Heterylation of 3-R1-5-R2-1,2,4-triazoles with derivatives of 3,5-dinitro-1,2,4-triazole

Article abstract of DOI:10.1023/A:1013100716588

Heterylation of 3-R¹-5-R²-1,2,4-triazoles (pK_a 3-12) with N-alkyl-, N-alkenyl-, N-alkoxycarbonyl-, N-oxoalkyl-, N-nitroxyalkyl, and N-nitroaminoalkyl-3,5-dinitro-1,2,4-triazoles results in substitution of a nitro group in 5 position of the dinitro compound yielding 1-R-methyl-3-nitro-5-(3-R¹-5-R ²-1,2,4-triazolyl)-1,2,4-triazoles. The side processes: Hydroxide-ion attack on C⁵ and (or) N¹ of the ring both in the substrate and in the target compound afford 1-R-methyl3-nitro-1,2,4-triazol-5-ones, 3,5-dinitro-1,2,4-triazole and NH-acids of N-C-bitriazole series. Optimal reaction media are aprotic dipolar substances, and for compounds prone to heterolysis ethyl acetate-water systems. The azole pK_a is the decisive factor controlling the composition and the ratio of reaction products. The process is promising for azoles with pK_a > 5, and the optimal range of pK_a is 8-10.

Full text of DOI:10.1023/A:1013100716588

Russian Journal of Organic Chemistry, Vol. 37, No. 8, 2001, pp. 1158 1168. Translated from Zhurnal Organicheskoi Khimii, Vol. 37, No. 8, 2001,
pp. 1217 1227.
Original Russian Text Copyright
2001 by Kofman.
Heterylation of 3-R1-5-R2-1,2,4-Triazoles with Derivatives
of 3,5-Dinitro-1,2,4-Triazole
T. P. Kofman
St. Petersburg State Technological Institute, St. Petersburg, 198013 Russia
Received October 27, 2000
Abstract Heterylation of 3-R1-5-R2-1,2,4-triazoles (pKa 3 12) with N-alkyl-, N-alkenyl-, N-alkoxy-
carbonyl-, N-oxoalkyl-, N-nitroxyalkyl, and N-nitroaminoalkyl-3,5-dinitro-1,2,4-triazoles results in
substitution of a nitro group in 5 position of the dinitro compound yielding 1-R-methyl-3-nitro-5-(3-R1-5-
R2-1,2,4-triazolyl)-1,2,4-triazoles. The side processes: Hydroxide-ion attack on C5 and (or) N1 of the ring
both in the substrate and in the target compound afford 1-R-methyl3-nitro-1,2,4-triazol-5-ones, 3,5-dinitro-
1,2,4-triazole and NH-acids of N C-bitriazole series. Optimal reaction media are aprotic dipolar substances,
and for compounds prone to heterolysis ethyl acetate water systems. The azole pKa is the decisive factor
controlling the composition and the ratio of reaction products. The process is promising for azoles with
pKa > 5, and the optimal range of pKa is 8 10.
In the 1,2,4-triazole series numerous modifications
of structure and syntheses of new derivatives of the
series were carried out starting from N-substituted
3,5-dinitro-1,2,4-triazoles. The most common are
reactions of nucleophilic substitution of the nitro
group in 5 position with various agents [1 6]. In
some of these studies as nucleophilic reagents were
used azoles, mostly 1,2,4-triazoles [2 5]. The NH-
acids of the 1,2,4-triazole series were subjected
mainly to N-arylation, publications on N-heterylation
besides those mentioned above are few in number.
Therewith reactions were frequently carried out with
haloderivatives where the group to be replaced was
activated by an acceptor located either out of the ring
or in the ring of the aromatic [6 10] or hetero-
aromatic system [11 14] respectively. For instance,
the N-arylation of 1,2,4-triazole with highly reactive
2,4-dinitro-1-fluorobenzene readily proceeds even
without bases (DMSO, 20 C) [8]. Besides 1,2,4-tri-
azole [7, 8, 10, 13] in reaction with halobenzenes
were used also its alkyl- [12], amino- [7, 11], and
even nitro derivatives [6, 9], in particular, 3,5-di-
nitro-1,2,4-triazole [6]. But whereas the target
N-arylation of the 3-nitro-1,2,4-triazole (DMSO,
20 C, yield 38% [9]) was fairly successful, the
attempt to picrylate the low-basic 3,5-dinitro-1,2,4-
triazole (acetonitrile, 20 C [6]) did not afford the
expected 1-picryl derivative. The isolated products of
the reaction originated from substitution in 5 position
of the triazole ring (3-nitro-1-picryl-5-chloro-1,2,4-
triazole and 3-nitro-1-picryl-1,2,4-triazol-5-one). It
was suggested that these products arose due to trans-
formations of the target product, 3,5-dinitro-1-picryl-
1,2,4-triazole which formed but was instable under
the reaction conditions. If it were true and substitu-
tion of nitro group in position 5 was preceded by
N-arylation then in [6] was achieved arylation of the
least basic among all 1,2,3-triazolides (pKa of the
conjugate NH-acid was 0.66 [15]). If the C5-sub-
stitution occurred in the reagent before the N-aryla-
tion then arylation proceeded also with anions of
relatively strong NH-acids: 3-nitro-5-chloro-1,2,4-
triazole and 3-nitro-1,2,4-triazol-5-one, pKa 3 4 [15].
The arylation or heterylation of triazoles is carried
out without solvent [11] or in polar aprotic media,
predominantly in DMF [8, 13], DMSO [7 9], or
acetonitrile [6, 8] both in the presence of acids [6, 8,
9, 11, 12, 14] or in their absence [7, 8, 11], at heat-
ing [8, 12] and in the cold [6 9]. In [10] arylation
was performed by Ulmann procedure (pyridine,
CuO). In the media of low polarity (benzene) the
substitution does not occur at all [8], and proton-
donor solvents are of low efficiency, especially with
triazolides of elevated basicity [8, 11].
