Alkylation of nitroaromatics with tetraalkylborate ion via electrochemical oxidation

Article abstract of DOI:10.1021/jo030158c

Aromatic nucleophilic substitution reaction (S_NAr) is one of the most thoroughly studied reactions. Alkylation of nitroaromatics with Grignard reagents via chemical oxidation of the σ^H-complexes is the most general method to introduc

Full text of DOI:10.1021/jo030158c

Alk yla tion of Nitr oa r om a tics w ith Tetr a a lk ylbor a te Ion via
Electr och em ica l Oxid a tion
Iluminada Gallardo,* Gonzalo Guirado, and J ordi Marquet
Departament de Qu´ımica, Universitat Auto´noma de Barcelona, E-08193 Bellaterra, Barcelona, Spain
iluminada.gallardo@uab.es
Received May 7, 2003
Aromatic nucleophilic substitution reaction (SNAr) is one of the most thoroughly studied reactions.
Alkylation of nitroaromatics with Grignard reagents via chemical oxidation of the σH-complexes is
the most general method to introduce an alkyl group into a nitroaromatic compound. This approach
has considerable drawbacks, especially when more than one nitro group are present in the aromatic
ring. In this article, we present an electrochemical approach, which offers a new very selective
methodology for obtaining alkyl polynitroaromatic compounds. Different strategies based on the
use of tetralkylborate anion as nucleophiles are used so as to increase efficiency and to reduce the
drawbacks associated with this reaction. A wide list of dinitro- and trinitro-aromatic compounds
are studied, the range of yields obtained being from fair (40%) to excellent (85%). The key to
improvement in the process is the use of electrochemical techniques for the oxidation of the mixture
σH-complexes/tetrabutylborate ion. The electroactive character of the nucleophile, which can be
oxidized to an alkyl radical, means that the SNAr of the hydrogen polar mechanism is not the only
mechanism operating during the electroxidation process, since the hydrogen radical SNAr mechanism
is running at the same time. Electrochemical mechanistic studies allow the participation of each
mechanism in the global product yield obtained to be quantified.
In tr od u ction
cases; consequently, it is also impossible to obtain the
substituted product.4
The main industrial process for obtaining alkyl ni-
troaromatic compounds is based on nitration. However,
this process is poorly selective and requires very drastic
conditions.1 These serious drawbacks are the main reason
new methods of alkylation of aromatic nitro compound
are of remarkable interest.2
The traditional and widely used approach for the
functionalization of aromatic compounds, which contain
electron-withdrawing groups, is the nucleophilic aromatic
substitution reaction (SNAr).3 However, lack of generality
and low yields, even when chemical oxidants are used,
linked with the fact that their use represents a significant
environmental hazard particularly when scaling-up,
make this approach difficult to apply in many cases.
Furthermore, the low oxidation potential usually shown
by chemicals oxidants and the high stability of the
σ-complexes makes its oxidation impossible in some
In recent years, we have been trying to develop a
general and environmentally favorable route for the
nucleophlic aromatic substitution reaction, avoiding the
use of chemical oxidants. It has recently been demon-
strated that the use of an electrode instead of a chemical,
in what we have called an electrochemical approach to
nucleophic aromatic substitution, allows the oxidation of
σ-complexes, leading to substitution products in good
yields, when cyanide, amine, amide, or enolate anions
are used as nucleophiles. Electrochemical oxidation of
σ-complexes can be considered a green process, since the
formation of the reduced form of the chemical oxidant is
avoided. Moreover, by using electrochemical techniques,
all of the potential range can be covered, thus allowing
the oxidation of all kinds of σ-complexes.5
We have reported that two general mechanisms take
place in the electrochemical SNAr reaction, the nucleo-
philic aromatic substitution of hydrogen (NASH) and the
nucleophilic aromatic substitution of heteroatom (NASX),
Scheme 1. In a first fast equilibrium step, the σH-complex
and/or the σX-complex are formed, depending on the
* To whom correspondence should be addressed. Fax: (+34) 93 581
2920.
(1) Franck, H.; Stadelhofer, J . W. Industrial Aromatic Chemistry.
Raw Materials, Processes, Products; Springer-Verlag: Berlin, Heidel-
berg, 1987.
(2) (a) Traynelis, V. J .; MacSweeney, J . Y. J . Org. Chem. 1966, 31,
243. (b) Russel, G. A.; Weiner, S. A. J . Org. Chem. 1966, 31, 248. (c)
Terrier, F.; Nucleophilic Aromatic Displacement; Feuer, H., Ed.;
VCH: New York, 1991. (d) Chupakhin, O. N.; Charushin, V. N.; van
der Plas, H. C. Nucleophilic Aromatic Sustitution of Hydrogen;
Academic Press: London, 1994.
(3) (a) Buncel, E.; Dust, J . M.; Terrier, F. Chem. Rev. 1995, 95, 2262.
(b) Makosza, M.; Winiarski, J . J . Org. Chem. 1980, 45, 1535. (c)
Makoska, M.; Winiarski, J . Acc. Chem. Res. 1987, 20, 282. (d) Makoska,
M.; Stalinski, K. Polish J . Chem. Res. 1999, 73, 282.
(4) (a) Bock, H.; J aculy, D. Angew. Chem., Int. Ed. Engl. 1984, 23,
305. (b) Huertas, I.; Gallardo, I.; Marquet, J . Tetrahedron Lett. 2001,
42, 3439.
