Revisiting the Synthesis of 4,6-Difluorobenzofuroxan: A Study of Its Reactivity and Access to Fluorinated Quinoxaline Oxides

FULL PAPER

DOI: 10.1002/ejoc.201402692

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan: A Study of Its Reactivity

and Access to Fluorinated Quinoxaline Oxides

Cyril Jovené,^[a]Morgane Jacquet,^[a]Jérome Marrot,^[a]Flavien Bourdreux,^[a]

Mikhail E. Kletsky,^[b]Oleg N. Burov,^[b]Anne-Marie Gonçalves,^[a]and Régis Goumont*^[a]

Keywords: Synthetic methods / Nucleophilic substitution / Nitrogen heterocycles / Fluorine / Drug discovery

New quinoxaline 1,4-dioxide derivatives have been synthe-

reactions. The first case of a reaction involving a nitro-substi-

sized from novel fluorinated benzofuroxans such as 4-fluoro- tuted benzoquinone 2-diazide and cyclopentadiene is also

benzofuroxan, which is prepared for the first time. Further-

more, the preparation 4,6-difluorobenzofuroxan has been re-

visited because we were unable to reproduce the reported

synthetic method. Several synthetic pathways have thus

been investigated, and the optimal way to prepare this disub-

stituted benzofuroxan was from the 3,5-difluoro-2-nitroanil-

ine. The various synthetic attempts have allowed the isola-

tion of interesting new compounds such as hydroxybenzotri-

azole-like heterocycles or benzoquinone 2-diazide. In the lat-

ter case, our study reveals some interesting features in the

mechanism of their formation. The first structural elucidation

of benzoquinone 2-diazide through an X-ray crystallographic

reported. To understand better the influence of the fluorine

atoms on the reactivity of benzofuroxans and benzofurazans,

a wide array of new fluorinated heterocycles were synthe-

sized together with some already known compounds. This

has enabled an extensive investigation of their electrophilic

behavior to be undertaken through a theoretical and an elec-

trochemical study. From these studies, it could be deduced

that the replacement of nitro groups by fluorine atoms results

in a significant decrease in the electrophilic character of

benzofuroxan. Nevertheless, these compounds could un-

dergo S_NAr processes, leading to new functionalized hetero-

cycles. The first examples of aromatic nucleophilic substitu-

study is also reported. This study has unambiguously shown tion of fluorine with these compounds are also described.

that benzoquinone 2-diazide can be involved in Diels–Alder

Introduction

Substituted benzofuroxans and benzofurazans constitute

a class of 10-π heterocyclic compounds that has attracted

interest over the past twenty years. It has been shown that

they react readily with weak nucleophiles to form stable an-

ionic σ adducts when they are substituted in the 4- and/or

6-position by electron-withdrawing groups. To exemplify

this behavior, one can mention the high susceptibility of

4,6-dinitrobenzofuroxan (DNBF, 1) to undergo σ complex-

ation in the absence of any added base in aqueous solutions

to form the hydroxy adduct C-1 according the Equa-

tion (1).^[1]More importantly, it has to be noted that the

nature of the substituent borne by the six-membered ring is

of primary importance in determining how these com-

pounds react with dienes or nucleophiles.

(1)

When benzofuroxans are substituted by electron-releas-

ing substituents, the chemical reactivity is transferred from

the carbocyclic ring to the furoxan ring. Such compounds

undergo heterocyclic transformations, with the most impor-

tant transformation being known as the “Beirut reaction”,

which was described by Haddadin and Issidorides^[2,3]. This

reaction can be used to prepare quinoxaline 1,4-dioxide,

benzimidazole and phenazine derivatives, which are also

known to exhibit biological properties. The interaction of

substituted benzofuroxans with electron-rich heterocycles

or carbanions is the key step in the synthesis of these new

biologically active compounds. In 1981, Gosh, Ternai and

Whitehouse^[4]were the first to publish a review in which

the biochemical activity of benzofuroxans was extensively

studied. More generally, benzofuroxan derivatives exhibit a

broad spectrum of biological activity including antibacte-

rial, antifungal, antileukemic, acaricide, and immunodep-

ressive properties. Recently, a review reporting the pharma-

[a] Institut Lavoisier de Versailles, UMR 8180, Université de

Versailles,

45 Avenue des Etats-Unis, 78035 Versailles Cedex, France

E-mail: regis.goumont@uvsq.fr

http://ww w .ilv.uvsq.fr/

[b] Department of Chemistry, Southern Federal University,

Sorge Street 7, 344090 Rostov-on-Don, Russian Federation

Supporting information for this article is available on the

WWW under http://dx.doi.org/10.1002/ejoc.201402692.

Eur. J. Org. Chem. 2014, 6451–6466

6451

R. Goumont et al.

FULL PAPER

ceutical properties of benzofuroxans and their derivatives, fluorobenzofuroxan 3. Together with the first synthesis of

obtained through reactions involving benzofuroxans them- these compounds, we will described a theoretical and an

selves, was published.^[5,6]Although the first members of the electrochemical study of a series of fluoro-substituted

benzofuroxan family were synthesized at the end of the 19th benzofuroxans, highlighting, for the first time, their electro-

century, papers in this major field of heterocyclic chemistry philic behavior.^[14]Finally, the use of some of these com-

are still being published, adding some consistency to the pounds in S_NAr processes or in the preparation of new

present work.^[7]Importantly, authors have been able to quinoxaline dioxides will also be discussed.

show that both the medicinal properties and the reactivity

of substituted benzofuroxans are closely related to the na-

ture and the position of the substituent(s) borne by these

heterocycles.^[5]The synthesis and the particular chemistry

of quinoxaline and phenazine dioxide derivatives have also

attracted considerable attention in the last years. Some of

them exhibit biological activities including antiviral, anti-

bacterial, antiinflammatory, antiprotozoal, and anticancer

Results and Discussion

properties.^[8–10]A recent review of Cerecetto has been dedi-

cated to the chemistry and the biological activity of these

heterocycles.^[11]

The formation of substituted benzofuroxans is readily

achieved by heating of the corresponding substituted α-

nitrophenyl azides. These azides are obtained from the cor-

responding nitroanilines either by diazotization (NaNO₂in

aqueous acidic media) and subsequent treatment with aque-

ous solution of azide, or by nucleophilic displacement of a

chlorine atom with sodium azide in dimethyl sulfoxide or

in a ternary mixture: acetone/methanol/water. The optimal

experimental conditions depend on the nature of the sub-

stituent: in the case of electron-withdrawing groups the

S_NAr process of the chlorine atom removal is easily

achieved under mild conditions (2 h at room temperature).

Another possible pathway is to prepare benzofuroxan from

the corresponding aniline by using oxidative conditions

such as the use of sodium hypochlorite in basic alcoholic

solution.^[15]

The peculiar case of fluorinated quinoxaline dioxide 2a–

c has been extensively studied; these compounds exhibit

biological activity as growth inhibitors of Trypanosoma

cruzy strains against parasitic diseases such as Chagas dis-

ease, tuberculosis or candida disease, which affect approxi-

mately 20 million people.^[12]Syntheses and studies on the

biological activity of 6-fluoro- (2a), 7-fluoro- (2b), and 6,7-

difluoro-quinoxaline (2c) derivatives have been reported

previously. Nevertheless, other isomers such as 5-fluoro-, 8-

fluoro-, 5,7-difluoro, or 6,8-difluoro-quinoxaline derivatives

have, to our knowledge, never been prepared, although they

could be easily isolated from 4,6-difluoro- and 4-fluoro-

benzofuroxan (for the numbering of quinoxaline deriva-

tives, see structures above). These new biological agents

could be used in future trials to extend the range of bio-

logical activity.

Synthesis of Fluoro-Substituted Benzofuroxans

4-Fluorobenzofuroxan 4 was prepared in two steps from

2,6-difluoronitrobenzene. In a first step, heating a solution

of 2,6-difluoronitrobenzene in acetonitrile with one equiva-

lent of sodium azide leads to the sole formation of the azido

intermediate, which is subsequently heated in toluene to

give 4-fluorobenzofuroxan as a pale-yellow solid in quanti-

tative yield (Scheme 1).

In this paper, we report on the revisited synthesis of 4,6-

difluorobenzofuroxan 3 and of other new fluorinated

benzofuroxans such as 4-fluorobenzofuroxan 4. We will

Scheme 1. Synthesis of benzofuroxan 4.

Some similar experimental conditions have been used in

show that the experimental conditions reported by Leyva^[13]the synthesis of 5- and 6-fluorobenzofuroxan and have been

to prepare 4,6-difluorobenzofuroxan did not allow the syn- improved in the preparation of 4, replacing dimethyl sulfox-

thesis of this latter compound but could be used to access ide (DMSO) by acetonitrile.^[16]This change results in easier

other interesting derivatives such as hydroxybenzotriazole- isolation of the benzofuroxan derivative. The ¹H NMR

like compounds or benzoquinone 2-diazide. Interestingly, spectra of 4 is characterized by a broad signal at δ =

we will also show that the reported NMR spectroscopic 7.39 ppm for the three protons H-5, H-6, H-7. In accord-

data in this paper are not those pertaining to 4,6-di- ance with the structure of 4 was the observation in the ¹⁹F

6452

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

NMR spectra of one broad signal at δ = –129.3 ppm, char-

acteristic of the fluorine atom at the fourth position. Inter-

estingly, an X-ray crystallographic study confirmed the

structure of the new benzofuroxan 4 in the solid state, re-

vealing a bond length of 1.344 Å for the C(4)–F bond (see

Figure 1).

Scheme 2. Interconversion between the 1- and 3-oxides of benzo-

furoxans.

When R is an electron donor, structure 7 is more abun-

dant, whereas an electron-withdrawing group favors struc-

ture 9. When an electron-withdrawing group is located in

the 4-position (as in the case of the 4-nitrobenzofuroxan,

10), the benzofuroxan exists in one form with the N-oxide

in the 1-position at all temperatures.^[17,18]When the nitro

group is in the 5-position, as in the case of 5-nitrobenzo-

furoxan 11, both isomers exist at room temperature

(Scheme 3).

Figure 1. X-ray structure of 4-fluorobenzofuroxan 4.

Scheme 3. Interconversion of 5-nitrobenzofuroxan 11.

The fact that the signals pertaining to benzofuroxan 4

The synthesis of 4,6-difluorobenzofuroxan 3 has been re-

are unresolved is clear evidence of the well-known 1-oxide/ ported by Leyva^[13]. It was mentioned that heterocycle 3 can

3-oxide tautomery. This rearrangement involving substi- be prepared from 4,6-difluoro-2-nitroaniline by using either

tuted benzofuroxan has been extensively studied. Benzo- basic or acidic conditions (Scheme 4). Keeping in mind that

furoxan itself (7; R = H) has been shown by low-tempera- 4,6-difluorobenzofuroxan will be used further in the synthe-

ture NMR studies to be a rapidly equilibrating system, with sis of new quinoxaline compounds, we attempted to repro-

the transformation between the 1- and 3-oxide structure duce the reported experimental conditions.

proceeding via o-dinitrosobenzene (8; R = H) as an inter-

By using first the diazotization methodology (right part

mediate. Ring–chain tautomerism of this type (see of Scheme 4), the synthesis of the diazonium salt was per-

Scheme 2) also occurs in substituted benzofuroxans. In the formed, followed by the addition of one equivalent of so-

case of benzofuroxans substituted in the 5-position (for the dium azide in water. This step led to the isolation of a dark-

numbering of benzofuroxan, see Scheme 2), the amount of red solid in 85% yield. However, the ¹⁹F NMR spectra re-

each tautomer is dependent on the nature of R.

corded in CDCl₃was not consistent with the structure of

Scheme 4. Synthetic pathways developed by Leyva.

