Decomposition of N-cyclopropyl-N-nitrosourea in the presence of reducing agents as a new way of generating the cyclopropyl radical

Article abstract of DOI:10.1007/s11172-006-0543-1

A new way of generating cyclopropyl radicals in the base-catalyzed decomposition of N-cyclopropyl-N-nitrosourea in the presence of organic reducing agents (1-phenylpyrazolidin-3-one and 4-methoxyphenol) was developed. The cyclopropyl radical generated und

Full text of DOI:10.1007/s11172-006-0543-1

Russian Chemical Bulletin, International Edition, Vol. 55, No. 11, pp. 2008—2012, November, 2006
2008
Decomposition of NꢀcyclopropylꢀNꢀnitrosourea
in the presence of reducing agents
as a new way of generating the cyclopropyl radical*
Yu. V. Tomilov,ꢀ I. V. Kostyuchenko, G. P. Okonnishnikova, I. P. Klimenko, and E. V. Shulishov
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences,
47 Leninsky prosp., 119991 Moscow, Russian Federation.
Fax: +7 (495) 135 6390. Eꢀmail: tom@ioc.ac.ru
A new way of generating cyclopropyl radicals in the baseꢀcatalyzed decomposition of
NꢀcyclopropylꢀNꢀnitrosourea in the presence of organic reducing agents (1ꢀphenylpyrazolidinꢀ
3ꢀone and 4ꢀmethoxyphenol) was developed. The cyclopropyl radical generated under these
conditions can not only abstract a proton from the substrate to give cyclopropane but also form
C—C or C—Br bonds in reactions with aromatic substrates or polybromomethanes.
Key words: cyclopropylnitrosourea, cyclopropyldiazonium, cyclopropyl radical, cycloproꢀ
panes, reducing agents, NMR spectra.
Chemical transformations of NꢀcyclopropylꢀNꢀ
chemical methods, their formation from compound 1 in
the presence of bases is confirmed by chemical methods
involving various trapping species (Scheme 1). The most
efficient way of trapping of the diazocyclopropane is to
use unsaturated compounds, which react with it accordꢀ
ing to the 1,3ꢀdipolar cycloaddition mechanism;2,3 in the
case of diazonium 3, azo coupling reactions are emꢀ
nitrosourea (1) have been investigated for more than
70 years;1 at the moment, this compound is the most
suitable source for generation of a number of highly reacꢀ
tive intermediates such as, e.g., diazocyclopropane (2)
and the cyclopropyldiazonium ion (3). Although these
intermediates cannot yet be detected by direct physicoꢀ
Scheme 1
* Dedicated to Academician O. M. Nefedov on the occasion of his 75th birthday.
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 11, pp. 1934—1938, November, 2006.
1066ꢀ5285/06/5511ꢀ2008 © 2006 Springer Science+Business Media, Inc.

Generation of the cyclopropyl radical
Russ.Chem.Bull., Int.Ed., Vol. 55, No. 11, November, 2006 2009
ployed.4,5 In the absence of appropriate trapping species,
intermediates 2 and 3 undergo spontaneous dediazotizaꢀ
tion accompanied by generation of some more unstable
species (cyclopropylidene and cyclopropyl and allyl
cations). Cyclopropylidene rearranges itself into allene
or cyclopropanates the double bond of bicyclopropylꢀ
idenes,6,7 while C3H5+ cations react with nucleophiles to
give cyclopropyl and allyl derivatives.8,9
when extensively screening compounds capable of azo
coupling with cyclopropyldiazonium ion (3).
Compound 1 was decomposed with moist K2CO3 in
the presence of pyrazolidinone 5 (0.8 equiv.) in CH2Cl2
according to a standard procedure.5 After primary workup
of the reaction mixture, the 1H NMR spectrum distinctly
showed characteristic signals for the methine protons of
the cyclopropane rings in three new compounds; the sigꢀ
nals appeared in a sufficiently high field (δH 1.9—2.9),
which is atypical of compounds with the azocyclopropane
fragment (δH ∼3.7). It is worth noting that the reaction
mixture contained no products involving the allyl cation,
though they usually form in similar reactions via partial
dediazotization of ion 3 and isomerization of the cycloꢀ
propyl cation. In these reactions, the number of allylꢀtype
compounds was the higher, the less efficient was the trapꢀ
ping species of intermediates 2 and 3 (see Ref. 9).