(5) (a) Gallardo, I.; Guirado, G.; Marquet, J . Chem. Eur. J . 2001, 7,
1759. (b) Gallardo, I.; Guirado, G.; Marquet, J . ES2000/489. (c)
Gallardo, I.; Guirado, G.; Marquet, J . Eur. J . Org. Chem. 2002, 2, 251.
(d) Gallardo, I.; Guirado, G.; Marquet, J . Eur. J . Org. Chem. 2002, 2,
261. (e) Gallardo, I.; Guirado, G.; Marquet, J . J . Org. Chem. 2002, 67,
2548.
10.1021/jo030158c CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/26/2003
7334
J . Org. Chem. 2003, 68, 7334-7341

Alkylation of Nitroaromatics
SCHEME 1
SCHEME 2
We have recently reported the alkylation of nitroarenes
by the electrochemical oxidation of the σ-complexes
arising from the reaction of the aromatic compounds with
RLi or RMgX. Both the NASH and the NASX mecha-
nisms (Scheme 1) have been observed in these reactions.
The yields obtained are comparable to or better than
those previously reported by using chemical oxidants. In
addition and for the first time, electrochemical oxidation
allowed the NASH alkylation of dinitroderivates to be
obtained in reasonable yields. Despite the use of electro-
chemical techniques being very attractive in SNAr, cer-
tain problems still remain unsolved: (1) no alkyltrini-
trobenzene derivatives have been obtained in any case;
(2) selectivity control is difficult because of the high
reactivity of the nucleophiles; and (3) RLi and RMgX
require the use of high resistive solvents, which are not
appropriate for performing exhaustive electrolysis (i.e.,
long reaction times).11
selectivity of the reaction. In the NASH process, the
oxidation of the σH-complex then arises via a three-step
mechanism. The first step is a single-electron oxidation
of the σH-complex, followed by a H+ elimination. The
resulting radical anion could be oxidized in a second
single-electron process on the electrode or, in some cases,
the radical generated in the first oxidation can act as
oxidizing agent, leading formally to a disproportioning
process. The NASX mechanism consists of a one-electron
oxidation process and the subsequent elimination of the
heteroatom as a radical.5d,e
In 1970, Taylor reported that 1,3,5-trinitrobenzene
reacts with tetraalkylborates to form the corresponding
σH-complex (Scheme 2).12
The reaction of organolithium or organomagnesium
compounds with nitroaromatics would seem to be the first
choice in order to produce alkylnitroaromatics via SNAr.
However, the first such reaction, performed by Severin,
provided evidence for the difficulty of reaction control.6
Thus, Grignard reagents (RMgX, with R ) Me, Et, Bu)
reacts with 1,3,5-trinitrobenzene in THF yielded the
corresponding 1,3,5-trialkyl-2,4,6-trinitrocyclohexanes,
after acidification of the reaction mixture. Similar reac-
tivity was found for 1,3-dinitrobenzene and 1-chloro-2,4-
dinitrobenzene. Using nitrobenzene, the reaction leads
to a mixture of alkyl nitrobenzenes and alkyl nitrosoben-
zenes.7
The chemical oxidation (with Br2 or KMNO4) of σ-com-
plexes has been reported. This leads to a mixture of
reaction products, from which alkyl-substituted nitro-
compounds are obtained in low yields.8 The use of DDQ
as an oxidizing agent improves the selectivity of the
reaction, but the cost of DDQ and the difficulty of the
separation of the reaction mixture hampers the use of
this reagent in preparative experiments.9
The use of the tetralkylborate anion as nucleophile
constitutes an interesting alternative to organolithium
or organomagnesium reagents in the SNAr reaction.
However, there is no report based on this approach,
probably because the chemical oxidant’s power was
4
limited to 0.6 V vs SCE and all the trinitro σ-adducts
have higher oxidation potentials.5,11 This problem can be
solved by using electrochemical oxidation.5,11
Here we report on the use of tetralkylborates in the
SNAr reaction to form alkyl dinitro and trinitro aromatic
compounds. This approach increases not only the yield
but also the selectivity of the reaction, when compared
with RLi or RMgX alkylating agents. To establish the
mechanistic details and the synthetic scope of the elec-
trochemical method, this study was carried out for a wide
series of nitrobenzene derivatives and related compounds
1-7: nitrobenzene (1), 1,3-dinitrobenzene (2), 1,3,5-
trinitrobenzene (3), 1,3-dinitronaphthalene (4), 1-meth-
oxi-2,4,6-trinitrobenzene (5), 1-chloro-2,4,6-trinitroben-
zene (6), and 1-methyl-2,4,6-trinitrobenzene (7) (Chart
1).
Recently, a good yield synthesis of alkyl nitro benzenes,
via nucleophilic aromatic substitution, has been reported,
using p-dinitrobenzene and alkyl boranes as reagents in
t
t
the presence of BuOK in BuOH. A mechanism based
on the reaction of the alkyl radical and the radical anion
of p-dinitrobenzene has been proposed, although no
evidence was provided.10
Resu lts a n d Discu ssion
Essentially following Taylor’s conditions,12 the σH-
complexes were prepared by the addition of nitroarene
(20 mM) to a solution of Me4NBBu4 (from 0.05 to 0.2 M),
in acetone under a nitrogen atmosphere. Initially, a large
excess of tetra-n-butylborate anion was used to shift the
equilibrium toward the σH-complexes.5,11 After 2 h, a
solution of DMF/0.1 M Et4NBF4 was added before per-
(6) (a) Severin, T.; Schmitz, R. Chem. Ber. 1963, 96, 3081. (b)
Severin, T. Angew. Chem., Int. Ed. Engl. 1958, 70, 164. (c) Artamkina,
G. A.; Egorov, M. P.; Beletskaya, I. P. Chem. Rev. 1982, 82, 427.
(7) (a) Kienzle, F. Helv. Chim. Acta 1978, 61, 449. (b) Bartoli, G.;
Leardini, R.; Lelli, M.; Rosini, G. J . Chem. Soc., Perkin Trans. 1 1997,
884. (c) Bartoli, G.; Leardini, R.; Medici, A.; Rosini, G. J . Chem. Soc.,
Perkin Trans. 1 1998, 692.