The N-substitution in the triazole ring was non-
selective only at arylation of the neutral molecule of
1,2,4-triazole proper or its C-alkylated analogs
(fusion of reagents, boiling thereof in ethanol [12]),
and among the arising N4- and N1-isomers the latter
prevailed (80 95%). In arylation (heterylation) of
5-amino-1,2,4-triazole the reaction first occurred at
the amino group [7, 11], and only on adding a base
the substitution was successfully performed at the
1070-4280/01/3708-1158$25.00 2001 MAIK Nauka/Interperiodica

HETERYLATION OF 3-R1-5-R2-1,2,4-TRIAZOLES
1159
heteroatom in the ring [7]. The identification of
products was commonly done on the evidence of
1H NMR spectra, but in most cases, also in [11], the
structure of products was not rigorously proved.
ditions and the structure of the substrates under con-
sideration. The heterylation was carried out with the
salts of sufficiently weak NH-acids, i.e., virtually all
reactions occurred in the media at pH over 7. Under
alkaline conditions the derivatives of 3,5-dinitro-
1,2,4-triazoles can react with the hydroxide anion
giving rise to the corresponding triazolones due to
the substitution of a nitro group. This reaction was
observed formerly as a side process with the majority
of the substrates of this group in their reactions with
any nucleophilic reagents [1 5], in particular, with
3-R1-5-R2-1,2,4-triazolides [2 5]. Besides both the
original and target compounds in the alkaline medium
sometimes suffer heterolysis at the bond connecting
the triazole ring with the functionalized N-alkyl
substituent [2 5], and in a number of compounds the
reactions at the functional groups also cannot be
excluded. All these processes significantly affect the
synthetic results; moreover, in some cases one or
another side process can dominate or be the only
direction of the reaction.
The most complete discussion on heterylation of
3-R1-5-R2-1,2,4-triazoles was done in publications
concerning replacement of a nitro group or a halogen
atom in the derivatives of 3,5-dinitro-1,2,4-triazole
[2 5] and in methylated 3- or 5-halo-1,2,4-triazoles
[12].
Among substrates brought to the fore are com-
pounds with the functional groups in the N-alkyl
fragment. This leads to complications during heteryl-
ation process but affords new reactive compounds.
And in this respect the interest arises not only
from N C-bitriazoles proper but also due to the
opportunity to estimate the effect of a substituent in
the substrate on the direction and selectivity of the
process. These data provide a possibility to reveal the
general rules of the reaction, the limits of its applic-
ation, and to forecast the results of a synthesis
depending on the structural characteristics of both
participants.
We established that under the conditions used in
this study (90% aqueous acetone, 60 C) the reaction
with the majority of the objects proceeded in the
target direction: Triazole heterylation occurred with
nitro group replacement in the position 5 of the
substrate, but the formation of the target bitriazoles
XXIII XLIII was accompanied with side processes
(the yield of N C-bitriazole at the total conversion of
substrate was always below 80%) (Tables 1 3). The
side reactions were due to the attack of hydroxide
anion on positions C5 and N1 in the ring. The first
reaction resulted in triazolones L LVII and was
observed practically always. A rare exception present
the cases where occurs only heterolysis by the type of
reverse Michael reaction: interaction of 3,5-dinitro-1-
(3-oxobutyl)-1,2,4-triazole (XV) with 5-amino-1,2,4-
triazole (I) (Table 2), or of 3,5-dinitro-1-(2-nitro-
ethyl)-1,2,4-triazole (XXII) and nitroamine XIX with
3-nitro-1,2,4-triazole (VI) in acetone (Table 3).
In the present study we used in heterylation mono-
(I VI) and disubstituted at the carbon atoms 1,2,4-
triazoles (VII XIII) with C-substituents of various
types. The compounds under study were characteriz-
ed by a wide range of pKa (6 11 and 3 12 respective-
ly [5, 15], As heterylating agents were applied
N-alkyl-, N-alkenyl-, N-alkoxycarbonyl-, N-oxo-
alkyl-, N-nitroxyalkyl, and N-nitroaminoalkylderiv-
atives of 3,5-dinitro-1,2,4-triazole (XIV XXVII)
[4, 16].
In some cases were introduced variations in the
reaction temperature, reagents concentration, their
ratio, and in the base type. As a rule the synthesis
was carried out at heating in an aprotic polar solvent,
most often in acetone with addition of 10% of water
to increase the solubility (Tables 1 3). The reagents
were used in equimolar amounts that had been
established before to be feasible [5]. The reaction
rate, completeness of substrate conversion, the
composition and ratio of the reaction products were
The best suitable for investigation of structure
(basicity) effect on the ratio of the main and side
reactions occurring at position C5 in the series of
NH-acids under consideration is 1-methyl-3,5-dinitro-
1,2,4-triazole (XIV) (set I, Table 1). The reactions
of the same reagents with a substrate that certainly
suffers N1 C-heterolysis, 3,5-dinitro-1-(2-oxobutyl)-
1,2,4-triazole (XV), are collected in set II (Table 2).
1
monitored by TLC and H NMR spectroscopy. The
products obtained as a rule were identical to those
prepared and described before [2 5, 17]; the struc-
ture of substances obtained for the first time was
confirmed by analytical data and spectral character-
istics that are given in Experimental.
The comparison of results permits evaluation of
the effect of azole pKa on the second side reaction,
N1-devinylation. Finally, the contribution of the
proper structure of the substrate into the overall
It should be specially noted that non-uniqueness of
the reaction direction originated both from its con-
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

1160
KOFMAN
Table 1. Process conditions and products yield in reaction of 1-methyl-3,5-dinitro-1.2.4-triazole (XIV) with triazoles
I IV, VI XIII [(XIV)/(triazole reagent)/NaOH = 1: 1: 1, 90% aqueous acetone]
Reagent
Products of C5-substitution
Temperature,
C
Time,
h
Compd.
no.
Compd.
no.
Bitriazole Triazolone L
Overall
yield, %
1
pKa
c, mol l
yield, %
yield, %
I
11.08
10.08
0.23
0.60
0.23
0.60
0.23
0.23
0.60
0.23
0.44
0.60
0.23
0.60
0.60
0.23
0.60
0.23
0.60
0.23
0.60
0.60
0.23
0.23
0.23
0.23
60
20
60
20
60
60
20
60
60
20
60
20
20
60
20
60
20
60
20
20
75b
60
60
60
0.5
24
0.5
24
1
0.5
24
7
XXIII
XXIV
0
70
53
83
79
75
36
60
62
60
0
55
65
82
54
85
40
50
23
46
47
42
18
0
0
28
11
17
19
22
10
30
30
30
0
35
35
10
8
11
7
45
32
40
46
52
66
60
Tarring
98
II
64a
100
98
III
IV
8.64
8.11
XXV
XXVI
97
46
90
92
VI
6.05
XXVII
4
144
0.5
3
24
1
24
1
24
3
24
72
8
3
24
24
90
VII
12.12
XXVIII
Tarring
90
90
92
62
96
47
95
55
86
93
94
84
60
VIII
IX
9.60
9.11
7.05
XXIX
XXX
X
XXXI
XI
XII
XIII
6.75
5.22
3.05
XXXII
XXXIII
a
b
All solvents contained 10 vol% of water. Tarring. Solvent EtOAc.