Apart from the aforementioned intermediates of the
baseꢀcatalyzed decomposition of compound 1, the genꢀ
eration of cyclopropyl radical (4) is also possible.10 For
instance, the decomposition of compound 1 with amines
1
gives cyclopropane (~2%), while the H NMR spectra
show the CIDNP effect for the protons of cycloꢀC3H6.
Cyclopropane is also detected in the generation of diazoꢀ
nium 3 in the presence of NaBH4 as a reducing agent.
However, no transformations of radical 4 into other prodꢀ
ucts (except for cyclopropane itself) in these processes
have been reported. Nevertheless, oneꢀelectron reducꢀ
tion of cyclopropyldiazonium (or the cyclopropyl cation)
to the cyclopropyl radical and its trapping by appropriate
substrates seem to be very likely, especially with considerꢀ
ation to some similarity of diazonium 3 with aromatic
diazonium salts, for which the reductive generation of
aryl radicals is well known.11
The compounds obtained were isolated in the indiꢀ
vidual state by column chromatography on SiO2 with benꢀ
zene—ethyl acetate (3 : 1) as an eluent and characterized
by data from elemental analysis, mass spectrometry,
1
and H and 13C NMR spectroscopy. The assigned strucꢀ
tures were 2ꢀcyclopropylꢀ1ꢀphenylpyrazolidinꢀ3ꢀone (6),
1ꢀ(2ꢀcyclopropylphenyl)pyrazolidinꢀ3ꢀone (7), and
1ꢀ(4ꢀcyclopropylphenyl)ꢀ3ꢀhydroxypyrazole (8). Along
with compounds 6—8, the starting pyrazolidinone 5
(~10%) was recovered and 3ꢀhydroxyꢀ1ꢀphenylpyrazole
(9) was obtained (Scheme 2).
Results and Discussion
Compound 9 was identified by comparing the 1H and
13C NMR spectra with the literature data.13 In the 1H and
13C NMR spectra of pyrazole 8, the signals for the pyrazole
ring are virtually identical with those in the spectra of
compound 9; however, the signals for the corresponding
Here we studied the possibility of generating radical 4
in the baseꢀcatalyzed decomposition of compound 1 in
the presence of organic reducing agents. We have discovꢀ
ered that possibility12 for 1ꢀphenylpyrazolidinꢀ3ꢀone (5)
Scheme 2

2010
Russ.Chem.Bull., Int.Ed., Vol. 55, No. 11, November, 2006
Tomilov et al.
atoms of the benzene ring unambiguously indicate the
presence of the paraꢀcyclopropyl substituent in the sysꢀ
tem (see Experimental). In contrast to pyrazoles 8 and 9,
ence of pyrazolidinone 5, which ensures the electron transꢀ
fer to cyclopropyldiazonium 3. A specific attack of the
cyclopropyl radical on the phenyl substituent* (orthoꢀsubꢀ
stitution for 1ꢀphenylpyrazolidine 5 and paraꢀsubstituꢀ
tion for 1ꢀphenylpyrazole 9) is probably due to various
effects of conjugation of the phenyl substituent with those
heterocycles.
The cyclopropyl radical mainly generated by thermoꢀ
lysis or photolysis of biscyclopropanoyl peroxide is
known14,15 to be a highly reactive species; its reactivity is
one to two orders of magnitude higher than most primary
aliphatic radicals. According to quantumꢀchemical calꢀ
culations,16 the energy barrier to an electrocyclic transꢀ
formation of this species into an allyl system is appreciaꢀ
bly higher than that for the cyclopropyl cation; the allyl
radical usually isomerizes under photolysis conditions.17
The high hydrogen affinity of the cyclopropyl radical has
also been reported.18 The above data explain well the
absence of allyl derivatives among the reaction products
and the sufficiently high yield of cyclopropane.
