The reaction of nitrosodicyclohexylamine with organolithiums

Article abstract of DOI:10.1002/poc.1451

The detailed study of the reaction of N-nitrosamines with organolithium compounds, under different reaction conditions, allowed the development of a useful methodology for the synthesis of substituted hydrazones and trialkyl hydrazines. The reaction proved to be very sensitive to the reaction conditions, and different main products can be obtained by ine tuning of several variables. With the purpose of searching into the reaction mechanism, a careful isolation, characterization, and quantitative determination of several minor products was carried out. N,N-dicyclohexylamine, N-cyclohexylidencyclohexyl amine, and N,N,N′-dicydohexylalkylhydrazines, were the main side products identified after the work up of the reaction mixtures. With the same aim, the kinetics of the reaction of N-nitrosodicyclohexylamine with n-BuLi, was determined at temperatures in the range 0°C room temperature. Under the conditions leading to the trialkyl hydrazine, the results show an abrupt slowdown of the reaction rate, after a short reaction time. This result, together with other observations, such as the recovering of hydrazone even when using high [RLi], is indicative of equilibrium involving different species in the reaction mixture. The isolation of reduction products, as well as additional experiments carried out with lithium dialkylamides and NO, allowed suggestion of an overall mechanistic scheme for the complex consecutive and parallel reactions, that is consistent with the afforded evidences. Copyright

Full text of DOI:10.1002/poc.1451

Special Issue
Received: 31 January 2008,
Revised: 26 August 2008,
Accepted: 26 August 2008,
Published online in Wiley InterScience: 30 September 2008
(www.interscience.wiley.com) DOI 10.1002/poc.1451
The reaction of nitrosodicyclohexylamine
with organolithiums
a
a
a
*
´
Alvaro J. Vazquez , Cristian Rodrı´guez and N. Sbarbati Nudelman
The detailed study of the reaction of N-nitrosamines with organolithium compounds, under different reaction
conditions, allowed the development of a useful methodology for the synthesis of substituted hydrazones and trialkyl
hydrazines. The reaction proved to be very sensitive to the reaction conditions, and different main products can be
obtained by fine tuning of several variables. With the purpose of searching into the reaction mechanism, a careful
isolation, characterization, and quantitative determination of several minor products was carried out. N,N-
dicyclohexylamine, N-cyclohexylidencyclohexyl amine, and N,N,N(-dicyclohexylalkylhydrazines, were the main side
products identified after the work up of the reaction mixtures. With the same aim, the kinetics of the reaction of
N-nitrosodicyclohexylamine with n-BuLi, was determined at temperatures in the range 0 -C room temperature. Under
the conditions leading to the trialkyl hydrazine, the results show an abrupt slowdown of the reaction rate, after a short
reaction time. This result, together with other observations, such as the recovering of hydrazone even when using
high [RLi], is indicative of equilibrium involving different species in the reaction mixture. The isolation of reduction
products, as well as additional experiments carried out with lithium dialkylamides and NO, allowed suggestion of an
overall mechanistic scheme for the complex consecutive and parallel reactions, that is consistent with the afforded
evidences. Copyright ß 2008 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: nitrosamines; organolithiums; nitrosonium ions; hydrazones; hydrazines; denitrosation; reaction mechanism
literature, but few dealt with the addition of lithium amides.[19]
INTRODUCTION
The first adduct between an amine and a carbonyl derivative was
isolated few years ago;[20] by 13C NMR we characterized the first
adduct between a formamide and R2NLi,[21] and a more complex
dilithiated adduct was lately characterized.[22] To the best of our
knowledge, scarce studies dealt with the addition of RLi on a
carbon–nitrogen double bond. The reaction described in this
paper is extremely sensitive to the reaction conditions and
precise determination of the reaction order was very difficult;
nevertheless, the kinetic studies together with other evidences
suggest the presence of complex equilibrium and competitive
reactions.