Table 2. Process conditions and products yield in reaction of 1-(3-oxobutyl)-3,5-dinitro-1,2,4-triazole (XV) with triazoles
1
I XII [(XV)/(triazole reagent)/NaOH = 1: 1: 1, c 0.44 mol l , conversion of substrate XV 100%]
Products of C5-substitution
Reagent
no.
Temperature,
C
Share of
heterolysis, %
Solvent
bitriazole
no.
yield of
overall
yield, %
yield LI, %
bitriazole, %
I
II
Acetone
Acetone
EtOAc
Acetone
EtOAc
20
60
80
60
80
60
80
60
60
20
60
80
80
80
0
21
69
65
68
79
63
68
39
40
35
57
67
25
Traces
45
28
30
31
22
35
28
20
20
20
20
31
Tarring
XXXIV
XXXV
XXXVI
69
97
95
99
92
98
96
59
60
55
77
98
95
34
3
5
1
8
III
IV
Acetone
EtOAc
2
4
Va
VI
Acetone
Acetone
Acetone
Acetone
CH3CN
EtOAc
XXXVII
XXXVIII
41b
40c
45d
23
2
Ethanol
70
5
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

HETERYLATION OF 3-R1-5-R2-1,2,4-TRIAZOLES
Products of C5-substitution
1161
Table 2. (Contd.)
Reagent
Temperature,
Share of
heterolysis, %
Solvent
no.
C
bitriazole
no.
yield of
bitriazole, %
overall
yield, %
yield LI, %
VII
Acetone
Acetone
Acetone
CH3CN
Acetone
CH3CN
Acetone
CH3CN
Acetone
Acetone
20
60
60
80
60
80
60
80
60
60
XXXIX
XL
25
0
3
0
28
0
62e
Tarring
25
VIII
IX
40
50
58
77
35
52
34
0
35
30
37
20
30
28
30
20
75
80
95
97
65
80
64
20
20
5
3
XLI
Œ
XLII
XLIII
35
20e
35
XI
XII
80
All solvents contained 10 vol% of water.
a
pKa 8.11.
Time 6 h.
Time 120 h, c 0.7 mol l .
Triethylamine base.
Conversion 90%, 24 h.
c 0.5 mol l .
b
c
d
e
f
1
1
Table 3. Process conditions and products yield in reaction of 3-nitro-1,2,4-triazole (VI) with derivatives of
1
3,5-dinitro-1,2,4-triazole XIV XXII [(XIV XXII)/(VI)/NaOH = 1: 1: 1, 60 C, c 0.23 mol l ]
Substrate
Products of C5-substitution
Triazolone
Time
of reaction,
h
N C-Bitriazole
Conver-
sion, %
Compd.
Overall
yield, %
Rf
no.
Compd.
no.
Yield,
%
Compd.
no.
Yield,
%
Rf
Rf
XIV
XV
0.58 XXVII
0.56 XXXVIII
0.48
0.46
60a
42a
69b
14a
30b
47a
55b
40a
0a
L
LI
0.28
0.20
30
20
30
54
15
34
41
40
50
28
30
30
0
90
62
99
66
45
81
96
80
50
98
90
80
0
95
100c
100c
93
7
6
12
3
6
8
12
8
2
4
7
XVId
XVII
0.62 XLIV
0.65 XLV
0.51
0.40
LII
0.32
0.25
90
LIII
100c
100
90
XVIII
XIX
0.67 XLVI
0.51 XLVII
0.43
0.39
LIV
LV
0.28
0.25
100c
100c
95
70b
60a
50a
0a
XX
XXI
XXII
0.51 XLVIII
0.60 XLIX
0.28
0.41
0.36
LVI
LVII
0.25
0.15
90
6
0.5
100e
a
b
c
d
e
90% aqueous acetone.
EtOAc: 80 C for compounds XV XVII, 20 C for compound XIX.
Partial heterolysis.
Yield of product LIX 27 and 45% in acetone and EtOAc respectively.
Yield of 1-(2-nitroethyl)-3-nitro-1,2,4-triazole (LVIII) 95%.
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

1162
KOFMAN
Scheme 1.
triazole (I) was definitely observed nonselective
heterylation, and the N1- and N2-isomers XXXIIIa, b
[5] were isolated and identified (Scheme 2).
Neither aminoazole I nor any other 3-R-5-amino-
1,2,4-triazole underwent heterylation at the amino
group: The structure of the heterylation products was
proved not only by spectral data but also by oxidation
of the N C bitriazole amino compounds into the cor-
responding azo derivatives [5]). The side product of
heterylation triazolone L originates exclusively from
the substrate. The target products, the corresponding
N C-bitriazoles, are not converted into triazolone L
under the synthesis conditions: Bitriazole XXVII
after heating for 6 h in 0.1 N water acetone solution
of NaOH was recovered quantitatively.
R2 = H: R1 = H (II, XXIV), N3 (III, XXV), Cl (IV,
XXVI), NO2 (V, XXVII); R2 = NH2: R1 = NH2 (VII,
XXXII), N3 (VIII, XXIX), Cl (IX, XXX), NO2 (X,
XXXI); R1 = NO2, R2 = Me (XI, XXXII), R1 = R2 =
Thus in the process under study occur two in-
dependent reactions: the C5 position of the substrate
is attacked by azolide and hydroxide anions. Con-
sequently the yields of the main and side products
should depend on the reactivity of these nucleophilic
agents and on their concentration. Since the con-
centration of the hydroxide ion under conditions of
the actual synthesis is controlled by the pKa of the
azole, i.e., pH of the medium changes with replace-
ment of a reagent, it is presumable that the final result
depends just on the acid-base characteristics of the
triazole.