(8) Bartoli, G.; Bosco, M.; Baccolini, G. J . Org. Chem. 1980, 45, 522.
(9) Makosza, M.; Surowiec, M. J . Organomet. Chem. 2001, 64, 167.
(10) Shifman, A.; Palani, N.; Hoz, S. Angew. Chem., Int. Ed. 2000,
39, 944.
(11) Gallardo, I.; Guirado, G.; Marquet, J . J . Org. Chem. 2003, 68,
631.
(12) Taylor, R. P. J . Org. Chem. 1970, 35, 3578.
J . Org. Chem, Vol. 68, No. 19, 2003 7335

Gallardo et al.
CHART 1
forming the electrochemical experiments. In the case
when the nitroarene was 1,3,5-trinitrobenzene (and the
acetone was removed before the addition of DMF/0.1 M
Et4NBF4 solution), typical voltammograms are depicted
in Figure 1. Figure 1a (starting with an anodic scan) and
Figure 1b (starting with a cathodic scan) show the elec-
trochemical behavior of tetrabutylborate ion and trini-
trobenzene solutions that have reacted, although not
completely, to form the corresponding σH-complex. The
third cyclic voltammogram (Figure 1c) shows the elec-
trochemical behavior of Me4NBBu4 in a DMF/0.1 M Et4-
NBF4 solution.
Starting with an anodic scan, the tetrabutylborate ion/
σH-complex mixture shows two oxidation peaks. The first
irreversible wave (Epa ) 0.35 V vs SCE) corresponds to
the oxidation of Me4NBBu4,11 and a second oxidation
wave (Epa ) 0.96 V vs SCE) arises from the oxidation of
the σH-complex (Figure 1a). In the following cathodic
scan, two peaks appear at -0.53 and -0.71 V. The first
corresponds to the reduction of 1,3,5-trinitrobenzene13 to
its radical anion, and the second corresponds to the reduc-
tion of the 1-butyl-2,4,6-trinitrobenzene to its radical
anion. The NASH product would be formed by oxidation
of the σH-complex during the previous anodic scan.
Starting with a cathodic scan (Figure 1b), the absence of
the cathodic peak at -0.71 V confirms that this peak
corresponds to an oxidation product formed during the
anodic scan. These results, linked with techniques of
chemical analysis (see below), undoubtedly indicate that
the oxidation of the σH-complex intermediate leads to the
above-mentioned alkyl NASH product (Scheme 3)
BBu4- exhibits a well-defined monoelectronic oxidation
wave at 0.35 V in DMF/0.1 M Et4NBF4 (Figure 1c).
Moreover, exhaustive electroxidation experiments of
F IGURE 1. (a) Cyclic voltammetry of 3 (20 mM) + 0.1 M Me4-
NBBu4 (2 h after addition) in DMF + 0.1 M Et4NBF4 at 10
°C. Scan rate 1.0 V s-1, glassy carbon disk electrode (0.5 mm
diameter). Scan is in the potential range 0.00/1.30/-1.00/0.00.
(b) Cyclic voltammetry of 3 (20 mM) + 0.1 M Me4NBBu4 (2 h
after addition) in DMF + 0.1 M Et4NBF4 at 10 °C. Scan rate
1.0 V s-1, glassy carbon disk electrode (0.5 mm diameter). Scan
is in the potential range 0.00/-1.00/1.30/0.00. (c) Cyclic vol-
tammetry of Me4NBBu4 (7.5 mM) in DMF + 0.1 M Et4NBF4
at 10 °C. Scan rate 1.0 V s-1, glassy carbon disk electrode (0.5
mm diameter)
SCHEME 3
-
14
BBu4 in ACN/0.1 M Et4NBF4 show a charge release
of 1 F, indicating the formation of n-Bu•. Spin electron
trapping experiments on the electroxidation of organobo-
rates have corroborated the presence of free-radical
intermediates.14,15 The oxidation is somewhat complicated
depending on the solvent, through the occurrence of
several consecutive reactions (Scheme 4).
(13) Gallardo, I.; Guirado, G.; Marquet, J . Chem. Commun. 2002,
2638.
(14) Bancroft, E. E.; Blount, H. N.; J anzen, E. G. J . Am. Chem. Soc.
1979, 101, 3692.
(15) (a) Bard, A. J .; Gilbert, J . C.; Goodin, R. D. J . Am. Chem. Soc.
1974, 96, 629. (b) Bard, A. J .; Kraeutler, B.; J aeger, C. D. J . Am. Chem.
Soc. 1978, 100, 4903.