New reactions on nitrosamines are being actively studied
specially on their relationship with carcinogenetic and mutagenic
properties;[1] recent studies showed that they are mainly
mutagenic through methylation of DNA.[2] There is special
concern on the important roles that environmental nitrosamines
play in the etiology of human cancer.[3] The tobacco-specific
nitrosamine, (nicotine-derived nitrosamine ketone, NNK) found in
cigarette smoke induces lung tumors in mice,[4] and esophageal
and gastroduodenal cancer in animals,[5,6] nitrosamines have
been also found in several food and beverages.[7–9]
On the other hand, hydrazones and hydrazines are receiving
renewed interest due to the recent discovery of remarkable
biological activities. These derivatives are well-known among
pesticides, drugs, amino acid precursors, and synthetic building
blocks for heterocyclic synthesis.[10] Several similar compounds
were shown to be effective for the treatment of various
diseases;[11] and hydrazine-based derivatives are found to be
potent agents against hepatitis, AIDS, and SARS.[12,13] Some
systematic synthetic methods have been recently developed,
using protective group methodology, but they demand many
steps (and the production of significant residues) for obtaining
the end product;[14,15] development of cleaner tandem meth-
odologies are desired.[16,17]
RESULTS AND DISCUSSION
—
Alkyl additions to the N O bond
—
The reaction of some organolithium compounds, R1CH2Li, with
N-nitrosodicyclohexylamine, 1, in THF, leads to the production of
hydrazones, 2, as indicated in the Eqn (1)
*
Correspondence to: N. S. Nudelman, Department of Organic Chemistry,
Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires,
Pab. II, P. 3, Ciudad Universitaria, (1428) Buenos Aires, Argentina.
E-mail: nudelman@qo.fcen.uba.ar
We recently reported the insertion of NO into the N—Li bond
of lithium dialkylamides[18] and the present paper describes an
extension of this reaction based on tandem additions of
organolithium reagents. Mechanistic studies on the addition of
RLi on a carbon–oxygen double bond are abundant in the
´
A. J. Vazquez, C. Rodr´ıguez, N. S. Nudelman
Departamento de Quimica Organica, Facultad de Ciencias Exactas
a
y
Naturales, Universidad de Buenos Aires, Pab. II, P. 3, Ciudad Universitaria,
(1428) Buenos Aires, Argentina
J. Phys. Org. Chem. 2008, 21 1098–1104
Copyright ß 2008 John Wiley & Sons, Ltd.

NITROSODICYCLOHEXYLAMINE WITH ORGANOLITHIUMS
hindrance of the n-pentyl groups could be responsible for the
lower reactivity. In the reactions with MeLi, normal addition to
form the hydrazone of formaldehyde (2a) occurred; but when a
second molecule of MeLi is added, loss of the elements of lithium
hydride spontaneously occurs and the hydrazone of acet-
—
aldehyde, (c-C H ) NN CHCH , is isolated as the main product
—
instead of the trialkylhydrazine expected. A similar LiH elimination
6
11 2
3
was early observed in the reaction of diethylnitrosamine with
—
formed.[23] In good agreement with these observations, the yields
of side products 4 and 5 are significantly higher than for the
other RLi.
MeLi, for which a tiny amount (<2%) of Et NN CHCH was
—
2
3
The reaction can be carried out at room temperature; in most
cases the reaction time was very short (5 min, approx.). 1 mmol of
1 was dissolved in 2 ml of THF and the organolithium reagent was
added in a [R1CH2Li]:[1] ¼ 1.05–1.2 molar ratio. The concentration
of organolithium was in the range 0.5–1.5 M approx., in n-hexane
solution. In the case of MeLi, (that is less soluble in hexane) 1 M
approx. THF solution was used, and a longer reaction time (2 h)
was required. Table 1 summarizes the results. The reaction with
PrLi (entry b) is particularly interesting, since the reaction is very
clean and high yields are obtained. In the case of n-C5H11Li (entry
d), the yields are slightly lower likely due to the higher steric
hindrance.
Kinetic study
The reaction of 1 with organolithium compounds, carried out
under the conditions previously mentioned renders the main
products described. Nevertheless, the reaction leads also to the
formation of some secondary products: dicyclohexylamine, 4,
N-cyclohexylidenecyclohexylamine, 5, and N,N,N0-dicyclohexyl-
N,N0-dicyclohexylalkylhydrazines, 6, in low variable yields. It is
interesting to comment the effect of adding more n-BuLi, after
the reaction has advanced (i.e., when some remaining 2c was still
present in the reaction mixture), in experiments carried out under
the conditions that lead to the formation of 3c as the main
product. In that cases, it was observed that a certain amount of
the remaining 2c could be converted to 3c, although not
completely. On the other hand, the amounts of 4, 5, and 6c
remained unchanged. These observations suggest the existence
of chemical equilibria involving the n-BuLi and the species
present in the reaction mixture, as it was reported before for
other reactions involving lithium amides.[21,22,24]
Tandem double alkylation
A second addition of another molecule of organolithium can be
carried out in a tandem sequence starting from 1. By carrying out
the reaction of 1 with RLi, at room temperature, using a relatively
high excess of RLi (Eqns (3)–(5)), a trialkyl hydrazine, 3, is
produced in good yields (Eqn (2))
To shed some light into the mechanism of this complex
reaction a kinetic study was undertaken. The formation of 2c
(starting from 1, and adding n-BuLi) is very fast (occurs in a few
seconds, even at 0 8C), therefore, a kinetic study of the formation
of this compound is difficult. In contrast, the formation of 3c (at
expenses of 2c) is slower, and could be studied at low
temperatures. A kinetic study of the formation of 3c at different
temperatures, starting from 1 and adding some excess of n-BuLi,
was carried out. Under the conditions of this study, the hydrazone
2c is formed in situ, and it reacts with more n-BuLi, affording 3c
(Eqn (3))
In all cases the reactions were carried out in THF for 2 h at room
temperature (runs at lower temperatures were not satisfactory).