Cl (XII, XXXIII), R2
= Br (XIII).
Scheme 2.
The rate of C5-substitution of the nitro group in
substrate XIV, runs set I, and the ratio of the main
and side products vary in a fairly wide range and
actually depend on the pKa of the reagent (Table 1).
The relationships are similar both for mono- (I VI)
and disubstituted (VII XII) triazoles, i.e., the steric
factors do not significantly affect either the selectivity
of heterylation or the ratio of the main and side
product.
process may be evaluated for example on the reaction
of 3-nitro-1,2,4-triazole (VI) with triazoles func-
tionally-substituted in the N-alkyl fragment
XIV XXII (set III, Table 3).
The rate of substrate conversion in general grows
with increasing basicity of triazolide. This can be
controlled by changing temperature and reagents
concentration. Therewith the temperature factor is
more important than the concentration one, and with
reagents of various basicity it is recommended to use
these parameters in different combinations depending
on pKa of the azole. Thus with triazoles with pKa
larger than 10 the heating of reagents in acetone
provides tarring, and to obtain the desired products
the synthesis temperature should be reduced to 18
20 C; to intensify the process with this group objects
a higher concentration of reagents (within reasonable
limits) is required, and the completeness of con-
version is achieved at longer reaction time. With the
It was shown that in heterylation 3,5-1-methyl-
1,2,4-triazole (XIV) is transformed according to the
target direction when reacting with the most of tri-
azoles I XII under study, and only with relatively
strong NH-acids (pKa 3 and less), e.g., with 5-bromo-
3-nitro-1,2,4-triazole (XIII), the only reaction pro-
duct is 5-bromo-1-methyl-3-nitro-1,2,4-triazol-5-one
(L). With reagents of pKa > 5 the triazolones were
obtained alongside the corresponding N C-bitriazoles
XXIII XXXIII (Scheme 1).
The heterylation occurred virtually selective at the
azole heteroatom N1, and only with 5-amino-1,2,4-
reagents of pKa in the range 9
5 both the tempera-
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

HETERYLATION OF 3-R1-5-R2-1,2,4-TRIAZOLES
1163
Scheme 3.
R2 = H: R1 = H (II, XXXIV), N3 (III, XXXV), Cl (IV, XXXVI), Br (V, XXXVII), NO2 (VI, XXXVIII); R2 =
NH2: R1 = NH2 (VII, XXXIX), N3 (VIII, XL), Cl (IX, XLI), NO2 (X, XLII); R1 = NO2, R2 = Me (XI, XLIII).
ture and concentration should be increased; therewith
for the azoles of low basicity (pKa below 6) the high
conversion is attained only at a long reaction time.
triazole XIV are conserved but appear some specific
features of the process. Alongside the formation of
bitriazoles XXXIV XLIII and the attendant tri-
azolone LI the second type of side process is evident:
attack of hydroxide ion on the N1 position and hetero-
lysis at the bond ring N-substituent both in the sub-
strate and the target product (a version of reverse
Michael reaction). As a result arise the salts of the
corresponding NH-acids XVa, XXXIVa XXXVIIa
(Scheme 3).
Note that at the corresponding choice of reaction
conditions the substrate conversion is fairly high, and
with reagents of pKa > 6 it is close to quantitative
(overall yield of C5-substitution products is over
95%). However the ratio of the target bitriazole and
the side product, triazolone, is considerably different
for reagents of unlike basicity. At the same time for
a given triazole the ratio of products of C5-substitu-
tion remains virtually unchanged at varying reagents
concentration, temperature, or replacing solvent:
ethyl acetate instead of acetone. In the latter case the
reaction occurs at the phase boundary, and with a
considerably lower rate.
We reported elsewhere [3] on the synthetic
opportunities provided by the reaction of butanone
XV with azoles, primarily the development of pre-
paration method for NH-acids of N C-biazole series
of XXXIVa XXXVIIa type based on heterylation
dealkylation. The substances obtained in this series
were described in [3, 5].
The fraction of the N C-bitriazoles in the reaction
products is over 65 70% only for the reagents with
pKa > 8. With decreasing pKa the yield of the target
product is reduced and accordingly grows the amount
of triazolone L. The latter is prevailing in heteryla-
tion of azoles of low basicity (pKa around 5), more-
over, it has been already mentioned that in the case
of 5-bromo-3-nitro-1,2-4-triazole triazolone L is the
only product of the substrate conversion. Unexpected-
ly with compounds of pKa over 10 the fraction of
triazolone L in the reaction products was greater than
with the reagents of pKa 8 10. This fact is apparently
due to the high alkalinity of the medium. Therefore it
should be avoided by heterylation with no base but
under more rigid conditions.
It should be noted that in general the synthesis of
the N C-bitriazole is successful only when dealkyla-
tion of the substrate and especially of the reaction
product is minimal.
Note that in all runs disregarding the synthesis
conditions and the reagent characteristics in the reac-
tion products were detected compounds originating
from decomposition of the substrate, butanone XV
(in the reaction mixture were present a sodium salt of
3,5-dinitro-1,2,4-triazole and methyl vinyl ketone,
and the yield of C5-substitution products was con-
siderably less than quantitative). The heterolysis of
the target N C-bitriazoles to the corresponding
NH-acids in its turn became detectable only with the
reagents of high basicity (at pKa of the conjugate
NH-acid 9 and more), and only at prolonged reaction
time. This fact is a practical confirmation of the
In a more complicated case, at heterylation of the
same triazole series I XII with butanone XV (runs
set II, Table 2) the general trends observed with
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

1164
KOFMAN
higher thermodynamic stability of the target products
as compared to the original compounds, and it is
reasonable if the pKa of the NH-acids in N C-bitri-
azole series (1 2 [3]) and that of 3,5-dinitro-1,2,4-
triazole ( 0.66 [15]) are considered.
Scheme 4.
To prevent the heterolysis of the reaction product
is especially important when it contains an amino
group (compounds XXXI XLI) since the eliminated
methyl vinyl ketone would attack the amino group
providing addition products (data of 1H NMR spectro-
scopy). The separation of the latter products is not
easy. But at careful monitoring of the substrate con-
sumption (TLC) and at timely workup of the reaction
mixture the bicyclic NH-acids form in the mentioned
reactions only in trace amounts.