It was therefore advisable to ascertain the electro-
chemistry of tetrabutylborate ion BBu4 in our work
-
7336 J . Org. Chem., Vol. 68, No. 19, 2003

Alkylation of Nitroaromatics
SCHEME 4
reaction, but overly large excess leads to the formation
of dialkylated substitution product as a result of the com-
petition between the nitroaromatic and monosubstitution
product, particularly at the end of the reaction when the
concentration of the unreacted nitroarene becomes low.
It should be noticed that in nearly all cases the yield
obtained is higher than the percent of σ-complex formed.
In summary, moderate-to-excellent yields (30-75%)
were obtained with high selectivity in the monoalkyl
product, since only minor amounts of disubstituted
products were found in certain cases (below 15%) as a
secondary product. The chemoselectivity obtained when
1-chloro-2,4,6-trinitrobenzene was used is also remark-
able (Table 1, entry 6), since NASX products were not
detected. Thus, the use of BBu4- instead of RLi or RMgX
reagents provides a higher yield and a more selective
synthetic route for alkylnitroderivatives.
SCHEME 5
conditions. Cyclic voltammetry and electrolysis experi-
ments in ACN, DMF, acetone, and mixtures ACN/acetone
or DMF/acetone confirm the oxidation mechanism of the
The main limitation of the synthetic method arises
from the initial thermal process, in which the concentra-
tion of σH-complex is determined. Following the litera-
ture,13,16 the reaction was carried out in several solvents
to test the best experimental conditions for forming the
maximum amount of σ-complexes. ACN was chosen as
the most appropriate solvent in order to favor their
formation. Table 2 summarizes the results in this solvent.
It is important to notice that in ACN the percentage
of σH-complex formed increases considerably (more than
50% in certain cases). In the cases of 1-chloro-2,4,6-
trinitrobenzene and 2,4,6-trinitrotoluene (Table 2, entries
2 and 3) the reaction is fully selective. An exception is
found in the case of 1,3,5-trinitrobenzene, where a
disubtituted product appears (Table 2, entry 1). However,
this fact can be used to obtain the dialkylated compound
in up to 36% yield, by using a large excess of BBu4- anion.
The NASH mechanism (Scheme 1) or SNAr polar
hydrogen mechanism requires an oxidation potential
higher than that of σH-complexes. However, we have
observed that the formation of σ-complexes increases
considerably by working at potentials lower than that
σ-complexes but high enough to oxidize the BBu4- anion.
This is consistent with an alternative mechanism for this
reaction. We have undertaken an electrochemical study
to elucidate this mechanism and to establish its contribu-
tion to the yield of the substitution product obtained.
Cyclic voltammetry of the tetrabutylborate ion/σH-
complex mixture shows, after electrolysis passing 0.5 F
at 0.6 V vs SCE, that the oxidation peak corresponding
to the tetrabutylborate ion diminishes, whereas the
oxidation peak of the σH-complex increases considerably
(Figures 2a,b vs Figure 1a,b). When the first scan is
cathodic (Figure 2b), it can be observed that the quantity
of unreacted 1,3,5-trinitrobenzene has diminished and in
the following anodic scan the concentration of tetrabu-
tylborate ion has been reduced to half, the concentration
of the corresponding σH-complex having considerably
increased in these conditions. Only a low amount (<10%)
of NASH product is found. Similar behavior is obtained
when the first scan is anodic (Figure 2a). However, in
this case the oxidation of the peak at 0.96 V leads to
1-butyl-2,4,6-trinitrobenzene, as is shown by an incre-
ment of the peak current at -0.71 V during a cathodic
BBu4-. BBu4 (in the range 0.1-0.2 M) shows a mono-
-
electronic oxidation wave at Epa, anodic peak potential,
from 0.35 to 0.52 V depending on the solvent. These
values are in the range of some σH-complexes. Thus, in
certain cases, the σH-complex oxidation waves may be
hidden inside the oxidation wave corresponding to the
-
large excess of BBu4 anion. Only a shift of the anodic
peak potential values indicates the presence of the σH-
complex. An accurate determination of the oxidation
potential of the σH-complexes can be achieved by cyclic
voltammetry experiments of solutions with low molar
ratio BBu4-/nitroarene. Furthermore, by the same tech-
nique, it is also possible to determine the percentage of
nitroaromatic converted to σH-complex by measuring the
cathodic current value corresponding to the reduction of
the nitroaromatic reactant.
5,11
On the other hand, the
high reactivity of n-Bu• radical is responsible for the
presence of nondesirable products in the reaction mixture
because the radical reacts with the ketone leading to the
ketone radical, which attacks nitroaromatics, leading to
the different substitution products (Scheme 5).
By using 1,3,5-trinitrobenzene and a 2.5-fold excess of
BBu4 in DMF/acetone mixtures, the σH-complex forms
-
in 45% yield. This complex shows an oxidation wave at
0.96 V vs SCE. After exhaustive electrolysis at 1.06 V vs
SCE, 1-butyl-2,4,6-trinitrobenzene (NASH product) is
obtained in 42% yield as major product and the reactant
is recovered in 43% yield. 2,4,6-Trinitrobenzyl-methyl-
cetone 5% and 1,3-dibutyl-2,4,6-trinitrobenzene 5% were
obtained as the minor products. We observe that the
alkyl-trinitrobenzene was obtained in excellent yield with
respect to the σH-complex available in solution (45%); the
limitation of the method therefore seems to be the
relatively low amount of the σH-complex formed.