The results are shown in the Table 2. As can be observed, good
yields of 3 were obtained for n-PrLi and n-BuLi (entries b and c).
With n-C5H11Li, instead, lower yields of 3d were obtained, and
significant amounts of 2d remained, likely the higher steric
Table 1. Reaction of dicyclohexylnitrosamine, 1, with RLi in
THF at room temperature. Relative % yields of dicyclohexyl-
alkylhydrazones, 2 and side products
Entry RLia
R1
% 2b,c % 4d % 5e % 6f
a
b
c
MeLi
n-PrLi
n-BuLi
n-C5H11Li
H
Et
n-Pr
n-Bu
63g
93
88
69
5
—
3
6
7
4
12
1
—
3
The reaction was followed by taking aliquots at different time
intervals after the addition of n-BuLi, and analyzing the aliquots
by GC. Tests were carried out at different temperatures and the
results are shown in the Fig. 1. As can be observed, formation of
3c is very fast at the beginning, but becomes very slow after
nearly 20 min of reaction at room temperature (it is even slower
at lower temperatures), remaining certain amounts of 2c. This
kinetic behavior, along with the observations commented
above, could be indicative of a complex mechanistic scheme,
involving equilibrium between the species and some lateral
reactions.
d
6
—
a [RLi]:[1] ¼ 1.05–1.2.
b
1
—
2: (C H ) NN CHR .
—
6
11 2
c No nitrosamine recovered.
d 4: (C6H11)2NH.
e 5: (C6H11)N(C6H10).
f 6: (C6H11)2NNH-CH2R1.
g 21% of 2c observed.
J. Phys. Org. Chem. 2008, 21 1098–1104
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
A. J. VAZQUEZ, C. RODRIGUEZ AND N. S. NUDELMAN
´
the second organolithium, R2Li, a higher excess was used
(4–5 equivalents of reagent), and longer reaction times were
needed (2 h). As can be observed in the Table 3, the yields of 7 are
good but variable amounts of 2 remained. For a couple of two
organolithiums, different products can be obtained depending
on the order in which the two organolithiums were added. On the
other hand, it can be observed that using PhLi, the corresponding
hydrazine is not obtained (even if the reaction is carried out at
50 8C), this result is, likely, a consequence of the lower reactivity of
PhLi, compared to the alkyllithium compounds. Consequently,
the yield of denitrosation products increases considerably.
Table 2. Reaction of dicyclohexylnitrosamine, 1, with RLi (in
high excess) in THF, at room temperature. Relative % yields of
trialkyl hydrazines, 3 and side productsa
Entry RLi
[RLi]:[1] R1
% 2b % 3c % 4d % 5e % 6f
a
b
c
MeLi
3
4
5
8
H
28
9
36g,h
77
11
3
8
1
1
1
4
2
4
5
n-PrLi
Et
n-BuLi
n-Pr
n-Bu
2
75
61i
3
d
n-C5H11Li
13
5
a Temperature: 20–25 8C. Reaction time 2 h.
b
1
—
2: (c-C H ) NN CHR .
—
6
11 2
c 3: (c-C6H11)2NNH-CH(R1)R.
Side products and mechanistic hints
d 4: (C6H11)2NH.
e 5: (C6H11)N(C6H10).
For the study of a reaction mechanism, not only the main
products are important, but the nature of the compounds that
are produced by lateral reactions used to be significant. In all the
above described reactions, some side products were also
produced in variable amounts. To shed more light into the
general mechanism of the global reaction mechanism, careful
efforts were devoted to the isolation, characterization, and
quantitative determination of the main side products. Those were
identified as dicyclohexylamine, 4, N-cyclohexylidenecyclohexylamine,
5, and N,N,N0-dicyclohexylalkylhydrazines, 6, which are shown
below.
f 6: (C6H11)2NNH-CH2R1.
g
—
The main product is (c-C H ) NN CHCH .
—
6
11 2
3
h 9% of 2b; side products 36% total yield.
i side products 26% total yield.
Reaction of nitrosamine, 1, with two different
organolithium compounds
With the purpose of gaining more information into the
mechanism of the reaction of 1 with organolithium compounds,
the reaction was tested at room temperature, using two different
organolithiums. Under these conditions, branched hydrazines, 7,
can be obtained as shown in Eqn (4).