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6453

R. Goumont et al.

FULL PAPER

the postulated azide 12. These spectra revealed the presence

of only one signal, pertaining to one fluorine atom, at δ =

–94.0 ppm. This was confirmed by the ¹³C NMR spectra,

which also showed only one doublet pertaining to a quater-

nary carbon substituted by a fluorine atom at δ =

167.15 ppm (¹J_C,F= 268 Hz). By using 2D NMR tech-

niques such as HMQC and HMBC experiments, the struc-

ture of this unexpected product was easily determined. This

compound can be formulated as the substituted benzoquin-

one 2-diazide 13, the formation of which can be accounted

for in terms of substitution of a fluorine atom by water

during the formation of the diazonium salt. Among the

structural data, typical diagnostic features for the structure

of 13 are: (1) the presence of two doublet of doublets that

are characteristic of the protons H-5 and H-7; (2) two sig-

nals pertaining to carbons C-1 and C-2 at δ = 79.4 and

175.7 ppm, respectively. The presence of two electron-with-

drawing groups such as a nitro group and the diazonium

moiety borne by the aromatic ring favors facile nucleophilic

aromatic substitution of a fluorine atom.^[19]As this syn-

thetic pathway does not allow 4,6-difluorobenzofuroxan 3

to be prepared, a second method using sodium hypochlorite

in basic media was envisioned. These synthetic conditions

have often been used to prepare substituted benzofuroxans

and especially the unsubstituted benzofuroxan 7 (R =

H).^[20]In this case also, the experimental conditions de-

scribed by Leyva^[13], i.e., the use of sodium hypochlorite in

ethanolic solution of base (sodium or potassium hydrox-

ide), did not lead to the formation of 4,6-difluorobenzo-

furoxan 3 but to the formation of a mixture of two com-

pounds 14 and 15, in 80% yield. Their formation can be

interpreted in terms of substitution of one or both fluorine

atoms in 4,6-difluorobenzofuroxan by one or two ethoxide

moieties. Decreasing the concentration of sodium or potas-

sium hydroxide together with the reaction time resulted in

the sole formation of monosubstituted compound 14, with

approximately 20% aniline remaining in the crude mixture.

NMR spectra were in complete agreement with the struc-

ture of benzofuroxans 14 and 15, revealing broad signals

that confirm a benzofuroxan structure and, more particu-

larly, the presence of quadruplet/triplet systems characteris-

tic of the ethyl groups. Interestingly, the structure of 14 was

also firmly established by an X-ray crystallographic study

(Figure 2).

Figure 2. ORTEP view of benzofuroxan 14.

sponding anisole 16^[22]with hydrazine hydrate in ethanol at

reflux.^{[23] 1}H and ¹⁹F NMR spectra recorded in acetonitrile

exhibited characteristic signals such as two multiplets at δ

= 8.41 and 7.66 ppm for the two aromatic protons, and two

signals at δ = –110.1 and –123.8 ppm for the two fluorine

atoms, respectively. Surprisingly, the diazotization of 17

leads to total recovery of the starting material, leading to

the conclusion that compound 17 could not be the expected

hydrazine 17.

A more detailed structural study showed that the nitro

group was absent, leading to a strong shielding of the two

tertiary carbon atoms C-5 and C-7, at δ = 102.1 and

93.3 ppm, respectively. A mechanism in which the nitro

group and the hydrazine moiety are involved is reminiscent

of the mechanism of formation of substituted 1-hydroxy-

benzotriazole (HOBt). The intramolecular dehydration in-

volving 17 leads to 18, which has been postulated to have a

Scheme 5. Synthesis of HOBt-like compound 18.

These results provide clear evidence for the intermediate

formation of 4,6-difluorobenzofuroxan in the reaction mix-

ture but also show that this compound is not stable in basic

media, in which it undergoes facile nucleophilic aromatic

substitution, as could be expected.

As it was impossible for us to reproduce the experimental

conditions of Leyva to prepare 4,6-difluorobenzofuroxan 3,

new synthetic pathways, involving other starting materials,

were investigated.

First, to prevent the formation of the benzoquinone

2-diazide, a synthesis of 3 starting from hydrazine 17 was

performed. It has been reported that aromatic hydrazine

could undergo diazotization to lead to azide formation.^[21]

Scheme 6. Synthesis of benzofuroxan 3 from 1,3,5-trifluoronitro-

Hydrazine 17 was easily prepared by treatment of the corre- benzene.

6454

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

Scheme 7. Tautomer equilibrium of benzofuroxane 3.

Scheme 8. Diazotization of 21 and definitive formation of benzofuroxan 3.

HOBt-like structure (Scheme 5).^[24]The NMR spectro- formation of benzoquinone 2-diazide 13, one could postu-

scopic data are in complete agreement with this structure late that 3,5-difluoro-2-nitroaniline 21 could lead to 3-N₃.

and HRMS data were also consistent, exhibiting a peak at Importantly, the switch of the two positions bearing the ni-

m/z 172.0323 (m/z calcd. for C₆H₄N₃OF₂: 172.0322).

tro and the amino groups reduces the likelihood of nucleo-

The second attempt to prepare benzofuroxan 3 was per- philic aromatic substitution of a fluorine atom by water (see

formed from 2,4,6-trifluoronitrobenzene by using the devel- below). The diazotization of 21 in acetic acid followed by

oped experimental conditions for the synthesis of 4-benzo- the addition of sodium azide in water led to the quantitative

furoxan. Heating of one equivalent of 2,4,6-trifluoronitro- formation of 20. Importantly, the NMR spectroscopic data

benzene^[25]with one equivalent of sodium azide in aceto- pertaining to this compound were similar to those obtained

1

nitrile led to the formation of a mixture of two substituted above. The H NMR spectra reveal two signals at δ = 7.37

azides 19 and 20 in a 1:2 ratio, respectively. The overall and 7.23 ppm, which were characteristic of the two aro-

yield was around 65% and the mixture of compounds was matic protons H-3 and H-5, respectively. The ¹⁹F NMR

obtained as an orange solid. The ¹⁹F NMR spectra of the spectra exhibit two signals at δ = –103.4 and –120.6 ppm

minor azide were consistent with the symmetrical structure for the two fluorine atoms. Finally, heating of 20 to reflux

of 19 (δF_2,6= –118.5 ppm). The formation of this latter in toluene led to the exclusive formation of benzofuroxan 3

azide resulted from substitution of the fluorine atom at the as an orange liquid in 90% yield (Scheme 8). The NMR

4-position (Scheme 6).

spectroscopic data were also similar to those obtained pre-

Unfortunately, separation of the two isomers 19 and 20 viously.

was not possible and this mixture of azides was heated over-

night to reflux in toluene. The crude reaction mixture was

obtained as a pale-orange liquid in 40% yield but in this Discussion

case also, it was not possible to separate the remaining sym-

Variable-Temperature NMR experIments

metrical azide from benzofuroxan 3 through column

chromatography on silica or alumina gel. Although this

Most of the benzofuroxans prepared in this study have

synthesis does not lead to pure benzofuroxan, the first been structurally studied by conducting variable-tempera-

NMR results are consistent with the structure of 3. The ¹⁹

F

ture NMR experiments. In this discussion, we will focus on

NMR spectra, recorded in [D₆]acetone, exhibited four the peculiar case of 4,6-difluorobenzofuroxan 3, with the

broad signals corresponding to the two tautomers 3-N₁and ¹H, ¹³C and ¹⁹F NMR spectra being recorded in aceto-

3-N₃at δ = –101.7/–126.2 ppm for the minor isomer and at nitrile[D₃] from 233 to 343 K together with 2D NMR ex-

δ = –104.9/–124.2 ppm for the major tautomer. At this stage periments. The numbering of the carbon atoms of both tau-

it was not possible to determine which tautomer corre- tomers is given in Scheme 7. The various spectra are sum-

sponded to which isomer (Scheme 7). These data are not marized in Figures 3, 4, and 5. At room temperature

1

consistent with those reported by Leyva. Nevertheless, the (298 K), H, ¹³C and ¹⁹F NMR spectra of 3 showed very

fact that the NMR spectroscopic data are consistent with broad signals due to the tautomeric equilibrium highlighted

the structure of azides and that the cyclization step takes in Scheme 7. When NMR experiments are carried out at

place in neutral media, i.e., a cyclization step in toluene at low temperature (233 K) the aromatic region exhibited nar-

reflux, make these results consistent.^[5]

row peaks, corresponding to both tautomers. In acetonitrile

To obtain definitive evidence for the structure of 3, a solution (approximately 10% w/v) at 343 K, the spectra re-

final synthesis was carried out. As shown in Scheme 7, the veal signals pertaining to only one isomer due to a complete

two tautomers 3-N₁and 3-N₃coexist in solution. If 4,6- coalescence. The recording of a complete series of spectra

difluoro-2-nitroaniline does not lead to 3-N₁, due to the at low and high temperature allowed straightforward attri-

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6455

R. Goumont et al.

FULL PAPER

bution of all the resonances and a determination of the ra- the quite close couplings between H-5 (or H-5Ј) with the

tio between the two isomers at low temperature. It should two fluorine atoms. Interestingly, it is known that the ¹³C

be mentioned that, at 343 K, it has not been possible to NMR spectra of benzofuroxans show some characteristic

observe signals pertaining to the two quaternary carbons features.^[26]The resonances of C-9 and C-8 appear to be the

C-8 and C-9. Low-temperature ¹H NMR spectra show key features of the ¹³C NMR spectra of benzofuroxans.

signal of characteristic shape. For example, the shape of With chemical shift of δ = 150 and 110 ppm, respectively,

signals at δ = 7.21 and 6.95 ppm (pseudo-triplet), which can the signals of C-9 and C-8 are quite independent of the

be attributed to H-5 and H-5Ј, respectively, could be due to position and of the nature of the substituent. The position

of the signal pertaining to C-8 compared with that of C-9

1

Figure 3. H NMR of 3 in CD₃CN from 343–233 K.

Figure 4. ¹⁹F NMR of 3 in CD₃CN from 343–233 K.

6456

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

Figure 5. ¹³C NMR of 3 in CD₃CN from 343–233 K.

could be explained by the mesomeric effect of the N-oxide

HMBC spectra recorded for these compounds exhibit

characteristic correlations. For example, two correlations

functionality (Scheme 9).^[27]

This substituent effect has been attributed to the pres- between C-9 (δ ≈ 145 –150 ppm) and H-7 (J_C9,H7= 5 Hz)

ence of a partial negative charge on C-8 resulting from a and H-5 (J_C9,H5= 7–9 Hz), respectively, can be observed,

significant contribution of the second resonance form de- whereas C-8 (δ ≈ 110–115 ppm) is only correlated with H-7

scribed in Scheme 9, whereas C-9, which is more distant (J_C8,H7= 2–3 Hz).^[26]The HMBC spectra recorded for 3 in

from the N-oxide function, remains unaffected or is only acetonitrile showed nice correlations as follows: (1) C-9 (δ

slightly affected.^[26]

= 145.7 ppm) is correlated with H-7 (δ = 7.01 ppm) and H-

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6457

R. Goumont et al.

FULL PAPER

Scheme 9. Tautomeric forms of substituted benzofuroxans.