We tried to extend the range of reagents capable of
generating the cyclopropyl radical in the decomposition
of compound 1 and, as far as possible, obtain derivatives
of the cyclopropane series. Experiments were carried out
in the tube of the NMR spectrometer at 20 °C, by adding
a double amount of Cs2CO3 to a mixture of the reagents
in CD2Cl2. Hydroquinone could be a good reducing agent
for diazonium 3; however, it is poorly soluble in diꢀ
chloromethane and thus is unfit for NMR experiments.
Instead, we used 4ꢀmethoxyphenol, which also exhibits
reductive properties.
Addition of Cs2CO3 and 4ꢀmethoxyphenol to a soluꢀ
tion of compound 1 caused gas evolution; cyclopropane,
allene, and 2ꢀ(cyclopropylazo)ꢀ4ꢀmethoxyphenol (11) as
an azo adduct were detected already during the first
minutes of the process, despite poor recording conditions
(Scheme 3). It should be noted that the CIDNP effect for
the signal of cyclopropane was observed during the reacꢀ
tion. After the reaction was completed (20—25 min), the
molar ratio of 11 : cyclopropane : allene was ~1.6 : 1.8 : 1.0.
To isolate and identify the earlier unknown azo comꢀ
pound 11, we carried out this reaction in a reaction vessel
with equimolar amounts of compound 1 and 4ꢀmethoxyꢀ
phenol and K2CO3 as a base (see Experimental).
1
the H and 13C NMR spectra of pyrazolidinones 6 and 7
are similar to those of the starting pyrazolidinone 5. The
1
difference is that the H NMR spectrum of compound 6
shows no lowꢀfield signal for the NH proton but exhibits
signals for the cyclopropyl protons (for the methine proꢀ
ton, δ 2.92); in contrast, 1H NMR spectrum of the strucꢀ
turally isomeric compound 7 contains a signal for the
NH proton at δ 9.1 but the range of signals for aromatic
protons shows a fourꢀspin system typical of the orthoꢀsubꢀ
stituted benzene ring and the signal for the methine proꢀ
ton of the cyclopropane ring appears at δ 2.18.
To confirm the formation of pyrazole 9 directly during
the reaction rather than during chromatographic separaꢀ
tion of the reaction mixture, we decomposed compound 1
in the tube of an NMR spectrometer (CD2Cl2, 0 °C).
Under these conditions, stirring is insufficient and the
reaction was catalyzed by Cs2CO3, which is a stronger
base than K2CO3. This experiment proved the formation
of hydroxypyrazole 9 during the course of the reaction
and additionally revealed two important results. The
1H NMR spectrum of the reaction mixture contains a
very intense singlet at δ 0.24 that indicates the formation
of cyclopropane in ~35% yield (estimated from the inteꢀ
1
gral intensities of the signals). In addition, the H NMR
spectrum shows another distinct signal (δ 3.42, tt) characꢀ
teristic of the methine proton of the cyclopropane fragꢀ
ment (most likely, (cycloꢀC3H5)—N=N). Immediately
after the reaction was completed, this signal was domiꢀ
nant in contrast to analogous signals for compounds 7
and 8 and especially to the signal for the methine proton
of Nꢀcyclopropylpyrazolidinone 6. However, this signal
gradually disappeared with time (1—20 h, 20 °C), which
was accompanied by a simultaneous increase in the intenꢀ
sities of the signals due to compound 6; the intensities of
the other signals remained unchanged.