Variable small amounts of the corresponding dialkyl amine
(such as 4), were also detected in the reaction mixture, when
nitrosamines were synthesized by the NO insertion into the N—Li
bond of lithium amides, following the procedure previously
reported.[18] The finding of this type of side product (i.e., dialkyl
amines) under those reaction conditions could, in principle, be
explained by small amounts of remaining lithium amide and/or
by the reagent slight hydrolysis in the reaction flask. Never-
theless, in the present case, the reaction starts from pure (amine
free) solid nitrosamine 1. The appearance of 4 and 5 in the
reaction mixture suggests the involving of some denitrosation
process, likely occurring through a nitrosiminium ion intermedi-
ate stabilized by resonance, as shown in Scheme 1.
The addition of the R1CH2Li was carried out using slight excess
of the reagent, and short reaction times (a few minutes, except in
the case of MeLi, that required longer times). These are the
conditions leading to the hydrazone, 2. For the addition of
Denitrosation of N-nitroso-N-dialkylamines is a subject of
active interest, since it is thought their carcinogenic and
mutagenic activity to be due in part to the fact that they
decompose to precursors of DNA-alkylating agents. Therefore,
the solution chemistry of these substances is of considerable
importance in understanding nitrosamine carcinogenesis. Fish-
bein and coworkers[25] reported the first evidence of
N-nitrosiminium ions in the non-enzymatic decomposition of
some a-acetoxy-N-nitrosodialkylamines. N-nitrosiminium ions are
ambident electrophiles and can be trapped by special agents and
solvents. Azide ion is known to be an efficient trap of some
unstable cations, reacting with a diffusion-limited rate constant.
The kinetic of the trapping of N-nitrosiminium ions by azide ions
was reported, and it demonstrated that formation of the
Figure 1. Reaction of dicyclohexylnitrosamine, 1, with 3 equivalent BuLi
in THF, at three temperatures. % yields of trialkylhydrazine, 3c, as a
function of the reaction time
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2008, 21 1098–1104

NITROSODICYCLOHEXYLAMINE WITH ORGANOLITHIUMS
Table 3. Reaction of dicyclohexylnitrosamine, 1, with two different organolithium compounds in THF, at room temperature. Relative
% yields of trialkyl hydrazines, 7 and side productsa
Entry
RLib
R2Li
[R2Li]:[1]
R1
% 2c
%4d
%5e
%6f
% 7g
a
b
c
d
e
MeLi
n-BuLi
n-BuLi
n-C5H11Li
n-PrLi
4
5
5
5
3
H
Et
Et
n-Pr
n-Pr
1
5
5
5
6
1
14
4
5
2
6
16
1
6
—
4
64h
69
n-PrLi
n-PrLi
n-BuLi
n-BuLi
2 (2b)
9 (2c)
61 (2c)
59i
64
PhLi
7
—
a Reaction time: 2 h.
b [RLi]:[1]:1.05–1.2.
c
1
—
2: (c-C H ) NN CHR .
—
6
11 2
d 4: (C6H11)2NH.
e 5: C6H11)N(C6H10).
f 6: (C6H11)2NNHCH2R1.
g 7: (c-C6H11)2NNH-CHR1R2.
h 15% of 3c determined.
i Side products 39% total yield.
—
another molecule of RLi to the double C N. Addition of
—
N-nitrosiminium ion is a reversible reaction, as shown in
Scheme 1.[26] Denitrosation of the N-nitrosiminium intermediate
(giving 5) and simultaneous reduction could be the route for
formation of side products 4 and 5.
—
—
organolithium to C N bond is less frequent than C O bond,
—
—
though it is a known reaction.
Nevertheless, in the present conditions, addition of a second
—
—
molecule of RLi to C N is much more difficult than to N O. It
—
—
takes much longer reaction times and never goes to completion.
Addition of fresh RLi after the reaction has advanced produces
some additional 3, but not complete conversion of 2 could be
afforded, even using relatively large excess of RLi. Similar results
were recently reported for the addition of a-lithio methoxyallene to
some hydrazones; surprisingly, no reaction was observed by using
2 M equivalents of the RLi. 6 M equivalents were required to engage
completely the substrate, and variable amounts of hydrazone were
recovered when intermediate quantities were used (no explanation
was given).[19] These observations together with the fact that the
kinetics of the reaction becomes much slower after a certain
reaction time (see Fig. 1) suggest that some RLi could be involved in
—
Addition of RLi to the N O bond
—
—
Though the reaction of organolithiums with the N O bond is
—
much less known than the synthetically very useful addition of
—
RLi to C O bonds, the efficient formation of hydrazones shown
—
in the present study can be interpreted by an addition
mechanism.