5 (δ = 7.21 ppm); (2) C-8Ј (δ = 153.2 ppm) is only correlated

It is also possible to describe ΔG^϶values by Equa-

with H-7Ј (δ = 7.19 ppm). Similarly, the correlations be- tion (4). In this equation, k_cvalues are used instead of Δν.

tween C-6 (or C-6Ј) with both protons H-5 and H-7 (or

Tc

ΔG^϶= 19.14T_c[10.32 + log( )] (Jmol^–1)

(4)

H-5Ј and H-7Ј) allow a rapid discrimination between the

resonances of C-4 and C-6. Such two-dimensional corre-

lations have been also used to determine the regioselectivity

of addition of diene on both isomers involved in the N-

oxide interconversion.^[6]Finally, from these results, it has

been determined that the major isomer was 3-N₁and that

the ratio in its favor was approximately 3:1. Similar studies

performed with 4 and 14 also revealed that the major iso-

mer has the N-oxide functionality at the N-1 position and

that the ratios are 15:1 and 4:1, respectively (Scheme 10,

low- and high-temperature NMR spectra for these benzo-

furoxans are given in the Supporting Information, Fig-

ures S1–S14).

kc

The coalescence temperatures T_c, k_c, and ΔG^϶values are

summarized in Table 1 for compounds 3, 4, and 14.

Table 1. k_cand ΔG^϶parameters for fluorobenzofuroxans 3, 4, and

14.

T_c

Δν

k_c

ΔG^϶

ΔG

[K]^[a]

[Hz]^[b][s^–1]

(kcalmol^–1)

[Eq. (3)]^[c,d]

[calmol^–1]

in favor of N₁

[e]

3

303

293

313

60

14

58

133.2

14.8 (14.5)

15.1 (15.9)

15.3 (16.0)

520

1340

790

4

31.1

14

128.8

[a] T_cvalues Ϯ5 K. [b] Average value. [c] ΔG^϶Ϯ0.5 kcalmol^–1.

[d] ΔG^϶values from Equation (4) are identical, values in parenthe-

ses are theoretical values. [e] ΔG = –RT ln(K), where T is the tem-

perature below which well-resolved spectra could be recorded,

[%N₁]

K =

.

[%N₃]

The estimated values for fluorobenzofuroxans 3, 4, and

14 are of approximately the same magnitude (ca.

15 kcalmol^–1) as those of the benzofuroxans reported ear-

lier (10–15 kcalmol^–1) by Katritzky.^[29]For example, ΔG^϶

values of 15 kcalmol^–1are reported for benzofuroxan and

5,6-dinitrobenzofuroxan and a value of 13.6 kcalmol^–1is

determined for 5-nitrobenzofuroxan 11.^[28]Interestingly, the

estimated experimental values for free enthalpies of acti-

vation ΔG^϶are in close agreement with those determined

theoretically (values in parentheses, Table 1). More impor-

tant, these values are also all in accord with those reported

for variously substituted benzofuroxans.^[28,29]

Scheme 10. 1-/3-Oxide interconversion of benzofuroxans 4 and 14.

From variable-temperature NMR experiments, the rate

constant k_cof the coalescence process and the activation

parameters for the dynamic equilibrium could be readily

obtained. A complete line-shape analysis was not always

necessary to extract these data, and simplified equations

could be used to estimate k_cand ΔG^϶values.^[28]For the

coalescence temperature T_c, the rate constant k_cis given by

Equation (2).

Formation of Benzoquinone 2-Diazide

The formation of benzoquinone 2-diazide 13 deserves

comment. Although the formation of this compound has

been previously reported, some interesting data have been

determined in our study and should be mentioned. Camps

et al. have shown that benzoquinone 2-diazide 13 formed

during their attempted synthesis of 2-amino-4,6-difluoro-

benzonitrile by a Sandmeyer cyanation of 2,4-difluoro-6-

nitrobenzene-diazonium cation.^[19]The authors mention

that diazonium salts having leaving groups at the ortho and

para position can give products derived from nucleophilic

substitution of the diazo group by nucleophiles (hydroxide

for example) present in the reaction medium, in competi-

tion with the expected nucleophilic substitution by the

cyanide moiety. In fact, no further nucleophile is needed to

form the benzoquinone 2-diazide 13. In our case, even when

πΔν

k_c=

= 2.22 Δν (s^–1)

(2)

√2

with Δν being the separation in Hz between the two signals

in the absence of exchange.

Free enthalpies of activation ΔG^϶were obtained from

the coalescence temperature T_cby the application of Equa-

tion (3), derived from Eyring’s equation:

Tc

ΔG^϶= RT_c[22.96 + ln( )] (Jmol^–1)

(3)

Δν

6458

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

sodium azide was added as a solid, the formation of benzo- Benzofurazan Synthesis

quinone 2-diazide could not be avoided, which suggests that

The formation of benzofurazans is usually achieved from

this reaction arises from adventitious water or from water

produced during the formation of NO⁺. A completely

water-free synthesis, involving the use of tertiobutyl nitrite

(tBuONO) as nitrosation agent, in acetonitrile was em-

ployed, but that also led to the sole formation of the benzo-

quinone 2-diazide.^[30]The structure of 13 was also con-

firmed through an X-ray crystallographic study, which re-

vealed that the length of the C(2)=C(3) double bond was

1.355 Å (Figure 6). Interestingly, this bond length is remi-

niscent of that of NBDF 22, with a bond length of

1.339 Å.^[31]This is typical of a nitro-olefinic fragment and

contrasts with the situation reported for DNBF 1 for which

values of 1.37 and 1.40 Å have been measured for the two

potentially reactive C(6)=C(7) and C(4)=C(5) double

bonds, respectively.^[31]

the corresponding benzofuroxan analogues by using tri-

phenylphosphine or triethylphosphite in toluene or in eth-

anol. The best experimental procedure to obtain benzo-

furazan 5 and 6 has been to heat fluorinated benzofuroxans

to reflux in toluene with one equivalent of triethylphos-

phite. The synthesis of 6 from 2,6-difluoroaniline with the

intermediate formation of an unstable nitroso compound

has been reported.^[33]Nevertheless, in our case, the deoxy-

genation process of 4 is easily carried out and the synthesis

leads to 6 in one step in a very good yield compared with

methods reported previously.^[33]The benzofurazans were

easily purified and were obtained in fair to moderate yields.

NMR spectra were in full agreement with the structure of

benzofurazan and with the disappearance of the N-oxide

functionality. Two major points have to be mentioned: The

NMR spectra reveal well-defined signals because of the ab-

sence of tautomery, and a large deshielding of the resonance

pertaining to C-8 going from δ = 110 ppm in benzofuroxan

to 155 ppm in benzofurazan.^[26]

To understand better the influence of the position of the

fluorine atom on the reactivity and electrophilicity of poly-

fluorobenzo-furoxan and -furazan, the syntheses of known

5-fluoro- (24),^[34]5,6-difluoro- (25),^[35]and 4,5,6-trifluoro-

benzofuroxan (26)^[13]were performed. To complete this

study, the syntheses of the new benzofurazan analogues 27–

29 of the previous heterocycles were carried out

(Scheme 12). The fluorobenzofurazans were fully charac-

Figure 6. ORTEP view of benzoquinone 2-diazide 13.

A first interesting result confirming the nitro-activated

character of the C(2)=C(3) double bond is that compound

13 reacts easily with cyclopentadiene to give cycloadduct 23

in quantitative yield (Scheme 11). This result is of signifi-

cant importance because it opens synthetic routes to new,

highly functionalized compounds. The result also high-

lights, for the first time, the dienophilic character of the

nitro-activated double bound of benzoquinone 2-diazide.

More interestingly, these aryl diazoketones may also be

used in the organic synthesis of highly functionalized com-

pounds such as Azamerone, a pyridazine natural prod-

uct.^[32]

Scheme 12. Structure of fluorobenzofuroxans and of the furazan

analogues.

Table 2. ¹H^[a]and ¹⁹F NMR^[b]parameters for fluorobenzofurazans.

H-4

–

H-5

7.39

H-6

–

H-7

7.58

F

5

–101.6

–117.7

–122.9

–104.7

–123.8

–121.0

–121.9

–146.9

–150.9

6^[c]

27

–

7.33

7.62

7.52

–

7.82

8.11

8.00

7.95

7.66

8.00

–

28^[d]

29

–

[a] Relative to internal SiMe₄, δ in ppm; solvent [D₆]acetone.

[b] Relative to internal CFCl₃, δ in ppm; solvent [D₆]acetone.

[c] Shifts δ are in agreement with those reported previously.^[31]

[d] Shifts δ are in agreement with those reported previously.^[33]

Scheme 11. Dienophilic behavior of the C(2)=C(3) double bond of

13.

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6459

R. Goumont et al.

FULL PAPER

terized by NMR experiments; it should be noted that liquid electrophilicity of the fluorinated heterocycles is dramati-

fluorobenzofurazans are highly volatile and that care must cally decreased upon substitution of a nitro group by a

be taken to limit the loss of those compounds, for example fluorine atom (ω = 3.0 eV for 3 and 5.46 eV for DNBF, 1;

through slow evaporation of reaction solvent under an inert ω = 2.8 eV for 4 and 4.21 for 4-NBF, 10). The replacement

gas stream. The NMR spectroscopic data are summarized of two nitro groups in compound 1 and of one nitro group

1

in Tables 2 and 3. Only the H NMR spectroscopic data of in 10 results in a Δω of 2.5 and 1.4 eV, respectively, leaving

28 were previously reported. This work allows a complete no doubt that these compounds will be unreactive towards

NMR characterization of the symmetrical 5,6-difluoro- common dienes such as cyclopentadiene in Diels–Alder re-

benzofurazan.^[36]

actions or towards neutral nucleophiles in nucleophilic sub-

stitution, such as water or methanol. Roughly, substitution

of one nitro group by a fluorine atom results in a decrease

in the electrophilicity of more than one ω unit. Previous

studies have allowed the demarcation line for σ complex-

ation and DA reactivity to be defined as a ω value of

roughly 3.5–3.8 as the boundary between super- and nor-

mal-electrophiles and between reactive dienophiles and in-

ert partners in Diels–Alder reactions.^[39]In fact, polyfluoro-

benzofuroxans and furazans fall below the border of super-

and normal electrophilic behavior, indicating a low electro-

philic character.

Table 3. ¹³C NMR parameters for fluorobenzofurazans.^[a]

C-4

C-5

C-6

C-7

C-8

C-9

5

152.0

150.8

99.5

102.2

138.7

109.9

115.21

164.7

155.4

143.1

164.5

133.9

126.9

155.4

156.4

97.8

151.2

152.7

148.6

147.2

146.9

143.2

144.3

150.6

147.2

142.7

6^[b]

27

114.1

120.5

102.2

98.7

28^[c]

29

[a] Relative to internal SiMe₄, δ in ppm; solvent [D₆]acetone.

[b] Shifts δ are in agreement with those reported previously.^[31]

[c] This work.

Electrophilicity of Fluorobenzofuroxan and Benzofurazan:

Theoretical and Electrochemical Studies

Table 4. Global properties and global electrophilicity scale and first

reduction potentials^[a]for fluorobenzofuroxans involved in this

work.