A second minor problem is the formation of the ketone
substitution products. To avoid this problem, the acetone
was removed by a continuous flow of dry nitrogen, after
the reaction of nitroaromatic and the tetrabutylborate
ion reaches equilibrium (ca. 2 h). The results of the
electroxidation methodology carried out in acetone-free
DMF solution are illustrated in Table 1. The tetrabu-
tylborate ion/nitroarene molar ratio reported was opti-
mized so as to obtain the best yield in the NASH product.
In principle, an excess of the nucleophile favors the
formation of the σH-complex, increasing the yield of the
(16) Atkins, J .; Gold, V.; Wassef, W. R. J . Chem. Soc., Perkin Trans.
2 1983, 1197.
J . Org. Chem, Vol. 68, No. 19, 2003 7337

Gallardo et al.
-
TABLE 1. NASH in Nitr oa r en es by Action of BBu 4 An ion (Aceton e/DMF ), via Electr och em ica l Oxid a tion
a
b
In initial conditions before performing the coulometry. The σ-complexes were carefully prepared by addition of the nucleophile to
solutions of the nitroarene 20 mM in acetone under inert atmosphere (see Experimental Section). The acetone was removed before
performing the coulometry. c Graphite working electrode. The oxidation products were analyzed by cyclic voltammetry, gas chroma-
d
tography/mass spectroscopy, and H NMR. e Determination of characteristic Epa value; the oxidation wave is not hidden inside the oxidation
1
-
peak of the BBu4
.
scan. After consumption of 1 F at 0.6 V vs SCE (Figure
2c, first scan solid line, second scan dotted line), the
results obtained are in total agreement with that previ-
ous explained. It is important to highlight that (1) the
tetrabutylborate ion has practically disappeared, (2) the
σH-complex is the major ”stable“ product present at the
end of the process, and (3) no 1-butyl-2,4,6-trinitroben-
zene (NASH product) is found as a main product (<10%)
at the end of the process. After exhaustive electrolysis
(2 F at 1.3 V vs SCE), Figure 2d shows two reduction
waves that can be attributed to 1-butyl-2,4,6-trinitroben-
zene and the dialkylated product. Scheme 6 summarizes
the overall process.
The controlled potential electrolysis performed at 0.6
V leads to the formation of n-Bu• radicals. The n-Bu•
radicals attack the unreacted trinitrobenzene via a
radical mechanism. The resulting radical evolves, by
elimination of a H+, as a result of the C-H acidity of
cyclohexadienyl radicals, leading to a nitroaromatic radi-
cal anion. This is then oxidized by the cyclohexadienyl
radical (according to the standard potentials of redox
couples, this is a disproportion step).17 This reaction is
7338 J . Org. Chem., Vol. 68, No. 19, 2003

Alkylation of Nitroaromatics
TABLE 2. NASH in Nitr oa r en es by Action of BBu 4 An ion (ACN/DMF ), via Electr och em ica l Oxid a tion
-
a
b
In initial conditions before performing the coulometry. The σ-complexes were carefully prepared by addition of the nucleophile to
solutions of the nitroarene 20 mM in ACN under inert atmosphere (see Experimental Section). The ACN was removed before to perform
the coulometry. c Graphite working electrode. The oxidation products were analyzed by cyclic voltammetry, gas chromatography/mass
d
spectroscopy, and 1H NMR. e Determination of characteristic Epa value; the oxidation wave is not hidden inside the oxidation peak of the
-
BBu4
.
similar to the termination step in the chain process SRN
1
Con clu sion s
aromatic substitution.10,18 The final products obtained
were NASH product (minor product) and the σH-complex
(major product) (Figure 2a,b). The electrolysis experi-
ments show the effects of the presence of n-Bu• radicals
(Table 2, entry 1), since thermal reactions yielding 48%
of σH-complex are observed (Figure 1a,b), but after the
oxidation of the BBu4- the percentage of σH-complex rises
to 85% (Figure 2c). Therefore, the presence of n-Bu•
radical significantly shifts the equilibrium to the σH-
complex. The n-Bu• radical can also attack the NASH
product, this process being more important at the end of
the reaction, because the concentration of the alkylated
aromatic compound is higher than that of the nitroaro-
matic. This fact is shown in Table 2 (entry 1), where a
large percentage of dialkylated products are found (36%).
Performing electrochemical oxidation, alkylnitroben-
zenes can be synthesized using RLi or RMgCl reagents,
via the NASH and NASX mechanism, and although
alkyldinitro- and trinitrobenzenes are not achieved using
RLi or RMgCl reagents, these compounds can be syn-
thesized using Me4NBR4 salts, via the NASH mechanism.
The BR4- nucleophile can easily be activated by electro-
chemical oxidation leading to n-Bu• radicals. This reactive
species allows improvement in the yields of NASH
products as well as reduction from days to hours in the
time required for synthetic procedures.
Finally, a SNAr of hydrogen polar mechanism and a
SNAr of hydrogen radical mechanism operate at the same
time during the electroxidation process. The participation
of each mechanism in the global product yield can thus
be established.
In summary, two kinds of SNAr mechanisms operate
concurrently in this reaction, a polar mechanism, more
effective at adequate potentials to oxidize σH-complexes
(and consequently to oxidize BBu4- anion; Table 1, entry
2) and a radical mechanism, more effective at adequate
potentials to oxidize BBu4- anion (and not oxidizing σH-
complexes; Table 2, entry 1). According to the results
(Table 2, entry 1), the SNAr radical mechanism can
contribute in a similar way as the SNAr polar mechanism
to the overall reaction yield.