RLi reagents are known to have some reducing properties and
one of the undesirable side reactions that usually occur in their
additions to simple carbonyl compounds, is the production of
alcohols and even hydrocarbons, from the reduction of the
starting carbonyl substrate (probably through the elements of
lithium hydride). Therefore, under the present conditions, the
appearance of 6 in low variable yield could, in principle, be
explained as the product resulting from of a hypothetic reduction
of the hydrazones 2. Formation of 3 could be a simple addition of
—
complexation of the N C bond, or in another type of more
—
complex equilibria as those previously observed in the insertion of
CO into the N—Li bond of lithium amides.[21,24,27]
One of the first studies of reactions of organolithiums with
nitrosamines reported variable products depending on the
substituents in the nitrosamine and the nature of the organo-
lithium, in most cases the yields were around 30% or below.[28] It
was suggested that azomethine intermediates are formed and
subsequently dimerize head to tail to form the isolated
sym-hexahydrotetrazines. The authors’ attempts to synthesize
hydrazones were fully unsuccessful, and only in one case a
trialkylhydrazine could be obtained. Thus, when N-methyl-tert-
butylnitrosamine was treated with tert-butyllithium and quenched
with ethanol 1,2-tert-butyl-1-(ethoxymethyl)hydrazine was
formed.[28] This compound did not eliminate ethanol to form
the sym-hexahydrotetrazine (as it was observed in the other cases),
probably due to steric hindrance of the tert-butyl groups.
Nevertheless, under the conditions of the present study,
hydrazones could be isolated in good yields. Azomethines could
also be suggested as likely intermediates in a complex reaction
scheme as outlined in Scheme 2, which is consistent with the
Scheme 1.
J. Phys. Org. Chem. 2008, 21 1098–1104
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
A. J. VAZQUEZ, C. RODRIGUEZ AND N. S. NUDELMAN
´
Scheme 2.
observed results for the formation of products 2, 3, and 6.
Reactions of nitrosamines with Grignard reagents were previously
studied, but the yields were not good, in some cases complex
mixtures (as least 8 products) or tar products were obtained, even
when the reactions were carried out at low temperatures; better
results were obtained with organolithiums, but only in one case a
good yield (80%) of the corresponding hydrazone was
obtained.[23]
An alternative approach
The procedure herewith described, could be combined with the
previously reported synthesis of N-nitrosamines from lithium
amides and NO,[18] affording a very convenient and efficient
tandem sequence for the synthesis of substituted hydrazones
and trialkyl hydrazines from amines as illustrated schematically in
Eqn (5).
Searching for an additional evidence for the suggested global
mechanism, an alternative approach was examined. The
N-dicyclohexylnitrosamine, 1, was prepared by the NO insertion
into the N—Li bonds of lithium N,N-dicyclohexylamide (as shown
in the first step of Eqn (5)) by the procedure described in the
experimental section, that afforded a quantitative conversion of
the lithium amide to the corresponding nitrosamine. Thus, the
reaction of lithium N,N-dicyclohexylamide with NO was carried
out at ꢀ78 8C, at atmospheric pressure, when the absorption of
NO ceased, BuLi (in 1–1.3 molar ratio) was added and 98% of the
expected 2c was obtained. Nevertheless, in a second approach,
conditions were changed in this way; after the NO absorption was
complete, the excess of NO was removed by vacuum out from the
reaction flask, before adding 1–1.2 equivalents of n-BuLi. Under
these conditions, dicyclohexylamine, 4, was obtained in yields
higher than 20%, and the yields of 2c diminished significantly. If 5
equivalent of BuLi is added under these conditions, (to produce
the corresponding 3c), the amount of 4 amounts to 26%, after 2 h
reaction, and the trialkyhydrazine is the only main product. This
approach was also tested by using the more sterically hindered
lithium cyclohexylisopropylamide in its reaction with NO. Under
The first step in the Scheme 2 involves attack by the
organolithium reagent on the nitroso moiety to give adduct 8
which by elimination of the elements of HOꢀ Liþ produces the
hydrazone 2c. In smaller amount, 8 could undergo similar
elimination but involving the proton of the a-carbon of one of the
cyclohexyl moiety, producing the azomethine 10, (presumably
through a diazenium salt 9). In the presence of tiny amounts of
LiH (an undesirable side product usually formed in the reactions
of organolithium compounds) intermediate 10 could undergo
reduction of the double bond forming 11 that on hydrolysis
would produce the hydrazine 6 which was isolated in small
—
—
amounts. By addition of a second molecule of RLi to the C
N
bond the adduct 12 is formed, which after quenching produces
the trialkylhydrazines 3. Under these conditions, dimerization of
the azomethine is avoided and several trialkylhydrazines could be
obtained in good yields. As can be observed, some trialkylhy-
drazines exhibit chiral carbon; though the purpose of this paper is
not synthetic, it is worthwhile to point out that, chiral hydrazines
could likely be obtained by carrying our the reaction in the
presence of a suitable chiral auxiliary. Examination of the
potential utility of this new reaction is under progress.