Recent density functional theory (DFT) studies on

Diels–Alder (DA) reactions have shown that the classifica-

tion of the diene/dienophilic pairs on a unique electrophilic-

ity scale is a convenient way to predict the feasibility and

the polar characteristics of the cycloaddition processes. The

question has been especially addressed by Domingo et

al.,^[37]using the global electrophilicity index ω, introduced

by Parr and defined by Equation (5).^[38]In this equation,

the electronic chemical potential μ and the chemical hard-

ness η of a substrate are two parameters that were evaluated

in terms of the one-electron energies of the frontier molec-

ular orbitals (FMO) HOMO and LUMO at the ground

state of the molecules.^[39]Another index used by Domingo

et al. is the so called ΔN_maxparameter, defined by Equa-

tion (6), which is a measure of the maximum amount of

electronic charge that the electrophilic partner can ac-

cept.^[37]According to the model, the polar character of a

DA interaction can be assessed from the difference, Δω, in

the global electrophilicities of the two reagents as well as

by the ΔN_maxvalues for the system at hand.^[36]

LUMO,

a.u.

HOMO,

a.u.

ω [eV] ΔN_max

E^I_1/2[V]

3

–0.10497

–0.10010

–0.10095

–0.09565

–0.10040

–0.10408

–0.10924

–0.09592

–0.09958

–0.10516

–0.13537

–0.23425

–0.23084

–0.25861

–0.25249

–0.22940

–0.23644

–0.24060

–0.25743

–0.26452

–0.26394

–0.25172

3.0 1.312

2.8 1.266

2.8 1.140

2.6 1.110

2.9 1.278

3.0 1.286

3.2 1.339

2.6 1.093

2.7 1.104

2.9 1.163

4.4 1.6635

–1.31

–1.22

–1.03

–1.30

–1.13

–1.35

–1.26

–1.33

–1.28

–

4

5

6

24

25

26

27

28

29

13

[a] In V vs. SRE in CH₃CN at room temperature.

The first half wave reduction potentials (E^I_1/2) associated

with a reversible monoelectronic transfer process of a large

set of substituted benzofuroxans can be determined

through a detailed electrochemical approach in acetonitrile

with a network of microelectrode by using ferrocene as an

internal reference potential. The electrochemical data per-

taining to the series of fluorobenzofuroxans are summa-

rized in Table 4. As can be seen, the major result emerging

from this study is that the reduction potentials are in the

same range (i.e., –1.03 to –1.33 V), confirming that these

benzofuroxan derivatives possess low electrophilicity com-

parable to that of benzotrithiadiazole (E^I_1/2= –1.00 V) and

μ²

ω =

(5)

2η

μ

η

ΔN_max= –

(6)

The various parameters required to compare the electro- of nitrobenzothiadiazole (E^I_1/2= –0.90 V).^[14]Qualitatively,

philicity of the fluorobenzo-furoxans and -furazans studied it has been shown that the E^I_1/2scale fits well the DA reac-

in this work are collected in Table 4. An important message tivity of this type of heteroaromatic derivative, and E^I_1/2val-

of Table 4 is that the ω values associated with all substrates ues below –0.4 V confirm that fluorobenzofuroxan cannot

are very low, being in the range 2.6–3.2 eV. This compares be engaged in Diels–Alder reactions but can be used in

in particular with ω values for nitroalkenes (e.g., ω = S_NAr processes with strong nucleophiles such as amines or

2.61 eV for nitroethylene), and cyano-activated olefins (e.g., alkoxides (see below).^[14]

ω = 2.99 eV for benzylidenemalononitriles) or 1,1-dinitro-

Microelectrodes have already proved to be efficient tools

2,2-diphenylethylene (ω = 3.16 eV).^[40]This shows that the for the investigation of reversible electrode reactions ac-

6460

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

cording rapid electron transfer kinetics.^[14]When measuring regioselectivity of the substitution was confirmed by ¹³C

time becomes sufficiently long, the current density reaches NMR spectroscopic analysis. For example, the signal of the

a steady-state value because of mass transport to the micro- characteristic carbon C-9 at δ = 148.2 ppm appears as a

electrode. Because the current at this potential is half the singlet, revealing that the fluorine at the 4-position has been

2

current limiting value, the potential is called the reversible removed, leading to the disappearance of the J_C9,Fcou-

half-wave potential (E^I_1/2), which can be associated with the pling. Such S_NAr reactions involving fluorobenzofuroxans

reversible formal potential of the Ox/Red couple (E°). Typi- and alkoxide ion or amines have been reported pre-

cal cyclic voltammograms obtained by this method for three viously.^[35]These first examples of substitution are of great

electron acceptors, 24 and 27–28 are presented in Figure 7. relevance because they lead to the regioselective formation

of highly functionalized compounds from appropriate nu-

cleophiles.

4,6-Difluorobenzofuroxan 3 in the Synthesis of

Quinoxalines

As mentioned in the introduction of this paper, one of

the goals of this study was to synthesize biologically active

compounds derived from fluorobenzofuroxans. To obtain

Figure 7. Linear scan voltammetry (20 mV/s) of a solution of the

variously substituted quinoxalines, the reaction of 4,6-di-

supporting electrolyte (tributylammonium perchlorate) in CH₃CN

fluorobenzofuroxan with dicarbonylated compounds such

as ethyl acetylacetate or acetylacetone, was taken as the

prototype reaction to establish the appropriate experimen-

tal conditions. Interestingly, heating of 4,6-difluorobenzo-

furoxan with a slight excess of carbonylated compound in

2-propanol, in the presence of a catalytic amount of

Ca(OH)₂, led to the formation of three types of new substi-

tuted quinoxalines 32–34, differing in the number of N-ox-

ide functionalities (Scheme 13).^[41]

under argon without benzofuroxan (linear curve) and of 24

(10^–2mol·dm^–3, E^I_1/2= –1.13 V vs. SRE), 27 (9ϫ10^–3mol·dm^–3,

E^I_1/2= –1.26 V vs. SRE), and 28 (4ϫ10^–3mol·dm^–3, E^I_1/2= –1.33 V

vs. SRE).

With a E^I_1/2value close to zero, DNBF 1 remains the

most electron deficient heterocycle of the series, with an

electrophilicity of the same order of magnitude as tetra-

nitrofluorenone (E^I_1/2= –0.06 V vs. SRE).^[14]

S_NAr Processes

Here we highlight the first examples of nucleophilic aro-

matic substitution involving 4,6-difluorobenzofuoxan.

First, as mentioned above, their low electrophilic character

does not allow reaction with either methanol or water. Nev-

ertheless, 4,6-difluorobenzofuoxan reacts readily with one

equivalent of sodium ethoxide, as described in the begin-

ning of this paper, to give 14 in quantitative yield. The re-

gioselectivity of this addition has been unambiguously as-

signed through an X-ray study (Figure 2). The addition of

a second equivalent of sodium ethoxide leads to the substi-

tution of the second fluorine atom at the 6-position to af-

ford 4,6-diethoxybenzofuroxan in quantitative yield. The

same reaction has been performed with 4-fluorobenzo-

furoxan. This reaction also leads to the quantitative forma-

tion of 4-ethoxybenzofuroxan 30. Some other nucleophiles

are currently being tested to extend the reactivity of fluoro-

benzofuroxans and furazans to the synthesis of new func-

tionalized compounds. For example, the reaction involving

4,6-difluorobenzofuroxan and piperidine led to quantitative

formation of 4-amino-substituted benzofuroxan 31. The Scheme 13. Formation of quinoxalines in 2-propanol.

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6461

R. Goumont et al.

FULL PAPER

The NMR spectroscopic data are in complete agreement

1

with the proposed structures of these quinoxalines. The H

NMR spectra reveal two multiplets for the two protons H-

6 and H-8, with the signal of H-6 being an apparent triplet

3

because of the two similar J coupling constants with the

fluorine atoms. The two fluorine atoms, which are two use-

ful probes, together with the HMBC spectra, allow straight-

forward assignment of the carbon atoms and especially the

quaternary atoms. The two carbon atoms C-9/C-10 are

clearly discriminated from the C-3/C-2 atoms through their

couplings with F-5 and F-7, and C-10 is assigned through

the two long-range ¹³C–¹H heteronuclear shift correlations,

3

revealed by the HMBC spectra and typical for the two J

couplings between C-10 and H-6 and between C-10 and H-

8. Similarly, the two doublet of doublets, characteristic of

C-5 and C-7 have been assigned through the HMBC spectra

and especially through the two correlations between C-7

and H-8 and H-6, shown by the HMBC spectra. Finally,

4

C-3 is assigned through the observation of a small J cou-

pling with F-5.

With all the quaternary carbon atoms being assigned, it

became easier to determine whether the quinoxaline bears

none, one, or two N-oxide functionalities. As was the case

for the benzofuroxan series, the presence of an N-oxide

group resulted in significant shielding of the carbon next to

the nitrogen atom bearing the oxygen.^[26]When the NMR

spectroscopic data of 32a are compared with those of 33a,

the resonance pertaining to the carbon C-2 shifts from δ =

148.8 ppm in 32a to δ = 139.0 ppm in 33a (Δδ ≈ 10 ppm)

whereas the chemical shift of carbon C-3 remains unaffec-

ted, confirming the position of the N-oxide group in the 1-

position. The chemical shift of C-9 follows a similar trend:

Figure 8. ORTEP views of quinoxaline 32a (top) and 33b (bottom).

Conclusions

For the first time, the synthesis of 4,6-difluorobenzo-

δ = 140.9 in 32a to δ = 136.6 ppm in 33a (Δδ ≈ 4.5 ppm). furoxan has been performed, using several synthetic strate-

The addition of a second N-oxide group in the 4-position gies. This study has shown that previous work reporting

results in a shielding of both C-10 and C-3 carbon atoms the synthesis of this benzofuroxan was erroneous. Variable-

[δ_C10= 133.1 ppm in 33c and δ_C10= 127.7 ppm in 34e (Δδ temperature NMR experiments confirm its structure and

≈ 5.5 ppm); δ_C3= 152.9 ppm in 33c and δ_C3= 140.3 ppm allow the quantification of the two tautomers 3-N₁and 3-

in 34e (Δδ ≈ 10.5 ppm)]. Surprisingly, the quinoxaline-1,4 N₃, which undergo N-oxide interconversion, in a 3:1 ratio.

dioxides were obtained in only one case; in most cases, a The various experimental pathways allow the isolation and

mixture of quinoxaline and quinoxaline-1 oxide was ob- characterization of new compounds such as HOBt-like

tained. These observations contradict previously reported compounds 18, which will be used in peptide couplings.

results. In fact, the use of these experimental conditions Perhaps more important, it has been shown that benzoquin-

(i.e., heating for one night in 2-propanol with a catalytic one 2-diazide 13 can be used as a dienophile in Diels–Alder

amount of Ca(OH)₂), only leads to the formation of quin- reactions involving cyclopentadiene. This reactivity is in

oxaline-1,4-dioxides. Modification of the experimental con- complete agreement with a ω value of 4.4 eV, determined by

ditions leads to similar results. One could anticipate that using the Parr methodology, revealing the high electrophilic

the formation of quinoxaline 32 and quinoxline-1 oxide 33 character of 13. To our knowledge, this is the first example

arises from the instability of the dioxide compounds 34, of Diels–Alder reaction involving benzoquinone diazide.

which suffer either one or two deoxygenation processes. Ex- We have also shown that these compounds can be prepared

periments are in progress to understand better the mecha- by using water-free conditions such as tertiobutylnitrite in

nism leading to the unexpected formation of both 33 and acetonitrile, indicating that water formed during the diazo-

34. The regioselectivity of the reaction has been unambigu- tization process is sufficient for substitution of the fluorine

ously assigned through the two X-ray structures of 32a and atom. Although the substitution of the nitro groups of

33b, showing that the fluorine atoms are in the 5- and 7- DNBF (1) by fluorine atoms results in a large decrease of

positions and the carbonyl moiety is in the 2-position. More the electrophilic character of fluorobenzofuroxans, the new

importantly, in the case of quinoxaline monoxides 33a–d, fluorinated benzofuroxans could be used to prepare new

the X-ray study has revealed that the N-oxide functionality quinoxaline derivatives, which will be biologically tested in

is located on the nitrogen atom N-1 (Figure 8).

further trials. Both the electrochemical and the theoretical

6462

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

nitrite (2.84 g, 41.14 mmol) in water (10 mL) was added dropwise,

immediately followed by the addition of a solution of sodium azide

(3.40 g, 52.36 mmol) in water (10 mL). The mixture was stirred for

1 h at room temperature, then water (200 mL) was added to obtain

3,5-difluoro-2-nitrophenylazide 20 (5.15 g, 70%; m.p. 55 °C) as a

studies have shown that the electrophilicity of fluorobenzo-

furoxans falls below the border demarcating super and nor-

mal electrophilic behavior, indicating a low electrophilic

character comparable to that of benzotrithiadiazole.