Exp er im en ta l Section
Gen er a l Rem a r k s. Electr och em ica l Mea su r em en ts.
The electrochemical cell and measurement procedures for
cyclic voltammetry have previously been described.14 All
potentials are reported vs an aqueous saturated calomel
electrode. A glassy carbon disk was used as the working
electrode (0.5 mm diameter). Electrolyses were carried out
using a PAR 273A potentiostat. A graphite rod was used as
the working electrode.
(17) (a) Amatore, C.; Saveant, J . M. J . Electroanal. Chem. 1979, 21,
102. (b) Amatore, C.; Gareil, M.; Save´ant, J . M. J . Electroanal. Chem.
1983, 147, 1.
(18) (a) Amatore, C.; Pinson, J .; Save´ant, J . M.; Thie´bault, A. J . Am.
Chem. Soc.1981, 103, 6930. (b) Amatore, C.; Gareil, M.; Oturan, M.;
Pinson, J .; Save´ant, J . M.; Thie´bault, A. J . Org. Chem. 1986, 51, 3757.
Ma ter ia ls. 2,4,6-Trinitrotoluene (7) was from Union Es-
pan˜ola de Explosivos. All commercially available reactants
were of the highest possible purity and were used without
further purification.
J . Org. Chem, Vol. 68, No. 19, 2003 7339

Gallardo et al.
SCHEME 6
Gen er a l P r oced u r e for NASH Descr ibed in Ta ble 1. A
solution of tetramethylammonium tetrabutylborate (Me4-
NBBu4) (0.1, 0.2 M in acetone, 1 mL) prepared under nitrogen
atmosphere was added to a solution of the nitroarene (20 mM)
in acetone (2 mL) also kept under a nitrogen atmosphere. The
corresponding σH-complex was prepared by careful addition
of the nucleophile to the solution of the nitroarene. After 2 h,
the acetone was removed by passing a dry nitrogen flow
through the reaction mixture.
Gen er a l P r oced u r e for NASH Descr ibed in Ta ble 2. A
solution of tetramethylammonium tetrabutylborate (Me4-
NBBu4) (0.05-0.1 M in ACN (1 mL)) prepared under a
nitrogen atmosphere was added to a solution of the nitroarene
(20 mM) in ACN (2 mL) also kept under a nitrogen atmo-
sphere. The corresponding σH-complex was prepared by careful
addition of the nucleophile to the solution of the nitroarene.
After 2 h, the ACN was removed by passing dry nitrogen flow
through the reaction mixture.
Gen er a l Electr och em ica l P r oced u r e for NASH De-
scr ibed in Ta bles 1 a n d 2. A 0.10 g portion of Et4NBF4
dissolved in 5 mL of DMF was added to the solid residue
obtained after the evaporation of acetone or ACN. The oxida-
tion peak potentials of the σH-complexes were measured by
cyclic voltammetry. Electrolysis was then carried out at
potentials ca. 100 mV more positive than the value measured
for each σH-complex, using a graphite rod as a working
electrode.
F IGURE 2. (a) Cyclic voltammetry (after electrolysis at 0.6
V vs SCE passing 0.5 F) of 3 (20 mM) + 0.1 M Me4NBBu4 (2
h after addition) in DMF + 0.1 M Et4NBF4 at 10 °C. Scan rate
1.0 V s-1, glassy carbon disk electrode (0.5 mm diameter). Scan
is in the potential range 0.00/1.30/-1.00/0.00. (b) Cyclic
voltammetry (after electrolysis at 0.6 V vs SCE passing 0.5 F)
of 3 (20 mM) + 0.1 M Me4NBBu4 (2 h after addition) in DMF
+ 0.1 M Et4NBF4 at 10 °C. Scan rate 1.0 V s-1, glassy carbon
disk electrode (0.5 mm diameter). Scan is in the potential
range 0.00/-1.00/1.30/0.00. (c) Cyclic voltammetry (after elec-
trolysis at 0.6 V vs SCE passing 1 F) of 3 (20 mM) + 0.1 M
Me4NBBu4 (2 h after addition) in DMF + 0.1 M Et4NBF4 at
10 °C. Scan rate 1.0 V s-1, glassy carbon disk electrode (0.5
mm diameter). Scan is in the potential range 0.00/1.30/-1.00/
0.00 (dashed line) and 0.00/-1.00/1.30/0.00 (solid line). (d)
Cyclic voltammetry after electrolysis at 1.3 V vs SCE passing
2 F of solution Figure 2c. Scan rate 1.0 V s-1, glassy carbon
disk electrode (0.5 mm diameter). Scan is in the potential
range 0.00/-1.00/0.00.
The electrolysis was then stopped when the current inten-
sity dropped to nearly 0. The reaction mixture was dissolved
in toluene and extracted with water. The organic layer was
dried with Na2SO4, and the solvents were evaporated to leave
a residue that was analyzed by gas chromatography. The final
products were analyzed by gas chromatography/mass spec-
trometry, 1H NMR, and cyclic voltammetry. Product yields
were calculated by gas chromatography and cyclic voltamme-
try, after verifying from the 1H NMR spectrum of the crude
product that only the substitution products and starting
material were present.