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2008, 21 1098–1104

NITROSODICYCLOHEXYLAMINE WITH ORGANOLITHIUMS
Table 4. Reaction of NO with R1R2NLi in THF, followed by in situ reaction with n-BuLia
1
2
R1R2NLi
BuLi (equivalent)
R1R2NH
% R1R2NNO
yields R R NN CHC H
Hydrazine
—
—
3
7
R1 ¼ R2 ¼ Cy
1.1
5
1.3
1.3
1.3
21
26
35
31
30
34
4
36
39
44
44
5
23
26
19
—
50 (3c)
—
—
—
R1 ¼ R2 ¼ Cy
R1 ¼ i-Pr, R2 ¼ Cy
R1 ¼ i-Pr, R2 ¼ Cy
R1 ¼ i-Pr, R2 ¼ Cy
a Reaction with NO at temperatures in the range from ꢀ19 to –0 8C, and atmospheric pressure. The excess of NO was removed by
vacuum before adding BuLi at room temperature.
regular conditions, the reaction of lithium cyclohexylisopropy-
lamide with NO at room temperature rendered cyclohexyliso-
propylnitrosamine in 96% yield.[18] Nevertheless, it can be
observed in Table 4, that by removing the excess of NO before
adding the RLi, the yield of cyclohexylpropylamine was higher
than 30% in the three cases, while the yields of the corresponding
nitrosamine and hydrazone diminishes significantly. The reac-
tions were repeated twice or thrice and it was observed that the
yield of amine was somehow dependent on the amount of NO
remaining in the reaction flask. This result is consistent with the
equilibrium for the nitrosamine denitrosation, that is shown in
Scheme 1.
Examination of a tandem sequence for a clean production of
trialkylhydrazines, as shown in Eqn (5), is under progress. This
new procedure would avoid the extra steps required by previous
methods that use protecting groups and the selective cleavage of
those for the synthesis of multisubstituted hydrazines.
well-ventilated hood. Contaminated materials were disposed in
special containers for further disposition of the still potentially
contaminated materials.
General remarks
All reactions involving organolithium reagents were carried out
by standard techniques for the manipulation of air- and
water-sensitive compounds, previously described.[29] The pro-
duct identification was carried out by melting point (when
available) and spectroscopic data. Product quantification was
carried out by using a 5890 Series II Plus Hewlett-Packard gas
chromatograph (equipped with a DB-5 column), at 70–280 8C
programmed temperature. Mass spectra were recorded on a BG
Trio-2 spectrometer. 1H- and 13C-NMR spectra were recorded in a
Brucker 200 and in a Brucker 500 MHz NMR spectrometers.
CONCLUSIONS
Solvents and reagents
The study of the reaction of dialkylnitrosamines with organo-
lithium reagents allowed the development of a simple route for
the synthesis of substituted hydrazones, and for the synthesis of
multisubstituted hydrazines. Trialkyl hydrazines have acquired
renewed interest for their wide versatile properties, and they can
be formed by the one-pot one-step reaction studied in this paper.
Some kinetic determinations, and the identification of lateral
products carefully isolated in small amounts, suggested likely
mechanistic routes that were confirmed by an alternative
approach preparing the starting nitrosamines by the reaction
of NO with lithium amides. If the excess of NO is removed before
adding the organolithium, significant amounts of the corre-
sponding dialkyl amine are determined in the reaction mixture,
thus confirming that extensive denitrosation occurs under these
conditions.
Hexane and THF (HPLC grade) were distilled from blue solutions
of sodium-benzophenone ketyl immediately before using. n-BuLi
(1 M in n-hexane solution) was synthesized as previously
described, starting from n-butyl-chloride.[22] n-PrLi and n-pentyl
Li (0.5 M hexane solutions) were prepared similarly to n-BuLi,
starting from the respective alkylbromides. Solid MeLi was
prepared from metal-halogen exchange, starting from n-BuLi
(5 mmol) and 0.62 ml of MeI (5 mmol) was added in several
aliquots, at 0 8C. The precipitated MeLi was centrifugated, the
solution was removed, and the white crystals were washed thrice
with 3 ml of anhydrous hexane followed by centrifugation each
time. The resulting solid was dried under vacuum at room
temperature, and dissolved in THF immediately before to use.