1

yellow solid. H NMR ([D₆]acetone): δ = 7.37 (m, 1 H, H-6), 7.23

(m, 1 H, H-4) ppm. ¹⁹F NMR ([D₆]acetone): δ = –103.4 (m, 1 F,

F-5), –120.6 (m, 1 F, F-3) ppm. ¹³C NMR ([D₆]acetone): δ = 164.2

(d, J = 254.2 Hz, C-5), 156.1 (d, J = 256.1 Hz, C-3), 138.2 (C-1),

129.4 (C-2), 105.2 (C-6), 102.3 (C-4) ppm.

Experimental Section

General: ¹H, ¹³C, and ¹⁹F NMR spectra were recorded with Bruker

200 and 300 MHz spectrometers. All spectroscopic data are re-

ported in ppm in the δ-scale relative to internal TMS (¹H and ¹³C

NMR) or CFCl₃(¹⁹F NMR). Abbreviations used are: br. s = broad

signal; d = doublet; m = multiplet. Melting points were determined

with a Büchi B-545 and are uncorrected. Starting materials were

obtained from commercial sources and were used without further

purification. Substituted benzofuroxans 24–25 were obtained ac-

cording to reported procedures.^[13,32,35]

3,5-Difluoro-2-nitrophenylazide 20 (5.15 g, 25.75 mmol) was

heated at reflux in toluene for 15 h. After cooling, the solvent was

evaporated under reduced pressure and the residue was purified by

column chromatography on silica gel (EtOAc/petroleum ether,

1:50) to obtain 4,6-difluorobenzofuroxan 3 (4.36 g, 98%) as an

1

orange liquid. H NMR ([D₃]acetonitrile, r.t.): δ = 7.05 (br. s, 2 H,

H-5–7) ppm. ¹⁹F NMR ([D₃]acetonitrile, r.t.): δ = –104.6 (br. s, 1 F,

F-4), –123.9 (br. s, 1 F, F-6) ppm.

4,6-Difluorobenzofuroxan (3): Obtained from 3,5-difluoro-2-nitro-

aniline (21).

4-Fluorobenzofuroxan (4): A mixture of 1,3-difluoro-2-nitrobenzene

(450 mg, 2.83 mmol) and sodium azide (202 mg, 3.11 mmol) was

heated at reflux in anhydrous CH₃CN (35 mL) for 15 h. The solu-

tion was then cooled and the solvent was evaporated under reduced

pressure. The residue was heated at reflux in toluene for an ad-

ditional period of 5 h to complete the cyclization of the remaining

azide. After evaporation of toluene, the residue was purified by

column chromatography on silica gel (EtOAc/petroleum ether,

1:10) to obtain 4-fluorobenzofuroxan 4 (435 mg, 99%; m.p. 45 °C)

as an orange solid. ¹H NMR ([D₃]acetonitrile): δ = 7.25 (br. s, 3

H, H-5–7) ppm. ¹⁹F NMR ([D₃]acetonitrile): δ = –129.3 (br. s, 1 F,

F-4) ppm.

3,5-Difluoro-2-nitroaniline (21): 3,5-Difluoroaniline (11.16 g,

86.4 mmol) was added to acetic anhydride (15 mL, 155.5 mmol) at

0 °C and the mixture was stirred for 1 h at room temperature. Water

(30 mL) was then added and the mixture was stirred for 30 min at

room temperature. The precipitate was filtered off to obtain 3,5-

difluoroacetanilide as a white solid (14 g, 95%; m.p. 130 °C), which

was used without further purification. 3,5-Difluoroacetanilide (7 g,

41 mmol) was dissolved in dichloromethane (50 mL) and a mixture

of fuming nitric acid (3.9 mL, 94 mmol), and sulfuric acid (3.9 mL)

was added dropwise at 0 °C. The mixture was stirred for 18 h at

room temperature, then the residue was diluted in water and ex-

tracted with dichloromethane to give a mixture of two products as

a white-yellow solid, which was separated by column chromatog-

raphy on silica gel (ErOAc/petroleum ether, 1:20) to obtain 3,5-

difluoro-2-nitroacetanilide (6 g, 68%, m.p.99.6 °C) and 3,5-di-

fluoro-4-nitroacetanilide (1 g, 11%, m.p. 137 °C).

2-Diazo-5-fluoro-3-nitrobenzoquinone (13): 2,4-Difluoro-6-nitro-

aniline (650 mg, 3.74 mmol) was dissolved in a mixture of acetic

acid (20 mL) and sulfuric acid (10 mL). The solution was cooled

to 0 °C and a solution of sodium nitrite (284 mg, 4.11 mmol) in

water (5 mL) was added dropwise. The mixture was stirred for 1 h

at room temperature. After the removal of the solvent under re-

duced pressure, the residue was extracted with EtOAc to obtain,

after evaporation, 2-diazo-5-fluoro-3-nitrobenzoquinone 13

3,5-Difluoro-2-nitroacetanilide: ¹H NMR ([D₆]acetone): δ = 7.84

(m, 1 H, H-6), 7.15 (m, 1 H, H-4) ppm. ¹⁹F NMR ([D₆]acetone): δ

= –102.5 (m, 1 F, F-5), –117.6 (m, 1 F, F-3) ppm. ¹³C NMR ([D₆]-

acetone): δ = 169.8 (C-7), 164.7 (d, J = 252.1 Hz, C-5), 156.9 (d, J

= 258.6 Hz, C-3), 135.9 (C-1), 129.1 (C-2), 107.0 (C-6), 101.4 (C-

4), 24.3 (CH₃) ppm. HRMS: m/z calcd. [MH⁺] 217.0425; found

217.0425.

1

(428 mg, 63%; m.p. 75 °C) as an orange solid. H NMR (CDCl₃):

δ = 7.18 (dd, J = 7.8 Hz, H-4), 6.68 (dd, J = 10.8 Hz, H-6) ppm. ¹⁹

F

NMR (CDCl₃): δ = –94.1 (m, 1 F, F-5) ppm. ¹³C NMR (CDCl₃):

δ = 175.6 (C-1), 167.2 (d, J = 267.3 Hz, C-5), 142.8 (C-2), 114.3

(C-4), 107.8 (C-6), 79.4 (C-3) ppm. HRMS: m/z calcd. for [MH⁺]

184.0158; found 184.0160. The experimental data were in agree-

ment with those reported previously.^[19]

3,5-Difluoro-4-nitroacetanilide: ¹H NMR ([D₆]acetone): δ = 7.58 (d,

J = 12 Hz, H-2–5), 2.16 (s, 3 H, CH₃) ppm. ¹⁹F NMR ([D₆]acet-

one): δ = –119.41 (d, J = 12 Hz, F-3–5) ppm. ¹³C NMR ([D₆]acet-

one): δ = 170.68 (C-7), 156.8 (d, J = 255.8 Hz, C-3–5), 145.7 (C-

1), 125.2 (br. s, C-4), 103.7 (C-2–5), 24.8 (CH₃) ppm. HRMS: m/z

calcd. for [MH⁺] 217.0425; found 217.0417.

4-Ethoxy-6-fluorobenzofuroxan (14) and 4,6-Diethoxybenzofuroxan

(15): Obtained from 4,6-difluorobenzofuroxan (3) upon treatment

with sodium ethoxide in ethanol. To a solution of 3 in ethanol was

added sodium ethoxide (1 equiv.). After 30 min at room tempera-

ture, ethanol was evaporated to give 14 (90%; m.p. 79 °C) as a pale-

yellow solid. ¹H NMR ([D₃]acetonitrile): δ = 6.62 (br. s, 2 H, H-5–

7), 4.28 (q, J = 7.0 Hz, 2 H, CH₂), 1.50 (t, J = 7.0 Hz, 3 H,

CH₃) ppm. ¹⁹F NMR ([D₃]acetonitrile): δ = –104.4 (br. s, 1 F, F-

5) ppm. HRMS: m/z calcd. for [MH⁺] 199.0519; found 199.0516.

3,5-Difluoro-2-nitroacetanilide (9 g, 41.65 mmol) was then heated

at reflux in HCl (3 m, 200 mL) for 4 h at 100 °C. The mixture was

diluted with water (300 mL) and filtered to obtain 3,5-difluoro-2-

nitroaniline 21 (6.5 g, 89%; m.p. 108 °C) as yellow solid, which was

engaged in the next step without further purification. ¹H NMR

([D₆]acetone): δ = 6.61 (m, 1 H, H-6), 6.42 (m, 1 H, H-4) ppm. ¹⁹

F

Compound 14 was then poured into ethanol and sodium ethoxide

(1 equiv.) was added. After 30 min stirring and the evaporation of

ethanol, 4,6-diethoxybenzofuroxan (15; quantitative) was isolated

as a yellow solid [m.p. 224 °C (decomp.)]. ¹H NMR ([D₆]aceto-

nitrile): δ = 6.32 (br. s, 1 H, H-7), 6.04 (br. s, 1 H, H-5), 4.10 (q, J

NMR ([D₆]acetone): δ = –103.2 (m, 1 F, F-5), –114.4 (m, 1 F, F-

3) ppm. ¹³C NMR ([D₆]acetone): δ = 166.1 (d, J = 250.9 Hz, C-5),

159.9 (d, J = 259.9 Hz, C-3), 149.2 (C-1), 121.0 (C-2), 100.1 (C-6),

94.1 (C-4) ppm.

3,5-Difluoro-2-nitroaniline 21 (6.5 g, 37.4 mmol) was dissolved in = 7.0 Hz, 2 H, CH₂), 3.54 (q, J = 7.0 Hz, 2 H, CH₂), 1.41 (t, J =

a mixture of acetic acid (40 mL) and sulfuric acid (20 mL). The 7.0 Hz, 3 H, CH₃), 1.13 (t, J = 7.0 Hz, 3 H, CH₃) ppm. HRMS:

reaction mixture was cooled to 0 °C, then a solution of sodium m/z calcd. for [MH⁺] 225.0875; found 225.0878.