To avoid the heterolysis it is sometimes feasible to
carry out the synthesis not in acetone solution but in a
two-phase system (ethyl acetate water). However this
procedure may be recommended, firstly, with the
highly reactive reagents (pKa > 8) for the reaction
rate at the phase boundary is notably slower, and,
secondly, the reaction products must be soluble in
ethyl acetate (this is not the case with amino com-
pounds).
R = Ac (XVI, XLIV, LII), CH2COOMe (XVII,
XLV,
N(NO2)Me (XIX, XLVII, LV), CH(ONO2)Me (XX,
XLVIII, LVI), Vin (XXI, XLIX, LVII).
LIII), COOMe (XVIII, XLVI, LIV),
The above stated shows that in the case in question
it is just the heterolysis that requires especially care-
ful adjustment of reaction conditions for azoles
heterylation. Therefore just in the runs of this set we
tried besides the variation of reaction temperature and
reagents concentration to introduce changes into the
reagents ratio, type of base, and solvent. Apart
acetone we applied acetonitrile, ethyl acetate, and
ethanol (the derivatives of 3,5-dinitro-1,2,4-triazole
are unstable in DMF and DMSO). The workup of the
reaction mixtures was performed either at complete
conversion of the substrate (TLC monitoring) or after
keeping for 24 h (with reagents of low activity). At
the use of a small excess of triazole salt (10%) even
with the reagent with the optimal pKa value (triazole
IX, pKa 9.11) we observed only intensified side
reactions, in particular, triazolone LI formation, and
therewith very sharply changed the ratio of the C5-
substitution products (XLI/LI) 3.85: 1 and 1.6 : 1
respectively]. This fact obviously is due to increased
pH of the reaction medium. Similar results were
obtained also at excess substrate: 10% excess sub-
strate naturally increased only the triazolone fraction.
Neither the use of organic base instead of alkali
changed the situation, although here we were able to
apply anhydrous solvents.
Although in the set II for butanone XV the overall
yield of C5-substitution products is notably lower in
acetone than in the set I with analogous reagents, the
ratio of the corresponding N C-bitriazole and tri-
azolone remains practically at the same level
(Tables 1, 2). The lower yield in the set II is due to
the N1-heterolysis. However the above data show that
the type of N1-substituent in these substrates virtually
does not affect the ratio of C5-substitution products.
Yet the range of application of butanone XV to
heterylation is considerably limited with respect to
the reagent structure (pKa). For instance, with a
group of highly basic azoles I, II, VII, pKa > 10,
under standard conditions prevailed the decomposi-
tion of the substrate, and the yield of bitriazoles
XXIV, XXXIX did not exceed 25%. The heterylation
product of 4-amino-1,2,4-triazole (I) was not obtained
at all due to virtually quantitative heterolysis of the
original butanone and to tarring from polymerization
of the arising methyl vinyl ketone. In its turn ketone
XXXIX was successfully isolated apparently because
of its low solubility: It separated into heterophase.
On the other hand, among the reagents of low basicity
(pKa < 6) we failed to obtain the product of heteryl-
ation with butanone XV of triazole XII, pKa 5.22.
Yet with substrate XIV triazole XII afforded the
corresponding bitriazole XXXVIII, although in a low
yield.
The problem was treated more efficiently by
choosing solvent [in this respect in Table 2 are
presented mainly the results obtained with 3-nitro-
1,2,4-triazole (VI) characterized by low activity, pKa
6.05]. We achieved its heterylation by going over
from acetone to ethyl acetate and acetonitrile. In the
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

HETERYLATION OF 3-R1-5-R2-1,2,4-TRIAZOLES
latter grew both the fraction of C5-substitution pro-
1165
Scheme 5.
ducts and bitriazole yield, but the substrate decom-
position still exceeded 20%. The heterolysis was
virtually fully suppressed in ethyl acetate and ethanol
but if in the former considerably enhanced the bi-
triazole fraction (XXXVIII/LI 2.16: 1 and 2: 1 for
ethyl acetate and acetone respectively) then in the
latter grew the triazolone fraction (XXXVIII/LI 1: 3).
It is presumable that in ethyl acetate the decomposi-
tion of substrate is prevented by its presence in the
organic phase during the reaction on the phase
boundary (ethyl acetate/aqueous alkali). This mode
of performing the process is feasible, especially with
the compounds from azole series having pKa over 10
and less than 8, although it requires longer time.
The main drawback of this group of compounds,
their tendency to heterolysis, significantly affects the
yield of heterylation products. At the same time just
this property, the possibility of further cleavage of a
labile group at nitrogen atom after substituting a nitro
group in such derivatives of 3,5-dinitro-1,2,4-tri-
azole, opens a way to new NH-acids of triazole
series, in particular, to N C-bitriazoles. This
opportunity was utilized in [3, 4]. To this end can be
applied both butanones and propionates, but two
points should be taken into consideration, the con-
servation of the substrate and the product under the
preparation conditions, and the ratio between the
target C5-substitution product and the attendant tri-
azolone. The data of Table 3 evidence that although
the esters of XVII type are more thermodynamically
stable, the use of ketone is more favorable: heteryl-
substitution in ester XVII is the prevailing direction
but the fraction of triazolone LIII in its substitution
products is 1.5 times greater. And since the applica-
tion of ethyl acetate practically suppresses the
heterolysis for both substrates, the problem is solved
by the use of two-phase system (ethyl acetate aqueous
alkali).
The relations disclosed in the study of sets I and II
were further used in heterylation of 3-nitro-1,2,4-tri-
azole (VI) with various substrates XVI XXII, set III
(Scheme 4).
Compounds XVIII, XIX XXI react analogously
to methyl derivative XIV, with similar rates and, save
ester XVIII, give nearly two times more bitriazoles
than triazolones (in the case of compound XVIII the
substitution products arise in equivalent amounts,
Table 3). The triazolones were mostly identified by
comparison with compounds described in [2, 18];
compound LV was prepared for the first time by
procedure [18].
With ester XVII, an analog of butanone XV, the
results were predictable and similar to those observed
with compound XV, the same side reactions: The
attack of hydroxide anion on positions 1 and 5 of the
substrate leading to N-heterolysis and to nitro group
replacement with triazolone formation. However in
acetone ester XVII decomposed twice slower than
ketone XV.