Solid PhLi was prepared from metal-halogen exchange, by the
same procedure described for MeLi, 1.2 ml of PhI (5.25 mmol)
replacing the MeI. All organolithiums reagents were titrated
with the double titration method, previously described.[30] N-
nitrosodicyclohexylamine was synthesized by the general
procedure developed by Zolfigol and cow, based on the in situ
generation of NOCl by the reaction of ZnCl2 and NaNO2,
in CH2Cl2, in the presence of wet SiO2.[30] It was purified by
crystallization from ethanol–water and washed with cold ethanol
(mp ¼ 104–105 8C).[18] The compound was kept under vacuum
overnight prior to use.
EXPERIMENTAL SECTION
WARNING: Most N-nitrosamines are powerful direct acting
carcinogens, they have to be handled and disposed with special
care avoiding skin contact. Manipulations were carried out using
frequently changed double pairs of disposable gloves and in a
J. Phys. Org. Chem. 2008, 21 1098–1104
Copyright ß 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

´
A. J. VAZQUEZ, C. RODRIGUEZ AND N. S. NUDELMAN
´
Reactions of N-nitrosodicyclohexylamine with
organolithium compounds
Acknowledgements
AV and CR are grateful recipients of the National Research Council
(CONICET) fellowships. The authors are indebted to the United
Nations Environmental Program (PNUD/ARG Proy 02/18, subproy
BC-53) and to the CONICET (PIP 6019) for financial support.
Synthesis of hydrazones 2: 210 mg (1 mmol.) of N-
nitrosodicyclohexylamine, 1, was put in a 10 ml round bottomed
flask, equipped with a magnetic stirrer and protected from light
by aluminum paper. The flask was tapped with a rubber septum,
it was placed in a bath at the desired temperature, and then
evacuated and filled with dry N2, alternatively several times.
Anhydrous THF of 2 ml of was added to dissolve the nitrosamine.
Then the organolithium solution (1.05–1.2 mmol in n-hexane)
was added by syringe with vigorous stirring, and allowed to react
for nearly 5 min. When MeLi is used, the reaction time should be
approximately 2 h. Then, the flask was placed in a water–ice bath,
and distilled methanol (ca. 0.15 ml) was syringed. The solvent of
the resulting mixture was distilled at reduced pressure, obtaining
an oil. Upon purification by silica-gel column, pure hydrazones
are obtained as pale yellow oils. For quantification, the crude
mixture is dissolved in CH2Cl2 and filtered, prior to CG-analysis.
Synthesis of hydrazines 3 and 7: the procedure is similar to the
hydrazone synthesis, but using 4–5 equivalents of organolithium
reagent, and allowing a reaction time for 2 h at room temperature.
When two different organolithiums are used, the first one is added
in a small excess (1–1.2 mmol, as previously described for
hydrazone synthesis), and the second (3–5 equivalents) is added
later (after 5 min), allowing to react for 2 h and then distilled
methanol (0.75 ml) is syringed. Upon purification by chromato-
graphic column, pure hydrazines are obtained as air-sensitive pale
yellow oils. For quantification, the crude mixture is dissolved
in CH2Cl2 and filtered, prior to CG-analysis.
Kinetic study of the reaction of 1 with n-BuLi: The reaction of 1
with BuLi was carried out at three working temperatures (0 8C,
12 8C and room temperature), following the conditions described
for the hydrazine synthesis ([n-BuLi]:[1] ¼ 3). Aliquots of ca. 0.5 ml
were withdrawn by a syringe at different time intervals, and
quenched with distilled MeOH (30 ml) in small vials, that had been
previously evacuated and purged with N2. The resultant solutions
were analyzed by GC.
Reaction of lithium dialkylamides with NO. Typical reaction
conditions are described for lithium dicyclohexylamide. A
round-bottomed reaction flask containing a teflon-coated stirring
bar and capped with a no-air stopper was evacuated and filled
with dry nitrogen alternatively several times, and then nitric oxide
was added at ca. 1013 mbar. After that, a solution of lithium
dicyclohexylamide (1 mmol) in THF (1 ml) was added with
vigorous magnetic stirring, for 3 h. The initial colorless solution
turned to orange at the beginning, and this color stayed along
the reaction. The reaction was worked out with 0.2 ml of distilled
methanol. Excess NO was removed and distilling the THF under
reduced pressure afforded slightly orange crystals of 1 in a
quantitative yield. Crystallization from acetone rendered white
crystals of m.p. 104.5–105–5 8C.
REFERENCES
[1] C. C. Abnet, Cancer Investig. 2007, 25, 189–196.
[2] R. Guza, M. Rajesh, Q. Fang, A. Pegg, N. Tretyakova, Chem. Res. Toxicol.
2006, 19, 531–538.