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6463

R. Goumont et al.

FULL PAPER

The same compounds were also obtained during the synthesis of 3

using basic and oxidative conditions.

F-5) ppm. ¹³C NMR ([D₆]acetone): δ = 164.7 (d, J = 256.0 Hz,

C-5), 150.6 (C-9), 148.6 (C-8), 126.9 (C-6), 120.5 (C-7), 99.5 (C-

4) ppm.

2,4-Difluoro-6-nitroanisole (16): Potassium 2,4-difluoro-6-nitro-

phenoxide (3 g, 14 mmol) was dissolved in anhydrous CH₃CN. Di-

methylsulfate (1.6 mL, 16.8 mmol) was added and the mixture was

then heated at 40 °C for 15 h until the red color disappeared. The

yellow reaction mixture was then cooled and the solvent was re-

moved under reduced pressure. The residue was purified by column

chromatography on silica gel (petroleum ether) to obtain 2,4-di-

fluoro-6-nitroanisole 16 (1.3 g, 50%) as a yellow oil. ¹H NMR

(CDCl₃): δ = 7.36 (m, 1 H, H-5), 7.15 (m, 1 H, H-7) ppm. ¹⁹F

NMR (CDCl₃): δ = –112.9 (m, 1 F, F-4), –121.6 (d, J = 10.4 Hz,

F-2) ppm. ¹³C NMR (CDCl₃): δ = 156.5 (CF_Ar), 156.3 (CF_Ar),

144.2 (C-6), 139.1 (C-1), 109.7 (C-3), 107.6 (C-5), 62.9 (CH₃) ppm.

5,6-Difluorobenzofurazan (28): Yield 62%; white solid; m.p. 114 °C.

¹H NMR ([D₆]acetone): δ = 8.00 (m, 2 H, H-4–7) ppm. ¹⁹F NMR

([D₆]acetone): δ = –122.4 (m, 2 F, F-5,6) ppm. ¹³C NMR ([D₆]acet-

one): δ = 155.4 (d, J = 262.5 Hz, C-5,6), 147.2 (C-8–9), 102.2 (C-

4–7) ppm.

4,5,6-Trifluorobenzofurazan (29): Yield 87%; colorless, highly vola-

tile liquid; ¹H NMR ([D₆]acetone): δ = 7.95 (m, 1 H, H-7) ppm.

¹⁹F NMR ([D₆]acetone): δ = –121.9 (m, 1 F, F-5), –146.9 (m, 1 F,

F-4), –150.9 (m, 1 F, F-6) ppm. ¹³C NMR ([D₆]acetone): δ = 156.4

(d, J = 261.5 Hz, C-6), 146.9 (C-8), 143.1 (C-5), 142.7 (C-9), 138.7

(C-4), 98.7 (C-7) ppm.

4,6-Difluoro-1-hydroxybenzotriazole (18): Hydrazine hydrate

(0.1 mL, 2.92 mmol) was added to a solution of 2,4-difluoro-6-ni-

troanisole (500 mg, 2.65 mmol) in absolute ethanol and the mixture

was heated at reflux for 15 h. After cooling and evaporation of the

solvent, the residue was washed with chloroform and the solid was

filtered to obtain 4,6-difluoro-1-hydroxybenzotriazole 18 (295 mg,

65%; m.p. 100 °C) as a red powder. ¹H NMR ([D₆]DMSO): δ =

7.53 (d, J = 8.4 Hz, H-7), 7.40 (m, 1 H, H-5) ppm. ¹⁹F NMR ([D₆]-

DMSO): δ = –117.8 (s, 1 F, F-4), –125.3 (s, 1 F, F-6) ppm. ¹³C

NMR ([D₆]DMSO): δ = 157.4 (d, J = 239.8 Hz, C-6), 151.7 (d, J

= 256.4 Hz, C-4), 130.2 (C-9), 128.4 (C-8), 98.0 (C-5), 92.7 (C-

7) ppm. HRMS: m/z calcd. for [MH]⁺172.0322; found 172.0323.

4-Ethoxy-6-fluorobenzofuroxan (30): Prepared in quantitative yield

by following the procedure used for the synthesis of 14. H NMR

1

([D₆]acetone): δ = 7.24 (br. s, 1 H, H-6), 6.93 (br. s, 1 H, H-7), 6.75

(br. s, 1 H, H-5), 4.30 (q, J = 7.0 Hz, 2 H, CH₂), 1.48 (t, J = 7.0 Hz,

3 H, CH₃) ppm. ¹³C NMR ([D₆]acetone): δ = 149.9 (C-4), 131.8

(C-6), 116.6 (C-8), 109.6 (C-5), 105.0 (C-7) ppm. Were were unable

to determine the chemical shift of the C-9 carbon even by recording

¹³C NMR spectra in other solvents. HRMS: m/z calcd. for [MH⁺]

181.0613; found 181.0616.

6-Fluoro-4-piperidinobenzofuroxan (31): A mixture of 3,5-difluoro-

benzofuroxan (200 mg) and piperidine (1.2 equiv.) in THF (20 mL)

was stirred at room temperature until completion of the reaction

(3 d). The solvent was removed under reduced pressure and the

crude product was purified by column chromatography on silica

gel (petroleum ether/EtOAc, 0–10%) to give 31 (56%) as an orange

23: 2-Diazo-5-fluoro-3-nitrobenzoquinone (100 mg, 0.55 mmol)

was dissolved in dichloromethane (10 mL). Cyclopentadiene

(1.4 mL, 16.5 mmol) was added and the reaction mixture was

stirred for 3 d at room temperature. After removal of the solvent,

the residue was purified by column chromatography on silica gel

(EtOAc/petroleum ether, 1:10) to obtain 23 (24 mg, 18%; m.p.

1

solid; m.p. 82.3 °C. H NMR (300 MHz, CDCl₃): δ = 6.26 (dd, J

= 6.7, 1.7 Hz, 1 H), 6.08 (d, J = 12.2 Hz, 1 H), 3.71–3.52 (m, 4 H),

1.85–1.63 (m, 6 H) ppm. ¹³C NMR (50 MHz, CDCl₃): δ = 163.6

(d, J = 253.4 Hz), 148.2, 142.5 (d, J = 12.8 Hz), 115.3 (d, J =

16.5 Hz), 101.1 (d, J = 34.5 Hz), 83.4 (d, J = 30.9 Hz), 50.5, 25.5,

24.3 ppm. ¹⁹F NMR (188 MHz, CDCl₃): δ = –103.19 (dd, J = 12.2,

6.7 Hz) ppm.

1

88 °C) as a yellow solid. H NMR (CDCl₃): δ = 6.60 (m, 1 H, H-

8), 6.05 (m, 1 H, H-9), 5.93 (d, J = 13.2 Hz, H-6), 3.63 (m, 2 H,

H-4–10), 3.43 (s, 1 H, H-7), 1.85 (s, 2 H, CH₂) ppm. ¹⁹F NMR

(CDCl₃): δ = –82.1 (m, 1 F, F-5) ppm. ¹³C NMR (CDCl₃): δ =

179.8 (C-1), 171.6 (d, J = 281.8 Hz, C-5), 140.3 (C-8), 132.5 (C-9),

108.6 (C-6), 93.4 (C-3), 68.7 (C-2), 54.0 (C-10), 47.3 (C-7), 46.5

(CH₂), 45.7 (C-4) ppm. HRMS: m/z calcd. for [MH]⁺250.0628;

found 250.0634.

Synthesis of Substituted Quinoxalines 32–34. General Procedure:

4,6-Difluorobenzofuroxan (200 mg, 1 equiv.) and carbonylated

compound (1.2 equiv.) was heated at reflux in the presence of cal-

cium hydroxide (2.3 mg) in 2-propanol (5 mL) for 18 h. The hot

mixture was filtered and the solvent was removed under reduced

pressure. The crude product was purified by column chromatog-

raphy on silica gel (petroleum ether/EtOAc, 0–5%) to give the

product in fair to modest yield. When fair yields were obtained,

the starting 4,6-difluorobenzofuroxan was recovered.

Synthesis of Substituted Fluorobenzofurazan. General Procedure:

Substituted fluoro-benzofuroxan (1 equiv.) was heated at reflux

with triethylphosphite (1.2 equiv.) in toluene overnight. After cool-

ing, the solvent was evaporated under reduced pressure and the

residue was purified by column chromatography on silica gel

(petroleum ether) to give the substituted fluorobenzofurazan.

32a: Yield 16%; white powder; m.p. 100.4 °C. ¹H NMR (300 MHz,

CDCl₃): δ = 7.95 (d, J = 8.4 Hz, 1 H, H-8), 7.36 (td, J = 9.5,

2.3 Hz, 1 H, H-6), 2.68 (s, 3 H, H-13), 2.63 (s, 3 H, H-12) ppm.

¹³C NMR (75 MHz, CDCl₃): δ = 196.2 (s, C-11), 162.0 (dd, J =

255.6, 12.0 Hz, C-7), 158.2 (dd, J = 244.9, 18.9 Hz, C-5), 152.3 (d,

J = 1.7 Hz, C-3), 139.0 (s, C-2), 136.7 (d, J = 12.0 Hz, C-9), 132.7

(dd, J = 14.8, 2.4 Hz, C-10), 108.4 (dd, J = 29.3, 22.3 Hz, C-6),

100.1 (dd, J = 27.4, 5.1 Hz, C-8), 29.7 (s, C-12), 22.3 (s, C-13) ppm.

4,6-Difluorobenzofurazan 5: Yield 50 mg (18%); pale-yellow, highly

volatile liquid. ¹H NMR ([D₆]acetone): δ = 7.58 (m, 1 H, H-7),

7.39 (m, 1 H, H-5) ppm. ¹⁹F NMR ([D₆]acetone): δ = –101.6 (d, J

= 8.7 Hz, F-4), –117.7 (m, 1 F, F-6) ppm. ¹³C NMR ([D₆]acetone):

δ = 164.5 (d, J = 256.5 Hz, C-6), 152.0 (d, J = 267.0 Hz, C-4),

151.2 (C-8), 143.2 (C-9), 109.9 (C-5), 97.8 (C-7) ppm.

4-Fluorobenzofurazan (6): Yield 42%; colorless, highly volatile li-

1

quid. H NMR ([D₆]acetone): δ = 7.82 (d, J = 9.0 Hz, H-7), 7.62 ¹⁹F NMR (188 MHz, CDCl₃): δ = –103.1 (dd, J = 8.4 Hz, F-5),

(m, 1 H, H-6), 7.33 (m, 1 H, H-5) ppm. ¹⁹F NMR ([D₆]acetone): δ

= –122.9 (m, 1 F, F-4) ppm. ¹³C NMR ([D₆]acetone): δ = 152.7 (C-

8), 150.8 (d, J = 254.9 Hz, C-4), 144.3 (C-9), 133.9 (C-6), 115.2 (C-

5), 114.1 (C-7) ppm.

–116.4 (dd, J = 13.7, 5.7 Hz, F-7) ppm. HRMS: m/z = 223.0681.