It turned out that heterolysis is characteristic not
only for the products of 3,5-dinitro-1,2,4-triazole
condensation with activated olefins XV, XVII, XXII,
but also for nitroamine XIX (Table 3). Compound
XIX alongside the main reaction underwent both side
processes. Therewith at more stringent conditions in
the substrate prevails heterolysis at the bond ring
N-substituent, and under the conditions acceptable
for the majority of the other substrates (acetone,
60 C) this compound totally decomposes to afford
sodium salt of 3,5-dinitro-1,2,4-triazole and tri-
azolone LV. Still we succeeded in performing
heterylation under very mild conditions (ethyl
acetate water, 20 C). The heterolysis of the substrate
was practically suppressed, and the triazolone yield
notably decreased. It seems that nitroamine XIX is
characterized by abnormally high reactivity since it
rapidly reacts with 3-nitro-1,2,4-triazole at the phase
boundary at room temperature.
At the same time 3,5-dinitro-1-(2-nitroethyl)-
1,2,4-triazole (XXII) as already mentioned under
similar conditions (acetone) underwent only hetero-
lysis: at the complete conversion of the original
compound that occurred very fast the products of
C5-substitution were not even detected, and the
eliminated highly reactive nitroethylene in situ
effected nitroethylation of the reagent to 3-nitro-1-
(2-nitroethyl)-1,2,4-triazole (LVIII) [2] in virtually
quantitative yield (Scheme 5).
Thus the thermodynamic stability of substrates, the
products of 3,5-dinitro-1,2,4-triazole addition to
compounds with the activated double bond, under
heterylation conditions decreased in the series XVII,
XV, XXII.
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

1166
KOFMAN
Thus a fragment CH2NNO2 introduces into
Alongside N C-bitriazole XLIV and triazolone LII
forms a product isomeric to the target ketone XLIV,
5-methyl-5-(3-nitro-1,2,4-triazol-1-yl)-5,6-dihyro-
oxazolo[3,2-b]-1,2,4-triazole (LIX) (Scheme 6).
position 1 of the triazole ring sharply increases the
lability of the nitro group in 5 position. This fact may
be understood considering the specific structure of
nitroaminomethyl group that firstly can act as
acceptor with respect to the ring, and secondly can
be conjugated with the N2-aza group of the ring
(along the chain :N CH2 N1 N2=). The latter inter-
action should change the general conjugation in the
ring and thus weaken the bond of C5 with the hetero-
atom of the leaving group. It is appropriate also to
mention here another uncommon fact that evidences
the specific effect of the nitroaminomethyl group as a
substituent in the N-alkyl chain of triazole on the
activity of the C5-substituent. It turned out that
similarly and also readily occurred the substitution by
3-nitro-1,2,4-triazole of a halogen atom in the analog
of nitroamine XIX, 5-bromo-3-nitro-1-(2-nitro-2-
azapropyl)-1,2,4-triazole, whereas failed all the other
attempts to use in this reaction alternative 1-R-5-
bromo-3-nitro-1,2,4-triazoles (120 180 h, 60 C,
acetone, trace amounts of N C bitriazoles XXVII,
XLVI, XLVIII).
Scheme 6.
Under standard conditions in this case formation of
side products dominated (LII/XLIV/LIX = 3: 1: 2).
The use of ethyl acetate instead of acetone sharply
reduced the triazolone amount and increased the yield
of azolyl derivatives, but the ratio of the latter
changed insignificantly (LII/XLIV/LIX = 0.5 : 1:
1.5]. Oxazolotriazole LIX arose apparently by the
parallel attack of triazolate anion on the carbon atom
of the carbonyl group in the side chain of the sub-
strate followed by its joining to the ring through
elimination of a nitro group in the intermediate. The
contribution of this reaction is greater than that of
the N-heterylation of 3-nitro-1,2,4-triazole. Note that
a similar transformation was observed once again
only with 5-methyl-3-nitro-1,2,4-triazole (XI) [17].
It is generally known that halogen atom in 1,2,4-
triazoles is of low lability: Its abnormally low
reactivity toward nucleophilic substitution in the
series of derivatives of 5-bromo-3-nitro-1,2,4-triazole
is proved by kinetic studies [19]. To N-methylated
3(5)-halo-1,2,4-triazoles in reaction with triazolides
were applied drastically rigid conditions (DMF,
150 220 C, up to 48 h) [13]. This inertness of the
substrates may be due to the specificity of the halogen
conjugation with the ring, and the most probable
reason of this anomaly is the already observed in the
series of nitrogen-containing heteroaromatic com-
pounds deactivation of the halogen atom by its direct
polar conjugation with the aza group in the ap-
propriate position in the ring [20] (in the 1,2,4-tri-
azole it is the aza group in 2 position: conjugation
along the chain C5 N4 C3 N2). Till recently there
were no exceptions from this rule in the 1,2,4-triazole
series. The first abnormally active halogen atom was
found in the 5-bromo-1-vinyl-3-nitro-1,2,4-triazole
[21], and now in 5-bromo-3-nitro-1-(2-nitro-2-aza-
propyl)-1,2,4-triazole. In both compounds in the
N1-substituent is present a group with +R-donor
properties. Thus the nitroamino group activates the
position 5 in triazole both for 5-nitro and 5-halo
derivatives by virtue of participation of the lone
electron pair of the nitroamine nitrogen in the general
conjugation.
Alkoxycarbonyl and nitroxy groups in esters
XVII, XVIII, XX, and also the double bond in allyl
derivative XXI do not suffer transformations in the
course of heterylation, and the only side process with
these substrates is triazolone formation (Table 3).