The results obtained suggest that the spectroscopically
detected intermediate is 2ꢀ(cyclopropylazo)ꢀ1ꢀphenylꢀ
pyrazolidinꢀ3ꢀone (10) formed via addition of diazonium
3 to the NH group of the starting pyrazolidinone 5. An
analogous process leading to cyclopropyltriazenes has been
observed earlier in the generation of ion 3 in the presence
of ethylꢀ and dimethylamines.10 Subsequent spontaneous
dediazotization of Nꢀazopyrazolidinone 10 is probably
sufficiently selective, yielding compound 6.
The presence of cyclopropane in the reaction mixture
suggests the generation of the cyclopropyl radical under
these conditions; therefore, radical 4 can also be trapped
by other substrates. It turned out that addition of Cs2CO3
and 4ꢀmethoxyphenol to a solution of compound 1 conꢀ
taining a small excess of CBr4 gives rise, along with azo
adduct 11 and allene, to cyclopropyl bromide (12) and
The results obtained (on the one hand, the presence of
products of oxidation of the saturated heterocycle into
hydroxypyrazoles 8 and 9 and, on the other hand, a suffiꢀ
ciently rare example of insertion of the cyclopropyl fragꢀ
ment into the C—H bond of the phenyl substituent14 and
the formation of cyclopropane itself) provide strong eviꢀ
dence for the generation of the cyclopropyl radical during
the decomposition of compound 1 with bases in the presꢀ
* We do not exclude the possible formation of paraꢀisomer 7 and
orthoꢀisomer 8; however, their amounts are too small for reliable
identification.

Generation of the cyclopropyl radical
Russ.Chem.Bull., Int.Ed., Vol. 55, No. 11, November, 2006 2011
Scheme 3
Decomposition of compound 1 in the presence of 1ꢀphenylꢀ
pyrazolidinꢀ3ꢀone (5). NꢀCyclopropylꢀNꢀnitrosourea (1) (0.62 g,
4.8 mmol) was added in small portions at 5 °C to a stirred mixꢀ
ture of 1ꢀphenylpyrazolidinꢀ3ꢀone (5) (0.65 g, 4.0 mmol) and
moist K2CO3 (1.32 g; ~20% water) in CH2Cl2 (45 mL). The
reaction mixture was stirred at this temperature for 3 h and
filtered. The dark red filtrate was concentrated and the residue
was separated by column chromatography on SiO2 with benꢀ
zene—AcOEt (3 : 1) as an eluent. After the removal of the
solvents, the starting pyrazolidinone 5 (0.064 g, ~10%) was reꢀ
covered and compounds 6—9 were obtained.
2ꢀCyclopropylꢀ1ꢀphenylpyrazolidinꢀ3ꢀone (6). The yield
was 0.16 g (20%). Found (%): C, 70.95; H, 6.87; N, 13.71.
C12H14N2O. Calculated (%): C, 71.26; H, 6.98; N, 13.85.
1H NMR (CDCl3), δ: 0.80 (m, 4 H, CH2CH2); 2.50 (t, 2 H,
H2C(4), J = 7.4 Hz); 2.92 (m, 1 H, CH); 3.75 (t, 2 H, H2C(5),
J
= 7.4 Hz); 7.35 and 7.05 (both m, 2 + 3 H, Ph).
bromoform; the spectrum of the reaction mixture conꢀ
tained no signal for cyclopropane. After the reaction was
completed, the integral intensities of the nonoverlapping
signals for bromide 12, azo adduct 11, bromoform, and
allene corresponded to their molar ratio ~8 : 6 : 2 : 1.
MS, m/z (Irel (%)): 202 [M]+ (39), 187 (16), 145 (82), 117
(100), 77 (92).