32b: Yield 14%; pale-yellow solid; m.p. 116.1 °C. ¹H NMR

(300 MHz, CDCl₃): δ = 7.64 (d, 1 H, H-8), 7.35 (ddd, J = 11.4,

2.2, 1.1 Hz, 1 H, H-6), 4.55 (q, J = 6.7 Hz, 2 H, H-12), 2.95 (s, 3

1

5-Fluorobenzofurazan (27): Yield 75%; yellow oil. H NMR ([D₆]- H, H-14), 1.48 (t, J = 7.1 Hz, 3 H, H-13) ppm. ¹³C NMR (50 MHz,

acetone): δ = 8.11 (m, 1 H, H-7), 7.66 (d, J = 8.4 Hz, H-4), 7.52 CDCl₃): δ = 165.2 (s, C-11), 162.1 (dd, J = 218.8, 13.3 Hz, C-7),

(m, 1 H, H-6) ppm. ¹⁹F NMR ([D₆]acetone): δ = –104.7 (m, 1 F,

6464

156.9 (dd, J = 230.2, 13.3 Hz, C-5), 152.5 (t, J = 2.8 Hz, C-3), 146.6

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan

(s, C-2), 141.0 (dd, J = 14.8, 2.1 Hz, C-9), 130.7 (dd, J = 12.1, 22.3 Hz, C-6), 100.3 (dd, J = 27.4, 5.1 Hz, C-8), 22.1 (s, C-16) ppm.

1.9 Hz, C-10), 109.5 (dd, J = 22.0, 5.3 Hz, C-8), 107.6 (dd, J = 29.9,

¹⁹F NMR (188 MHz, CDCl₃): δ = –103.1 (q, J = 8.4 Hz, F-5),

–116.4 (m, F-7) ppm.

22.2 Hz, C-6), 62.9 (s, C-12), 23.6 (s, C-14), 14.3 (s, C-13) ppm. ¹⁹

F

NMR (188 MHz, CDCl₃): δ = –105.3 (dd, J = 16.2, 8.4 Hz, F-5),

–120.4 (t, J = 8.3 Hz, F-7) ppm. HRMS: m/z = 253.0791.

34e: Yield 54.5%; yellow powder; m.p. 100.4 °C. ¹H NMR

(300 MHz, CDCl₃): δ = 8.15 (d, J = 8.1 Hz, 1 H, H-8), 7.77 (d, J

= 7.5 Hz, 2 H, H-13), 7.60 (t, J = 7.2 Hz, 1 H, H-15), 7.41 (m, 9

H, H-6 and Ar-H) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 185.1

(s), 163.2 (dd, J = 258.7, 11.6 Hz), 156.3 (dd, J = 273.1, 12.9 Hz),

141.7 (dd, J = 3.8, 2.6 Hz), 140.3 (s), 140.1 (s), 135.2 (s), 134.5 (s),

131.0 (s), 130.1 (s), 129.4 (s), 129.1 (s), 129.0 (s), 127.7 (m), 126.2

(s), 109.7 (dd, J = 28.7, 24.7 Hz), 102.6 (dd, J = 27.8, 5.7 Hz) ppm.

¹⁹F NMR (188 MHz, CDCl₃): δ = –99.1 (m, F-5), –108.1 (m, F-

7) ppm. HRMS: m/z = 379.0898.

32c: Yield 13%: yellow powder; m.p. 93.2 °C. ¹H NMR (300 MHz,

CDCl₃): δ = 7.75 (dd, J = 4.5, 2.3 Hz, 2 H, Ar-H), 7.66 (d, J =

8.7 Hz, 1 H, H-8), 7.51 (dd, J = 3.9, 2.3 Hz, 3 H, Ar-H), 7.37 (td,

J = 9.4, 2.3 Hz, 1 H, H-6), 4.34 (q, J = 7.1 Hz, 2 H, H-12), 1.18

(t, J = 7.1 Hz, 3 H, H-13) ppm. ¹³C NMR (75 MHz, CDCl₃): δ =

166.2 (s, C-11), 162.2 (dd, J = 253.8, 12.1 Hz, C-7), 157.9 (dd, J =

265.6, 14.4 Hz, C-5), 151.6 (d, J = 2.3 Hz, C-3), 147.8 (s, C-2),

141.0 (dd, J = 14.8, 1.8 Hz, C-9), 137.0 (s, C-14), 130.6 (dd, J =

12.1, 1.5 Hz, C-10), 130.1 (s, C-17), 128.84 (s, C-Ar), 128.80 (s, C-

Ar), 109.3 (dd, J = 22.1, 5.3 Hz, C-8), 107.6 (dd, J = 30.0, 22.1 Hz,

C-6), 62.8 (s, C-12), 13.8 (s, C-13) ppm. ¹⁹F NMR (188 MHz,

CDCl₃): δ = –103.9 (dd, J = 16.6, 8.5 Hz, F-5), –119.1 (t, J =

8.4 Hz, F-7) ppm.

33a: Yield 21%; yellow powder; m.p. 114.7 °C. ¹H NMR

(300 MHz, CDCl₃): δ = 7.57 (d, J = 8.7 Hz, 1 H, H-8), 7.34 (td, J

= 9.4, 2.5 Hz, 1 H, H-6), 2.96 (s, 3 H, H-12), 2.81 (s, 3 H, H-

13) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 200.9 (s, C-11), 161.6

(dd, J = 252.5, 12.2 Hz, C-7), 157.2 (dd, J = 264.1, 14.4 Hz, C-5),

152.9 (t, J = 2.9 Hz, C-3), 148.8 (s, C-2), 140.9 (dd, J = 14.6, 2.2 Hz,

C-9), 130.8 (dd, J = 11.8, 2.0 Hz, C-10), 109.4 (dd, J = 21.6, 5.3 Hz,

C-8), 107.7 (dd, J = 29.9, 22.3 Hz, C-6), 27.9 (s, C-12), 24.4 (s, C-

13) ppm. ¹⁹F NMR (188 MHz, CDCl₃): δ = –105.9 (dd, J = 15.8,

8.5 Hz, F-5), –120.4 (dd, J = 12.4, 4.6 Hz, F-7) ppm. HRMS: m/z

= 223.0684.

33b: Yield 14%; white solid; m.p. 116.1 °C. ¹H NMR (300 MHz,

CDCl₃): δ = 7.95 (d, J = 8.6 Hz, 1 H, H-8), 7.34 (td, J = 9.4,

2.7 Hz, 1 H, H-6), 4.55 (q, J = 7.1 Hz, 2 H, H-12), 2.67 (s, 3 H,

H-14), 1.43 (t, J = 7.1 Hz, 4 H, H-13) ppm. ¹³C NMR (75 MHz,

CDCl₃): δ = 161.7 (dd, J = 253.5, 11.25 Hz, C-7), 161.3 (s, C-11),

158.2 (dd, J = 262.0, 11.4 Hz, C-5), 152.0 (dd, J = 2.9, 1.1 Hz, C-

3), 136.6 (dd, J = 12.8, 3.6 Hz, C-9), 135.0 (s, C-2), 132.7 (dd, J =

14.8, 2.4 Hz, C-10), 108.4 (dd, J = 29.3, 22.3 Hz, C-6), 100.3 (dd,

J = 27.5, 5.1 Hz, C-8), 63.5 (s, C-12), 22.0 (s, C-14), 14.41 (s, C-

13) ppm. ¹⁹F NMR (188 MHz, CDCl₃): δ = –103.2 (dd, J = 8.4 Hz,

F-5), –116.5 (dd, J = 38, 1.5 Hz, F-7) ppm. HRMS: m/z =

253.0784.

33c: Yield 15%; white solid; m.p. 102.9 °C. ¹H NMR (300 MHz,

CDCl₃): δ = 8.05 (d, J = 8.6 Hz, 1 H, H-8), 7.77 (dd, J = 6.8,

1.3 Hz, 2 H, H-Ar), 7.48 (m, 3 H, H-Ar), 7.40 (td, J = 8.8, 2.7 Hz,

1 H, H-6), 4.40 (q, J = 7.1 Hz, 2 H, H-12), 1.21 (t, J = 7.1 Hz, 3

H, H-13) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 162.2 (dd, J =

255.2, 10.0 Hz, C-7), 161.4 (s, C-11), 158.9 (dd, J = 255.2, 10.0 Hz,

C-5), 152.9 (dd, J = 2.9, 1.2 Hz, C-3), 136.6 (dd, J = 12.9, 3.3 Hz,

C-9), 135.6 (s, C-2), 135.0 (s, C-14), 133.1 (dd, J = 15.0, 2.3 Hz, C-

10), 130.8 (s, C-17), 129.0 (s, C-Ar), 128.6 (s, C-Ar), 108.7 (dd, J

= 29.4, 22.2 Hz, C-6), 100.4 (dd, J = 27.5, 5.2 Hz, C-8), 63.4 (s, C-

12), 13.8 (s, C-13) ppm. ¹⁹F NMR (188 MHz, CDCl₃): δ = –103.1

(q, J = 8.4 Hz, F-5), –116.4 (m, F-7) ppm.

X-ray Structural Analysis: X-ray intensity data were collected with

a Bruker X8-APEX2 CCD area-detector diffractometer using Mo-

K_αradiation (λ = 0.71073 Å). Data reduction was accomplished

using SAINT V7.03. The substantial redundancy in data allowed

a semiempirical absorption correction (SADABS V2.10)^[42]to be

applied, on the basis of multiple measurements of equivalent reflec-

tions. The structure was solved by direct methods, developed by

successive difference Fourier syntheses, and refined by full-matrix

least-squares on all F²data using SHELXTL V6.12.^[43]Hydrogen

atoms were included in calculated positions and allowed to ride on

their parent atoms.

Computational Information: Full geometry optimizations for the

dienes and dienophiles not previously studied have been performed

at the B3LY8/6-31G* level of theory,^[44–46]which is implemented in

the Gaussian 03 package of programs.^[46]The global electrophilicity

power (ω) was evaluated by means of Equation (4). The electronic

chemical potential (μ) and chemical hardness (η) values were ap-

proximated in terms of the one-electron energies of the frontier

molecule orbitals (FMO), ε_Hand ε_L, respectively, by using μ = (ε_H

+ ε_L)/2 and η = ε_H– ε_L,^[39]respectively, at the ground state (GS) of

the molecules.

Crystal Structure Analysis: See the Supporting Information. The

crystal structures (Figure 1, Figures 2 and 8) have been deposited

at the Cambridge Crystallographic Data Centre and allocated the

deposition numbers CCDC-977343 (for 13), -977344 (for 14),

-977345 (for 4), -996899 (for 32a), and -996900 (for 33b). These

data can be obtained free of charge from The Cambridge Crystallo-

graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Supporting Information (see footnote on the first page of this arti-

1

cle): Copies of the H, ¹³C, and ¹⁹F NMR spectra of the key inter-

mediates together with the low- and high-temperature NMR spec-

tra of the two benzofuroxans 4 and 14.

Acknowledgments

The authors are grateful for the support of this research by the

Centre National de la Recherche Scientifique (CNRS) and the

Ministry of Research of Russia (grant RCF 14-13-00103).

1

33d: Yield 30%; pale-yellow oil. H NMR (300 MHz, CDCl₃): δ =

[1] a) F. Terrier, S. Lakhdar, R. Goumont, T. Boubaker, Chem.

Commun. 2004, 2586–2587; b) L. Forlani, A. Tocke, E.

Del Vecchio, S. Lakhdar, R. Goumont, F. Terrier, J. Org. Chem.