Regretfully, the number of samples in the set III is
not sufficient for final conclusions on the influence
produced by substituent in substrate on the heteryla-
tion process, but in general when the N1-substituent
is insensitive to the bases its type insignificantly
affects the products composition: the reaction affords
only C5-substitution products with prevalence as a
rule of the N-substituted N C-bitriazole. A trend is
observed to acceleration of the reaction rate and
increase in the triazolone fraction with growing
acceptor character of the N-alkyl substituent in the
substrate, and nitroamine XIX holds a special posi-
In reaction of 3-nitro-1,2,4-triazole with pro-
panone XVI occurs quite unusual side process:
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

HETERYLATION OF 3-R1-5-R2-1,2,4-TRIAZOLES
1167
tion. Summing up all the above stated we can assume
that heterylation of azoles and 1,2,4-triazoles in
particular is the easiest and sufficiently general
synthetic method for N C-biazoles, also those with a
functional group in a side chain.
was extracted into ethyl acetate, the organic layer was
separated, and the water layer was evaporated till
dryness. The residue was treated with 2% sulfuric
acid, the reaction product was extracted into ethyl
acetate. On evaporating the solvent the residue was
recrystallized. All the compounds isolated were
identical to those described in the literature: N C-
bitriazoles XXIII XLVI, XLIX [2 5], 1-R-methyl-3-
nitro-1,2,4-triazole-5-ones L LIV, LVI, LVII [2,
18], 3-nitro-1- (2-nitroethyl)-1,2,4-triazole LVIII [3],
5-methyl5-(3-nitro-1,2,4-triazol-1-yl)-5,6-dihydro-
oxazolo[3,2-b]-1,2,4-triazole LIX [17].
EXPERIMENTAL
1H NMR spectra were registered on spectrometer
Perkin-Elmer R-12 (60 MHz) in DMSO-d6, internal
reference HMDS. IR spectra were recorded on
Specord 75IR instrument from films. Molecular
weights were determined by the reversed ebullioscopy
in acetone. TLC was performed on Silufol UV-254
plates, solvents system hexane acetone ethyl acetate,
5: 2: 1; Rf 0.58 (XIV), 0.56 (XV).
1-R-Methyl-3-nitro-5-(3-R1-5-R2-1,2,4-triazol-1-
yl)-1,2,4-triazoles XXIII XLIX. To a solution of
0.02 0.03 mol of 3-R1-5-R2-1,2,4-triazole I XIII in
5 ml of water containing 0.02 0.93 mol of sodium
hydroxide (triethylamine) at 18 20 C while stirring
was poured a solution of 0.025 mol of N-substituted
3,5-dinitro-1,2,4-triazole XIV XXII [2, 4, 15] in
45 ml of acetone (acetonitrile, ethyl acetate, ethanol).
The reaction mixture was kept at 20 C or in the
boiling solvent till 90 100% conversion of the sub-
strate (TLC monitoring). The solvent was evaporated,
the residue was weighed and analyzed by TLC and
1H NMR spectroscopy. Then 20 ml of water was
added to the residue, the mixture was stirred for
20 min, the insoluble residue (bitriazole) was filtered
off, washed with water on filter (2 10 ml), dried at
50 C, weighed, and analyzed (TLC, 1H NMR
spectroscopy; the target compound may contain the
initial substrate as impurity). N C-bitriazoles XXIII
XLIX was purified by recrystallization.
3-Nitro-1-(2-nitro-2-azapropyl)-5-(3-nitro-1,2,4-
triazol-1-yl)-1,2,4-triazole (XLVII). mp 197 198 C
1
(ethanol). IR spectrum ( , cm ): 840 s, 850 w, 865
w, 990 s, 1020 w, 1050 w, 1080 w, 1180 m, 1220 w,
1280 s, 1305 s, 1340 v.s, 1380 w, 1430 s, 1450 m,
1
1500 m, 1525 m, 1560 1580 v.s. H NMR spectrum
( , ppm): 9.90 s (1H, C5H), 6.65 s (2H, CH2), 3.69 s
(3H, CH3). Found, %: C 23.3; H 2.1; N 44. M 320.
C6H6N10O6. Calculated, %: C 22.94; H 1.93; N
44.58. M 314.17.
3-Nitro-1-(2-nitroxypropyl)-5-(3-nitro-1,2,4-tri-
azol-1-yl)-1,2,4-triazole (XLVIII). mp 124 125 C
1
(chloroform). IR spectrum ( , cm ): 840 v.s, 890 m,
940 w, 1000 m, 1060 w, 1140 w, 1180 w, 1250 w,
1280 s, 1315 s, 1350 w, 1390 w, 1440 s, 1460 w,
1
1515 s, 1570 v.s, 1585 s, 1650 v.s. H NMR spec-
trum ( , ppm): 9.50 s (1H, C5H), 5.80 d (2H, CH2),
5.05 m (1H, CH), 1.55 s (3H, CH3). Found, %: C
25.8; H 2.1; N 38.5. M 323. C7H7N9O7. Calculated,
%: C 25.5; H 2.1; N 38.3. M 329.18.
3-Nitro-1-(2-nitro-2-azapropyl)-1,2,4-triazole-5-
one (LV). mp 181 182 C (ethanol). IR spectrum ( ,
1
cm ): 805 m, 870 s, 940 s, 1020 s, 1070 m, 1100 m,
1140 m, 1240 w, 1260 m, 1280 s, 1340 m, 1360 m,
1380 m, 1420 w, 1470 m, 1550 s, 1750 v.s. H NMR
The combined filtrate and washings were acidified
by 10% sulfuric acid till pH 1, extracted with ethyl
acetate (3 25 ml), the extract was washed with water
and dried with calcined magnesium sulfate, the
solvent was removed in a vacuum of a water-jet
pump, the residue was weighed and analyzed (TLC
and 1H NMR spectroscopy). In reaction products
were present predominantly the corresponding tri-
azolones L LVIII and sometimes the original azole
and impurity of a heterolysis product from the target
compound, NH-acids of N C bitriazole series
XXXIVa XXXVIIa. When in the crude product
prevailed triazolones the latter (L LVIII) were
isolated by crystallization. The mother liquor after
crystallization was evaporated, the residue was put
into water, treated with sodium carbonate till pH was
2 units lower than pKa of the original azole, the azole
1
spectrum ( , ppm): 6.00 s (2H, CH2), 3.56 s (3H,
CH3). Found, %: C 21.9; H 2.9; N 38.1. M 210.
C4H6N6O5. Calculated, %: C 22.0; H 2.8; N 38.5. M
218.12. The compound was also prepared by an
independent synthesis by treating 1-(2-nitro-2-aza-
propyl)-3,5-dinitro-1,2,4-triazole (XIX) with a base
by procedure [18].