1ꢀ(2ꢀCyclopropylphenyl)pyrazolidinꢀ3ꢀone (7). The yield was
0.11 g (14%), m.p. 152—153 °C. Found (%): C, 71.03; H, 6.90;
N, 13.69. C12H14N2O. Calculated (%): C, 71.26; H, 6.98;
N, 13.85. 1H NMR (CDCl3), δ: 0.74 and 1.03 (both m, 2 H each,
CH2CH2); 2.18 (m, 1 H, CH); 2.55 (t, 2 H, H2C(4), J = 7.6 Hz);
3.87 (t, 2 H, H2C(5), J = 7.6 Hz); 6.85 (dd, 1 H, H(3´), J =
1.6 Hz, J = 7.3 Hz); 7.10 and 7.18 (both m, 1 H each, H(4´) and
H(5´); 7.28 (dd, 1 H, H(6´), J = 1.6 Hz, J = 7.6 Hz). 13C NMR
(CD3OD), δ: 9.36 (CH2CH2), 11.88 (CH), 30.33 (C(4)), 56.35
(C(5)), 117.70 (C(6´)), 125.56 and 125.71 (C(4´) and C(5´)),
126.84 (C(3´)), 136.40 (C(2´)), 151.77 (C(1´)), 177.71 (C=O).
MS, m/z (Irel (%)): 202 [M]+ (24), 173 (68), 131 (100), 119 (27),
91 (27), 77 (39), 55 (35).
1ꢀ(4ꢀCyclopropylphenyl)ꢀ3ꢀhydroxypyrazole (8). The yield
was 0.096 g (12%), m.p. 168—170 °C. Found (%): C, 71.72;
H, 6.00; N, 13.75. C12H12N2O. Calculated (%): C, 71.98;
H, 6.04; N, 13.99. 1H NMR (CDCl3), δ: 0.72 and 1.00 (both m,
2 H each, CH2CH2); 1.92 (m, 1 H, CH); 5.88 (d, 1 H, H(4), J =
2.5 Hz); 7.14 and 7.42 (both m, 2 H each, paraꢀC6H4); 7.62 (d,
1 H, H(5), J = 2.5 Hz). 13C NMR (CDCl3), δ: 9.29 (CH2CH2),
15.06 (CH), 93.83 (C(4)), 118.97 (C(2´) and C(6´)), 126.05
(C(3´) and C(5´)), 129.07 (C(5)), 137.33 (C(4´)), 142.11 (C(1´)),
163.86 (C(3)). MS, m/z (Irel (%)): 200 [M]+ (100), 165 (26), 117
(69), 91 (25), 77 (29), 63 (28).
3ꢀHydroxyꢀ1ꢀphenylpyrazole (9) was identified by compariꢀ
son with the literature data.13 The yield was 0.17 g (27%).
2ꢀ(Cyclopropylazo)ꢀ4ꢀmethoxyphenol (11). Potassium carꢀ
bonate (2.76 g) containing water (~20%) was added to a solution
of 4ꢀmethoxyphenol (0.62 g, 5 mmol) and compound 1 (0.65 g,
5 mmol) in CH2Cl2 (60 mL). The reaction mixture was stirred at
2—5 °C for 6 h and filtered. The filtrate was concentrated in vacuo
and the residue was separated by column chromatography on
SiO2 with benzene—ether (70 : 1) as an eluent. The isolated
fractions contained the starting 4ꢀmethoxyphenol (0.35 g, ~55%)
and azo adduct 11 (0.36 g, 37%).
2ꢀ(Cyclopropylazo)ꢀ4ꢀmethoxyphenol (11), m.p. 57—58 °C.
Found (%): C, 62.63; H, 6.40; N, 14.29. C10H12N2O2. Calcuꢀ
lated (%): C, 62.49; H 6.29; N, 14.57. 1H NMR (CD2Cl2), δ:
1.32 and 1.45 (both m, 2 H each, CH2CH2); 3.68 (tt, 1 H, CH,
J = 3.3 Hz, J = 7.0 Hz); 3.75 (s, 3 H, OMe); 6.95 (dd, 1 H, H(6),
Scheme 4
Under analogous conditions, the use of bromoform
instead of CBr4 also afforded cyclopropyl bromide 12; the
CIDNP effect was observed in the spectrum recorded two
minutes after the beginning of the reaction. The final
reaction mixture contained bromide 12, azo adduct 11,
CH2Br2, allene, and cyclopropane in the molar ratio
24 : 15 : 6 : 4.5 : 1 (see Schemes 3, 4).