Reaction P r odu cts. 4-Bu tyl-1,3-din itr oben zen e.22 (Table
Syn th esis of Sta r tin g Ma ter ia ls. 1-Methoxy-2,4,6-trini-
trobenzene (5) and 1-chloro-2,4,6-trinitrobenzene (6) were
synthesized, following described procedures.20,21
1
1, entry 1) H NMR (250 MHZ, CD3CN) δ 8.67 (d, J ) 2.33
Hz, 1H, Ar-H), 8.54 (dd, J ) 8.16, J ) 2.33 Hz, 1H, Ar-H),
7.75 (d, J ) 8.16 Hz, 1H, Ar-H), 2.75 (t, 2H, -CH2), 1.64-
1.71 (m, 2H, -CH2), 1.34-1.40 (m, 2H, -CH2), 0.93 (t, 3H,
-CH3); MS (70 eV) m/z (%) 224 (1) [M]+, 207 (9), 190 (26), 175
(12), 164 (95), 161 (35), 146 (35), 143 (30), 118 (27), 115 (28),
(19) Andrieux, C. P.; Larrumbre, D.; Gallardo, I. J . Electroanal.
Chem. 1991, 304, 241.
(20) Reese, C. B.; Pei-Zhuo, Z. J . Chem. Soc., Perkin Trans. 1 1993,
2291.
(21) Lam, K.-B.; Miller, J .; Samenho, P. J . J . Chem. Soc., Perkin
Trans. 2 1977, 457.
(22) Zou, G.; Falck, J . R. Tetrahedron Lett. 2001, 42, 5817.
7340 J . Org. Chem., Vol. 68, No. 19, 2003

Alkylation of Nitroaromatics
91 (36), 77 (43), 57 (22), 43 (100); MS (NCI, CH4) 224 (M-,
100), 225 (12), 195 (M- - NO, 3), 176 (M- - NO - NO, 91).
2-Bu tyl-1,3,5-tr in itr oben zen e.12,23,24 (Table 1, entry 2,
(s, 1H, Ar-H), 3.58 (t, J ) 6.35 Hz, 2H, -CH2), 1.52-1.75 (m,
2H, -CH2), 1.30-1.50 (m, 2H, -CH2), 0.93 (t, J ) 7.19 Hz,
3H, -CH3); MS (70 eV) m/z (%) 303 (1) [M]+, 304 (0.3), 285
(18), 244 (16), 213 (5), 168 (1), 99 (11), 87 (16), 75 (23), 74 (11),
59 (14), 57 (21), 43 (100).
1
Table 2, entry 1) H NMR (250 MHZ, CD3CN) δ 8.84 (s, 2H,
Ar-H), 3.62 (t, J ) 7.75 Hz, 2H, -CH2), 1.53-1.75 (m, 2H,
-CH2), 1.30-1.50 (m, 2H, -CH2), 0.93 (t, 3H, -CH3); MS (70
eV) m/z (%) - [M]+, 253 (3), 252 (33), 235 (8), 210 (29), 178
(6), 163 (5), 89 (20), 77 (23), 76 (26), 57 (30), 43 (100); MS (NCI,
CH4) 269 (M-, 100), 270 (M- + 1, 13), 239 (M- - NO, 78), 209
(M- - NO - NO, 65).
2,4-Dibu tyl-1,3,5-tr in itr oben zen e.30-32 (Table 1, entry 2;
1
Table 2, entry 1) H NMR (250 MHZ, CD3CN) δ 8.77 (s, 1H,
Ar-H), 3.60 (m, 4H, -CH2), 1.60-1.80 (m, 4H, -CH2), 1.35-
1.55 (m, 4H, -CH2), 0.93 (m, 6H, -CH3); MS (70 eV) m/z (%)
325 (4) [M]+, 309 (4), 237 (10), 210 (2), 128 (15), 115 (20), 91
(14), 89 (12), 77 (18), 71 (14), 63 (12), 57 (26), 43 (100); MS-
(NCI, CH4) 325 (M-, 100), 326 (M- + 1, 17), 295 (M- - NO,
13).
1
1-Bu tyl-2,4-d in itr on a p h th a len e.25 (Table 1, entry 3) H
NMR (250 MHZ, CD3CN) δ 9.19 (d, J ) 0.95 Hz, 1H, Ar-H),
8.45 (dd, J ) 7.50, J ) 1.35 Hz, 1H, Ar-H), 7.99 (m, J ) 8.50
Hz, J ) 6.85 Hz, J ) 0.85 Hz, 1H, Ar-H), 7.74 (m, J ) 8.50
Hz, J ) 1.35 Hz, 1H, Ar-H), 7,57 (m, J ) 7.50, J ) 6.85 Hz,
J ) 1.35 Hz, 1H, Ar-H), 3.45 (t, 2H, -CH2), 1.43-1.68 (m,
2H, -CH2), 1.29-1.42 (m, 2H, -CH2), 0.96 (t, 3H, -CH3); MS
(70 eV) m/z (%) 274 (35) [M]+, 275 (6), 228 (4), 201 (4), 196 (8),
185 (21), 196 (26), 157 (13), 140 (50), 139 (71), 128 (43), 115
(38), 101 (13), 89 (14), 71 (22), 57 (8), 43 (72), 41 (33).
2-Bu tyl-4-m eth oxy-1,3,5-tr in itr oben zen e.25 (Table 1, en-
try 4) 1H NMR (250 MHZ, CD3CN) δ 8.95 (s, 1H, Ar-H), 4.06
(s, 3H, -OCH3), 3.51 (t, J ) 6.50 Hz, 2H, -CH2), 1.53-1.73
(m, 2H, -CH2), 1.29-1.51 (m, 2H, -CH2), 0.97 (t, 3H, -CH3);
MS (70 eV) m/z (%) 299 (2) [M]+, 282 (46), 249 (5), 223 (5), 174
(3), 147 (4), 77 (21), 57 (24), 43 (100).