[3] M. Ikeda, K. Masumura, Y. Sakamoto, B. Wang, M. Nenoi, K. Sakuma, I.
Hayata, T. Nohmi, Mutat. Res. – Genet. Toxicol. Environ. Mutagen. 2007,
626, 15–25.
[4] S. Loft, P. Svoboda, H. Kasai, A. Tjonneland, P. Moller, M. Sorensen, K.
Overbad, H. Autrup, O. Raaschou-Nielsen, Int. J. Cancer 2007, 121,
1579–1584.
[5] P. Bonde, D. Gao, L. Chen, T. Miyashita, E. Montgomery, J. W. Harmon,
C. Wei, Ann. Thorac. Surg. 2007, 83, 433–440.
[6] C. A. M. Krul, M. J. Zeilmaker, R. C. Schothorst, R. Havenaar, Food Chem.
Toxicol. 2004, 42, 51–63.
[7] X. N. Zou, S. H. Lu, B. Liu, Int. J. Cancer 1994, 59, 155–158.
[8] X. Zou, J. Li, S. Lu, X. Song, X. Wang, L. Guo, J. Li, J. Ye, Chin. Med. Sci. J.
1992, 7, 201–204.
[9] W. Lijinsky, Mutat. Res. – Genet. Toxicol. Environ. Mutagen. 1999, 443,
129–138.
[10] A. Bredihhin, U. M. Groth, U. Ma¨eorg, Org. Lett. 2007, 9, 1097–
1109.
[11] U. Ragnarsson, Chem. Soc. Rev. 2001, 30, 205–213.
[12] R. Zhang, J. P. Durkin, W. T. Windsor, Bioorg. Med. Chem. Lett. 2002, 12,
1005–1008.
[13] M. M. Cherney, C. Huitema, J. Liu, K. E. James, J. C. Powers, L. D. Eltis,
M. N. James, J. Mol. Biol. 2005, 354, 1137–1151.
[14] N. Brosse, M. F. Pinto, B. Jamart-Gregoire, J. Org. Chem. 2000, 65,
4370–4374.
[15] N. Brosse, M. F. Pinto, J. Bodigue, B. Jamart-Gregoire, J. Org. Chem.
2001, 66, 2869–2873.
[16] A. Vazquez, R. Goldberg, N. S. Nudelman, Organolithiums as Useful
Synthetic Intermediates for Tandem Reactions in the Chemistry of
Functional Groups: Lithium Chemistry (Eds.: S. Patai, Z. Rappaport),
Wiley, Chichester, 2006, ch. 2, pp. 63–137.
[17] G. V. Garc´ıa, N. S. Nudelman, Org. Prep. Proc. Int. 2003, 35, 445–
500.
[18] N. S. Nudelman, A. E. Bonatti, Synlett 2000, 1825–1827.
´
[19] V. Breuil-Desvergnes, J. Gore, Tetrahedron 2001, 57, 1939–1950.
[20] M. Adler, M. Marsch, N. S. Nudelman, G. Boche, Angew. Chem, Int. Ed.
Engl. 1999, 38, 1261–1263.
˜
[21] N. S. Nudelman, H. Schulz, G. Garc´ıa Linares, A. E. Bonatti, Organo-
metallics 1998, 17, 146–150.
[22] N. S. Nudelman, G. Garc´ıa Linares, J. Org. Chem. 2000, 65, 1629–
˜
1635.
[23] C. J. Michejda, R. W. Schluenz, J. Org. Chem. 1973, 38, 2412–
2415.
[24] N. S. Nudelman, E. Lewcowikz, J. P. Furlong, J. Org. Chem. 1993, 58,
1847–1852.
[25] M. Rajama¨ki, A. Vigroux, L. Chahoua, J. C. Fishbein, J. Org. Chem. 1995,
60, 2324–2325.
[26] L. Chahoua, A. Vigroux, Y. Chiang, J. C. Fishbein, Can. J. Chem. 1999,
77, 1148–1161.
[27] N. S. Nudelman, Pure Appl. Chem. 1998, 70, 1939–1946.
[28] P. R. Farina, H. Tieckelmann, J. Org. Chem. 1973, 38, 4259–4263.
´
[29] A. J. Vazquez, N. S. Nudelman, ARKIVOC 2005, 332–340.
[30] M. A. Zolfigol, F. Shirini, A. G. Choghamarani, A. Taqian-Nasab, H.
SUPPORTING INFORMATION
All the compounds herewith described were fully characterized
by 1H- and 13C–NMR and MS spectroscopy. The data correspond-
ing to every compound are provided as supplementary material.
Keypour, S. Salehzadeh, J. Chem. Res. (S) 2000, 420–422.
www.interscience.wiley.com/journal/poc
Copyright ß 2008 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2008, 21 1098–1104