REFERENCES
1. Bagal, L.I., Pevzner, M.S., and Samarenko, V.Ya.,
Khim. Geterotsikl. Soed., 1970, no. 2, pp. 269 274;
1972, no. 1, pp. 117 119; no. 4, pp. 568 571;
Bagal, L.I., Pevzner, M.S., Samarenko, V.Ya., and
Egorov, A.P., Khim. Geterotsikl. Soed., 1970, no. 5,
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001

1168
KOFMAN
1981, vol. 112, no. 5, pp. 675 678.
pp. 702 704; no. 7, pp. 997 1000; no. 12, pp. 1701
1703; Kofman, T.P., Manuilova, V.I., and Timofe-
eva, T.N., Khim. Geterotsikl. Soed., 1975, no. 5,
pp. 705 707; Kofman T.P., and Pevzner M.S., Khim.
Geterotsikl. Soed., 1981, no. 10, pp. 1403 1405;
Kofman, T.P., Kirpenko, Z.V., and Pevzner, M.S.,
Khim. Geterotsikl. Soed., 1982, no. 8, pp. 1113 1117.
14. Wartenberg, Ch., Sharrue, P., and Laval, F., Propel-
lants, Explos., Pyrotechn., 1995, vol. 20, no. 1,
pp. 23 26; Lee, K.Y., Storm, C.B., Hickey, M.A.,
and Coburn, M.D., J. Energ. Mater., 1991, vol. 9,
no. 5, pp. 415 428; Chem. Abstr., 1992, vol. 116,
217504; Kofman, T.P. and Kartseva, G.Yu., Zh. Org.
Khim., 2000, vol. 36, no. 6, pp. 899 905.
15. Kofman, T.P., Uspenskaya, T.L., and Pevzner, M.S.,
Zh. Org. Khim., 1993, vol. 29, no. 11, pp. 2328
2330; Kofman, T.P., Pevzner, M.S., Zhukova, L.N.,
Kravchenko, T.A., and Frolova, G.M., Zh. Org.
Khim., 1980, vol. 16, no. 2, pp. 420 425; Pevz-
ner, M.S., Kofman, T.P., and Gorbunova, N.N., Zh.
Org. Khim., 1978, vol. 14, no. 10, pp. 2024 2029;
Beuer, H., Z. Chem., 1965, vol. 5, no. 10,
pp. 381 382.
2. Pevzner, M.S., Kofman, T.P., Kibasova, E.N.,
Sushchenko, L.F., and Uspenskaya, T.L., Khim.
Geterotsikl. Soed., 1980, no. 2, pp. 257 261; USSR
Inventor,s Certificate 717048, 1978. SSSR Byull.
Izobr., 1978, no. 7; Novoe v khimii azotistykh getero-
tsiklov (News in Chemistry of Heterocycles), Riga:
Zinatne, 1979, pp. 120 122.
3. Kofman, T.P., Sushchenko, L.F., and Pevzner, M.S.,
Khim. Geterotsikl. Soed., 1980, no. 11, pp. 1553
1557; USSR
Inventor,s Certificate 753849, 1980.
16. Bagal, L.I., Pevzner, M.S., Sheludyakova, N.I., and
Kerusov, V.M., Khim. Geterotsikl. Soed., 1970,
no. 2, pp. 265 268; Kofman, T.P., Uspenskaya, T.L.,
Medvedeva, N.Yu., and Pevzner, M.S., Khim.
Geterotsikl. Soed., 1976, no. 7, pp.991 994; Ostap-
kovich, A.M., Kofman, T.P., Lisitsyna, L.V., and
Pevzner, M.S., Izv. Vuzov, Ser. Khim. i khim. tekhnol.,
1979, vol. 22, no. 4, pp. 402 406; Stotskii, A.A. and
Tkacheva, N.P., Zh. Org. Khim., 1974, vol. 10,
no. 10, pp. 2232 2234; Kofman, T.P. and Kirpen-
ko, Z.V., Zh. Org. Khim., 1994, vol. 30, no. 5,
pp. 765 769; Sickman, D.V., US Patent 2987520;
Chem. Abstr., 1962, vol. 56, no. 1, 4776.
17. Kofman, T.P., Zh. Org. Khim., 1995, vol. 31, no. 8,
pp. 1223 1226.
18. Kofman, T.P., Zhukova, L.N., and Pevzner, M.S.,
Khim. Geterotsikl. Soed., 1981, no. 4, pp. 553 558.
19. Pevzner, M.S., Samarenko, V.Ya., and Bagal, L.I.,
Khim. Geterotsikl. Soed., 1972, no. 6, pp. 848 851.
20. Pozharskii, A.F., Teoreticheskie osnovy khimii getero-
tsiklov (Theoretical Principles of Chemistry of
Heterocycles), Moscow: Khimiya, 1985.
SSSR Byull. Izobr., 1980, no. 29.
4. Kofman, T.P., Krivosheeva, G.S., and Pevzner, M.S.,
Zh. Org. Khim., 1993, vol. 29, no. 11, pp. 2304 2310.
5. Kofman, T.P., Kartseva, G.Yu., Namestnikov, V.I.,
and Paketina, E.A., Zh. Org. Khim., 1998, vol. 34,
no. 7, pp. 1084 1090.
6. Sitzman, M.E., J. Org. Chem., 1978, vol. 43, no. 17,
pp. 3389 3391.
7. Coburn, M.D. and Jakson, T.B., J. Heterocyclic
Chem., 1968, vol. 5, no. 2, pp. 199 203.
8. Wilshire, J.F.K., Austral. J. Chem., 1966, vol. 19,
no. 10, pp. 1935 1945.
9. Khan, M.A. and Lynch, B.M., J. HeterocyclicChem.,
1970, vol. 7, no. 6, pp. 1237 1239.
10. Ainkhan, M. and Polya, J.B., J. Chem. Soc., 1970,
no. 1, pp. 85 87.
11. Ivashchenko, A.V., Garicheva, O.N., Shmelev, L.V.,
Ryabokobylko, Yu.S., and Adamova, G.M., Khim.
Geterotsikl. Soed., 1980, no. 12, pp. 1673 1678.
12. Aouial, M., Bernaedini, A., and Viallefont, Ph.,
J. Heterocyclic Chem., 1977, vol. 14, no. 3,
pp. 397 400.
21. Kofman, T.P. and Kartseva, G.Yu., Zh. Org. Khim.,
2001, vol. 37, no. 5, pp. 744 754.
13. Khan, M.A. and Freitas, A.C., Monatsh. Chem.,
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 37 No. 8 2001