Thus, the baseꢀcatalyzed decomposition of Nꢀcycloꢀ
propylꢀNꢀnitrosourea in the presence of reducing agents
such as 1ꢀphenylpyrazolidinꢀ3ꢀone (5) or 4ꢀmethoxyꢀ
phenol allows generation of the cyclopropyl radical, which
can not only abstract proton from a substrate to give cycloꢀ
propane but also alkylate the benzene ring or form C—Br
bonds in reactions with polybromomethanes.
Experimental
1
H and 13C NMR spectra were recorded on a Bruker AMꢀ300
spectrometer (300 and 75.5 MHz, respectively) in CDCl3,
CD3OD, or CD2Cl2 all containing 0.05% Me4Si as the internal
standard. Mass spectra were recorded on a Finnigan MAT
INCOSꢀ50 instrument (EI, 70 eV, direct inlet probe). For chroꢀ
matography, silica gel 60 (0.040—0.063 mm; Merck) was
used. 1ꢀPhenylpyrazolidinꢀ3ꢀone (Reakhim), 4ꢀmethoxyphenol,
cyclopropyl bromide, tetrabromomethane, bromoform, and ceꢀ
sium carbonate (Merck) were used as purchased. NꢀCyclopropylꢀ
Nꢀnitrosourea (1) was prepared according to a known proꢀ
cedure.1
Jpara = 0.6 Hz, Jortho = 9.0 Hz); 6.90 (dd, 1 H, H(5), Jmeta
2.8 Hz, Jortho = 9.0 Hz); 7.29 (dd, 1 H, H(3), Jmeta = 2.8 Hz,
=

2012
Russ.Chem.Bull., Int.Ed., Vol. 55, No. 11, November, 2006
Tomilov et al.
Jpara = 0.6 Hz); 11.0 (br.s, 1 H, OH). 13C NMR (CDCl3), δ:
10.77 (CH2CH2), 49.33 (CH), 55.96 (OMe), 113.66 (C(3)),
118.40 (C(5)), 119.44 (C(6)), 135.85 (C(2)), 146.46 (C(1)),
152.70 (C(4)). MS, m/z (Irel (%)): 192 [M]+ (43), 177 [M – Me]+
(8), 165 (75), 122 (26), 108 (42), 95 (22), 79 (22), 68 (18), 52
(26), 41 (49), 39 (100).
Reactions in the tube of an NMR spectrometer (general proꢀ
cedure). A solution of compound 1 (10 mg, 0.077 mmol)
in CD2Cl2 (0.4 mL) and 1ꢀphenylpyrazolidinꢀ3ꢀone (5) or
4ꢀmethoxyphenol (0.07 mmol) were placed in the tube. In exꢀ
periments with polyhalomethanes, CBr4 or CHBr3 (0.1 mmol)
was also added. Then Cs2CO3 (50 mg, 0.153 mmol) was added
in one portion at 0 °C and the tube was placed in the probe of
the NMR spectrometer. Spectra were recorded at regular periꢀ
ods of time (e.g., 2, 10, 30 min, etc.). The spectra were compared
with each other and with the spectra of known compounds; the
ratio of the products was determined from the integral intensiꢀ
ties of the nonoverlapping signals.
References
1. V. P. Gol´mov, Zh. Obshch. Khim., 1935, 5, 1562 [Chem.
Zentralbl., 1936, 107 (II), 1905].
2. Yu. V. Tomilov, I. V. Kostyuchenko, and O. M. Nefedov,
Usp. Khim., 2000, 69, 507 [Russ. Chem. Rev., 2000, 69, 461
(Engl. Transl.)].
3. Yu. V. Tomilov and I. V. Kostyuchenko, Sovremennye
problemy organicheskoi khimii [Modern Problems of Organic
Chemistry], St.ꢀPetersburg State University, St.ꢀPetersburg,
2001, Issue 13, 113 (in Russian).