Su bp r od u cts Obta in ed (1-15%). 1,3-Dibu tyl-2,4-d in i-
tr on a p h th a len e. (Table 1, entry 3) This product could not
be isolated and was therefore tentatively assigned by GC/MS
spectroscopy. MS (70 eV) m/z (%) 330 (49) [M]+, 331 (5), 253
(12), 242 (15), 197 (14), 154 (22), 152 (32), 141 (27), 127 (26),
115 (30), 57 (10), 43 (100).
2,4-Dibu tyl-6-m eth yl-1,3,5-tr in itr oben zen e.31-33 (Table
1, entry 5) This product could not be isolated and was therefore
tentatively assigned by GC/MS spectroscopy. MS (70 eV) m/z
(%) 339 (5) [M]+, 323 (6), 322 (30), 280 (7), 266 (18), 210 (12),
143 (10), 128 (22), 115 (31), 91 (21), 57 (100), 43 (95), 41 (96).
2,4-Din itr op h en yl-a ceton e.5 1H NMR (250 MHz, CD3CN)
δ 8.98 (d, J ) 2.50 Hz, 1H), 8.45 (dd, J ) 8.45 Hz and J )
2.50 Hz, 1H), 7.65 (d, J ) 8.45 Hz, 1H), 4.41 (s, 2H), 2.29 (s,
3H)
2-Bu tyl-4-m eth yl-1,3,5-tr in itr oben zen e.26 (Table 1, entry
5 and Table 2, entry 3) 1H NMR (250 MHZ, CD3CN) δ 8.75 (s,
1H, Ar-H), 3.48 (t, J ) 6.35 Hz, 2H, -CH2), 2.36 (s, 3H, Ar-
CH3), 1.65-1.75 (m, 2H, -CH2), 1.38-1.50 (m, 2H, -CH2), 0.93
(t, J ) 7.19 Hz, 3H, -CH3); MS (70 eV) m/z (%) 283 (2) [M]+,
266 (25), 267 (4), 223 (25), 214 (2), 194 (4), 148 (7), 128 815),
115 (20), 103 (12), 90 (14), 89 (15), 78 (12), 77 (29), 57 (21), 43
(100).
2,4,6-Tr in itr op h en yl-a ceton e.11 1H NMR (250 MHz, CD3-
CN) δ 8.95 (s, 2H), 4.49 (s, 2H), 2.23 (s, 3H).
2,4-Din itr on a p h th yl-a ceton e.14 1H NMR (250 MHz, CD3-
CN) δ 8.61 (s, 1H), 8.10 (d, J ) 3.84 Hz, 1H), 8.10 (d, J ) 3.84
Hz, 1H), 8.04 (m, 1H), 7.79 (d, J ) 7.37 Hz, 1H), 7.71 (m, 1H),
4.04 (s, 2H), 2.12 (s, 3H).
3-Bu tyl-1-ch lor o-2,4,6-tr in itr oben zen e.27-29 (Table 1, en-
try 6 and Table 2, entry 2) 1H NMR (250 MHZ, CD3CN) δ 8.75
Ack n ow led gm en t. We gratefully acknowledge fi-
nancial support from the DGI (Spanish MCyT) through
project BQU2000-0336 and from the Generalitat de
Catalunya through project 2001SGR00180.
(23) Zhang, N.; Yu, B.; Qin, R.; Xie, B.; Zhong, D.; Gong, Z.; Gu, Q.;
J in, X.; Wang, Y. Nucl. Sci. Tech. 1991, 2, 60.
(24) Schroeder, G.; Leska, B.; Przybyl, J .; J arczewski, A. Polish J .
Chem. 1993, 67, 1131.
J O030158C
(25) Sergeev, A. M.; Nurmukhametov, R. N.; Kaminskii, A. Ya. Dokl.
Akad. Nauk SSSR 1986, 291, 399.
(26) Chastrette, M.; Zakarya, D.; Elmouaffek, A. Eur. J . Med. Chem.
1986, 21, 505.
(27) Wright, H. Explosivstoffe 1968, 16, 176.
(28) Sharnin, G. P.; Buzykin, B. I.; Moisak, I. E. Zh. Org. Khim.
1969, 5, 182.
(29) Sharnin, G. P.; Buzykin, B. I.; Nurgatin, V. V.; Moisak, I. E.
Zh. Org. Khim. 1967, 3, 82.
(30) J etti, R. K. R.; Thallapally, P. K.; Xue, F.; Mak, T. C. W.;
Nangia, A. Tetrahedron 2000, 56, 6707.
(31) Lewin, A. H.; Ramesh, U. J . Labelled Compd. Radiopharm.
1995, 36, 549-52.
(32) Ryzhova, G. L.; Rubtsova, T. A.; Vasil’eva, N. V. Primen. Mol.
Spektrosk. Khim., Sb. Dokl. Sib. Soveshch., 3rd; Krasnoyarsk: USSR,
1966; Meeting Date 1964; pp 78-81.
(33) Bubnov, Yu. N.; Frolov, S. I.; Kiselev, V. G.; Mikhailov, B. M.
Zh. Obsh. Khim. 1970, 40, 1316.
J . Org. Chem, Vol. 68, No. 19, 2003 7341