2006, 71, 5527–5537; c) S. Kurbatov, R. Goumont, V. Minkin,

F. Terrier, Chem. Commun. 2006, 4279–4281; d) S. Lakhdar, M.

Westermaier, R. Goumont, T. Boubaker, H. Mayr, F. Terrier,

J. Org. Chem. 2006, 71, 9088–9095; e) L. Forlani, E. Del Vec-

chio, R. Goumont, F. Terrier, Chem. Eur. J. 2007, 13, 9600–

9607; f) R. Goumont, E. Jan, M. Makosza, F. Terrier, Org.

7.96 (td, J = 10.6, 4.2 Hz, 1 H, H-8), 7.81 (d, J = 10.6, 4.2 Hz, 2

H, H-13), 7.66 (dd, J = 10.6, 4.2 Hz, 1 H, H-12), 7.51 (t, J = 7.7 Hz,

2 H, H-14), 7.39 (ddd, J = 9.4, 8.5, 2.8 Hz, 1 H, H-6), 2.59 (s, 3 H,

H-16) ppm. ¹³C NMR (75 MHz, CDCl₃): δ = 188.6 (s, C-11), 161.9

(dd, J = 251.1, 7.6 Hz, C-7), 158.3 (dd, J = 259.3, 8.7 Hz, C-5),

153.0 (dd, J = 2.9, 1.1 Hz, C-3), 138.4 (s, C-2), 136.8 (dd, J = 12.7,

3.9 Hz, C-9), 135.3 (s, C-15), 134.4 (s, C-12), 133.0 (dd, J = 14.8,

2.4 Hz, C-10), 129.6 (s, C-14), 129.0 (s, C-13), 108.4 (dd, J = 29.2,

Eur. J. Org. Chem. 2014, 6451–6466

www.eurjoc.org

6465

R. Goumont et al.

FULL PAPER

Biomol. Chem. 2003, 1, 2192; g) S. Kurbatov, P. Rodrigez-

DaFonte, R. Goumont, F. Terrier, Chem. Commun. 2003,

2150–2151; h) M. Sebban, R. Goumont, J. C. Halle, J. Marrot,

F. Terrier, Chem. Commun. 1999, 1009–1010.

a) M. J. Haddadin, C. H. Issidorides, Heterocycles 1976, 4, 767;

b) M. J. Haddadin, C. H. Issidorides, J. Org. Chem. 1966, 31,

4067–4068.

H. Suzuki, T. A. Kawakami, Synthesis 1997, 855–857.

P. Ghosh, B. Ternai, M. Whitehouse, Med. Res. Rev. 1981, 1,

159–187.

C. Jovene, E. Chugunova, R. Goumont, Mini-Rev. Med. Chem.

2013, 13, 1089–1136.

C. Jovene, M. Sebban, J. Marrot, R. Goumont, in: Targets in

Heterocyclic Systems, Chemical Society of Italy, Rome, Italy,

2012, vol. 16, p. 90–112.

a) S. D. Jorge, F. Palace-Berl, K. F. M. Pasqualoto, M. Ishii,

A. K. Ferreira, C. M. Berra, R. V. Bosh, D. A. Maria, L. C.

Tavares, Eur. J. Med. Chem. 2013, 64, 200–214; b) L. Wang, C.

Li, Y. Zhang, C. Qiao, Y. Ye, J. Agric. Food Chem. 2013, 61,

8632–8640.

J. B. Laursen, J. Nielsen, Chem. Rev. 2004, 104, 1663–1685.

Y. Deepika, P. Surendra Nath, K. Sachin, S. Shewta, Int. J.

Curr. Pharm. Rev. Res. 2011, 1, 33–46.

A. Carta, P. Corona, M. Loriga, Curr. Med. Chem. 2005, 12,

2259–2272.

10; b) H. Gunther in: NMR Spectroscopy, 2nd ed., Wiley,

Chichester, UK, 1995, chapter 9.

[29] a) A. R. Katritzky, M. F. Gordeev, Heterocycles 1993, 35, 483–

518; b) A. J. Boulton, P. J. Halls, A. R. Katritzky, J. Chem. Soc.

B 1970, 636–1970; c) A. J. Boulton, A. R. Katritzky, M. J. Sew-

ell, B. Wallis, J. Chem. Soc. B 1967, 914–919.

[2]

[30] K. Barral, A. D. Moorhouse, J. E. Moses, Org. Lett. 2007, 9,

1809–1811.

[3]

[4]

[31] a) S. Kurbatov, R. Goumont, S. Lakhdar, J. Marrot, F. Terrier,

Tetrahedron 2005, 61, 8167–8176; b) E. A. Chugunova, A. D.

Voloshina, R. E. Mukhamatdinova, I. V. Serkov, A. N. Proshin,

E. M. Gibadullina, A. R. Burilov, N. V. Kulik, V. V. Zobov,

D. B. Krivolapov, A. B. Dobrynin, R. Goumont, Lett. Drug

Des. Discovery 2014, 11, 502–512.

[32] J. M. Winter, A. L. Jansma, T. M. Handel, B. S. Moore, Angew.

Chem. Int. Ed. 2009, 48, 767–770; Angew. Chem. 2009, 121,

781.

[5]

[6]

[7]

[33] M. E. Jung, T. A. Dong, X. Cai, Tetrahedron Lett. 2011, 52,

2533–2535.

[34] S. Uchiyama, K. Takeshira, S. Kohtani, T. Santa, R. Nakagaki,

S. Tobita, K. Imai, Phys. Chem. Chem. Phys. 2001, 4, 4514–

4522.

[35] S. K. Kotovskaya, S. A. Romanova, V. N. Charushin, M. I. Ko-

dess, O. N. Chupakhin, J. Fluorine Chem. 2004, 125, 421–428.

[36] E. Pretsch, P. Bühlmann, M. Badertscher, in: Structure Deter-

mination of Organic Compounds, Springer, Berlin/Heidelberg,

2009, p. 109–260.

[8]

[9]

[10]

[11]

[37] a) P. Arroyo, M. T. Picher, L. R. Domingo, J. Mol. Struct.

2004, 709, 45–52; b) L. R. Domingo, M. J. Aurell, P. Perez, R.

Contreras, Tetrahedron 2002, 58, 4417–4423; c) L. R. Domingo,

M. J. Aurell, P. Perez, R. Contreras, J. Phys. Chem. A 2002,

106, 6871–6875; d) P. Perez, L. R. Domingo, M. J. Aurell, R.

Contreras, Tetrahedron 2003, 59, 3117–3125.

a) M. Gonzales, H. Cerecetto, A. Monge, Top. Heterocycl.

Chem. 2007, 11, 179–211; b) M. Gonzales, H. Cerecetto, Expert

Opin. Ther. Pat. 2012, 22, 1289–1302.

a) J. J. Cazzulo, Curr. Top. Med. Chem. 2002, 2, 1261–1271; b)

P. Hotez, D. H. Molyneux, A. Fenwick, J. Kumaresan, S. E.

Sachs, J. D. Sachs, L. N. Savioli, N. Engl. J. Med. 2007, 357,

1018–1027.

S. Leyva, V. Castanedo, E. Leyva, J. Fluorine Chem. 2003, 121,

171–175.

G. Berionni, A. M. Gonçalves, C. Mathieu, T. Devic, A. Etch-

eberry, R. Goumont, Phys. Chem. Chem. Phys. 2011, 13, 2857–

2869.

[12]

[38] R. G. Parr, L. Von Szentpaly, S. Liu, J. Am. Chem. Soc. 1999,

121, 1922–1924.

[13]

[14]

[39] a) R. G. Parr, W. Yang, in: Density Functional Theory of Atoms

and Molecules, Oxford University Press, New York, 1989. b)

R. G. Parr, R. G. Pearson, J. Am. Chem. Soc. 1983, 105, 7512–

7516.

[40] S. Lakhdar, G. Berionni, R. Goumont, F. Terrier, Lett. Org.

Chem. 2011, 8, 108–118.

[15] P. B. Ghosh, M. W. Whitehouse, J. Med. Chem. 1968, 11, 305–

311.

[16] P. B. Ghosh, B. Ternai, M. W. Whitehouse, J. Med. Chem.

1972, 15, 255–260.

[17] a) A. J. Boulton, P. B. Gosh, Adv. Heterocycl. Chem. 1969, 10,

[41] A. F. Kluge, M. L. Maddox, G. S. Lewis, J. Org. Chem. 1980,

45, 1909–1914.

[42] G. M. Sheldrick, SHELXTL, version 5.1, Bruker AXS Inc.,

Madison, WI, USA, 1999.

1–41; b) A. Gasco, A. J. Boulton, Adv. Heterocycl. Chem. 1981, [43] APEX2, version 1.0–8, Bruker AXS, Madison, WI, 2003.

29, 251–340.

[44] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.

[18] P. B. Ghosh, M. W. Whitehouse, J. Med. Chem. 1968, 11, 305– [45] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.

311.

[46] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,

M. A. Robb, J. R. Cheeseman, J. A. Montgomery, T.

Vreven Jr., K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-

gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,

N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara,

K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima,

Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox,

H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gom-

perts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.

Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A.

Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dap-

prich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,

A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q.

Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov,

G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin,

D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayak-

kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,

M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03, revision

B.04, Gaussian, Inc., Pittsburgh PA, 2003.

[19] P. Camps, J. Morral, D. Munoz-Torrero, J. Chem. Res. Synop.

1998, 144–145.

[20] F. B. Mallory, Org. Synth. 1963, 4, 74–75.

[21] M. Y. Hang, V. Huynh, M. A. Hiskey, D. E. Chavez, R. D. Gil-

ardi, J. Energ. Mater. 2005, 23, 99–106.

[22] S. C. Waller, Y. A. He, G. R. Harlow, Y. Q. He, E. A. Mash,

J. R. Halpert, Chem. Res. Toxicol. 1999, 12, 690–699.

[23] M. D. Coburn, J. Heterocycl. Chem. 1974, 11, 1099–1100.

[24] F. Novelli, F. Sparatore, Farmaco 2002, 57, 871–882.

[25] G. C. Finger, F. H. Reed, J. L. Finnerty, J. Am. Chem. Soc.

1951, 73, 153–155.

[26] M. Sebban, C. Jovene, P. Sepulcri, D. Vichard, F. Terrier, R.

Goumont, in: Magnetic Resonance Spectroscopy (Ed.: Dongh-

yun Kim.), INTECH Publisher, Croatia, 2012, chapter 10, p.

183–206.

[27] a) F. Terrier, J. C. Hallé, P. MacCormack, M. J. Pouet, Can. J.

Chem. 1989, 67, 503–507; b) F. A. L. Anet, I. Yavari, Org.

Magn. Reson. 1976, 8, 158–160.

[28] a) H. Friebolin in: Basic One- and Two-Dimensional NMR

Spectroscopy, Wiley-VCH, Weinheim, Germany, 1998, chapter

Received: June 4, 2014

Published Online: September 4, 2014

6466

www.eurjoc.org

Eur. J. Org. Chem. 2014, 6451–6466

Article Doi

Revisiting the Synthesis of 4,6-Difluorobenzofuroxan: A Study of Its Reactivity and Access to Fluorinated Quinoxaline Oxides

DOI: 10.1002/ejoc.201402692

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.201402692

Full text of DOI:10.1002/ejoc.201402692

Products guided by the article

R&D Labs maybe for 149440-45-9

Relevant to this article

Hot Product