
9.156(1)
15.032(2)
98.03(1)
1345.7(3)
4
˚
b [A]
˚
c [A]
β [°]
3
˚
V [A ]
Z
ρcalcd [g·cmϪ3
F(000)
]
1.363
692
Data collection
H 3.91, N 10.18, Cl 25.77; found C 52.33, H 3.90, N 9.65, Cl 25.49.
diffractometer
Enraf-Nonius CAD4
˚
˚
radiation [A]
λ(Mo-Kα) ϭ 0.71073 A
0.40 ϫ 0.20 ϫ 0.20
graphite
crystal size [mm]
(b) Synthesis of Methyl 2,2-Diphenyl-5-(trichloromethyl)-∆3-1,3,4-
triazoline-1-carboxylate (6b): Yield: 70%, white powder, m.p.(dec)
monochromator
reciprocal lattice segment
0 Յ h Յ 12
Ϫ11 Յ k Յ 0
Ϫ18 Յ l Յ 18
ωϪ2θ scans
2.33Ϫ26.29
4.62
1
109°C. Ϫ IR (KBr): ν˜ ϭ 1710 cmϪ1 (CϭO). Ϫ H NMR (CDCl3):
δ ϭ 3.25 (br. s, 3 H, CO2CH3), 6.70Ϫ7.90 (m, 11 H, 10 arom.
CH ϩ CHCCl3). Ϫ 13C NMR (CDCl3): δ ϭ 53.0 (q, J ϭ 148 Hz,
CO2CH3), 96.0 (s, CCl3), 109.3 (d, J ϭ 166 Hz, CHCCl3), 116.0 (s,
Cq, C-2), 127.3Ϫ129.9 (4 signals for 10 arom. CH), 135.6 (s, arom.
C), 137.2 (s, arom. C), 156.0 (s, CO). Ϫ C17H14Cl3N3O2 (398.68):
calcd. C 51.22, H 3.54, N 10.54, Cl 26.68; found C 51.17, H 3.62,
N 10.32, Cl 26.65.
scan type
θ range [°]
linear abs., µ [cmϪ1
no. of refls. measd.
no. of refls. unique
]
2881
2721
cut-off for obsd. data
no. of unique obsd. data
R(F2)
I > 2σ(I)
2053
0.0406
no. of parameters
163
Rw(F2)
0.0652
[1]
M. Regitz, H. Heydt, in 1,3-Dipolar Cycloaddition Chemistry
G.O.F.
1.059
3
˚
(Ed.: A. Padwa), John Wiley & Sons, Inc., New York, 1984,
ρmax/ρmin [e/A ]
0.36/Ϫ0.47
1, 653Ϫ660.
[2]
´
J. Perrocheau, R. Carrie, Bull. Soc. Chim. Belg. 1993, 102, 749.
[3]
´
J. Perrocheau, R. Carrie, J.-P. Fleury, Can. J. Chem. 1994, 72,
C18H16Cl3NO2 (384.70): calcd. C 56.20, H 4.19, N 3.64, Cl 27.65;
found C 55.98, H 4.22, N 3.50, Cl 27.78.
2458.
[4]
[5]
´
J. Perrocheau, R. Carrie, J. Chem. Res. 1995, (S) 303, (M) 1927.
R. Consonni, P. Dalla Croce, R. Ferraccioli, C. La Rosa, J.
Chem. Res. 1992, (S) 32.
Methyl 2,2-Diphenyl-3-(trichloromethyl)aziridine-1-carboxylate (4b):
Yield: 7%, m.p. 120°C. Ϫ IR (KBr): ν˜ ϭ 1725 cmϪ1 (CϭO). Ϫ 1H
NMR (CDCl3): δ ϭ 3.65 (s, 3 H, CO2CH3), 4.55 (s, 1 H, CHCCl3),
7.30Ϫ7.65 (m, 10 H, arom. H). Ϫ 13C NMR (CDCl3): δ ϭ 53.6 (q,
J ϭ 148 Hz, CO2CH3), 58.2 (d, J ϭ 174 Hz, CHCCl3), 60.6 (s, Cq),
96.1 (s, CCl3), 127.5Ϫ129.5 (6 signals for 10 arom. CH), 135.7 (s,
arom. C), 136.9 (s, arom. C), 160.0 (s, CO). Ϫ C17H14Cl3NO2
(370.67): calcd. C 55.10, H 3.81, N 3.78, Cl 28.69; found C 55.19,
H 3.76, N 3.81, Cl 28.76.
[6]
[7]
A. Belaissaoui, C. Morpain, B. Laude, Bull. Soc. Chim. Belg.
1995, 104, 491.
A. Belaissaoui, S. Jacquot, C. Morpain, G. Schmitt, B. Laude,
Can. J. Chem. 1997, 75, 523.
[8]
[9]
R. Huisgen, G. Szeimies, Chem. Ber. 1965, 98, 1153.
We thank the referee who suggested the alternative structure
5Јa for the transient species.
[10]
[11]
[12]
J. B. Miller, J. Org. Chem. 1959, 24, 560.
O. Diels, C. Seib, Chem. Ber. 1909, 42, 4062.
H. Ulrich, B. Tucker, A. A. R. Sayigh, J. Org. Chem. 1968,
33, 2887.
Reaction at 20°C: A solution of diphenyldiazomethane (10 mmol)
and N-ethyloxy- or N-methyloxycarbonyl-N-(2,2,2-trichloroethyli-
dene)amine (10 mmol) in toluene (15 mL) was stirred at room tem-
perature for about 30 h. The solvent was then evaporated at room
temperature, and the remaining white solid was collected by fil-
tration. Methyl- and ethyl 2,2-diphenyl-5-trichloromethyl-∆3-1,3,4-
triazoline-1-carboxylates 6a and 6b were obtained as white powders
(each 7 mmol, 70%).
[13]
Crystallographic data for the structure 3 reported in this paper
(excluding structure factors) have been deposited at the Cam-
bridge Crystallographic Data Centre as supplementary publi-
cation number CCDC-104494. Copies of the data can be ob-
tained free of charge on application to The Director, CCDC,
12 Union Road, Cambridge CB2 1EZ, U.K. [Fax: (internat.)
ϩ44 (0)1223 336033; E-mail: deposit@ccdc.cam.ac.uk].
Received November 5, 1998
[O98497]
1544
Eur. J. Org. Chem. 1999, 1541Ϫ1544