4. Yu. V. Tomilov, I. V. Kostyuchenko, E. V. Shulishov, and
O. M. Nefedov, Mendeleev Commun., 2002, 104.
5. Yu. V. Tomilov, I. V. Kostyuchenko, E. V. Shulishov, and
G. P. Okonnishnikova, Izv. Akad. Nauk, Ser. Khim., 2003,
941 [Russ. Chem. Bull., Int. Ed., 2003, 52, 993].
6. L. Fitjer and J. Conia, Angew. Chem., 1973, 85, 349.
7. A. de Meijere and S. I. Kozhushkov, Chem. Rev., 2000,
100, 93.
8. W. Kirmse and H. Schütte, Chem. Ber., 1968, 101, 1674.
9. I. P. Klimenko, V. A. Korolev, Yu. V. Tomilov, and O. M.
Nefedov, Zh. Org. Khim., 2006, 42, 1320 [Russ. J. Org. Chem.,
2006, 42, 1299].
10. W. Kirmse and U. Seipp, Chem. Ber., 1974, 107, 745.
11. H. Zollinger, Diazo Chemistry I, Weinheim, VCH
Verlagsgesellschaft, 1994, 1, 188.
12. Yu. V. Tomilov, I. V. Kostyuchenko, I. P. Klimenko, and
G. P. Okonnishnikova, Abstrs VII Conf. on the Chemistry of
Carbenes and Related Intermediates, Kazan, 2003, 22.
13. H. H. Adam, B. D. Baigrie, and T. A. Joslin, J. Chem. Soc.,
Perkin Trans. 2, 1977, 1287.
14. T. Shono and I. Nishiguchi, Tetrahedron, 1974, 30, 2183.
15. L. J. Johnston, J. C. Scaiano, and K. U. Ingold, J. Am.
Chem. Soc., 1984, 106, 4877.
A. Cesium carbonate was added to a solution of compound 1
and pyrazolidinone (5). After 1 h, compounds 6—10 and cycloꢀ
propane itself were detected in the molar ratio ~1 : 2 : 2 : 6 : 4 : 3.
After a 20ꢀh exposure at 20 °C, the ratio of compounds 6—9 was
~5 : 2 : 2 : 6, while compound 10 was virtually absent.
B. Cesium carbonate was added to a solution of compound 1
1
and 4ꢀmethoxyphenol (5). After 2 min, the H NMR spectrum
showed a strong emission of the signal for cyclopropane
(CIDNP effect). After the reaction was completed, azo adꢀ
duct 11, cyclopropane, and allene were detected in the molar
ratio 1.8 : 1.6 : 1.0.
C. Cesium carbonate was added to a solution of compound 1,
4ꢀmethoxyphenol, and CBr4. After the reaction was completed,
azo adduct 11, cyclopropyl bromide, bromoform, and allene
were detected in the molar ratio ~6 : 8 : 2 : 1. Cyclopropane was
virtually absent from the reaction mixture.
D. Cesium carbonate was added to a solution of compound 1,
4ꢀmethoxyphenol, and CHBr3. After 2 min, the 1H NMR specꢀ
trum showed a CIDNP effect for the signals of cyclopropane
and cyclopropyl bromide. After the reaction was completed, azo
adduct 11, cyclopropyl bromide, CH2Br2, allene, and cycloproꢀ
pane were detected in the molar ratio ~15 : 24 : 6 : 4 : 1.
16. D. J. Mann and W. L. Hase, J. Am. Chem. Soc., 2002,
124, 3208.
17. (a) G. Maier and S. Senger, Angew. Chem., Int. Ed., 1994,
33, 558; (b) K. B. Wiberg and C. G. Österle, J. Org. Chem.,
1999, 64, 7756.
18. I. G. Green and B. P. Roberts, J. Chem. Soc., Perkin Trans. 2,
1986, 1597.
This work was financially supported by the Russian
Foundation for Basic Research (Project No. 06ꢀ03ꢀ
33149).
Received September 22, 2006;
in revised form October 11, 2006