Cyano-activated fluoro displacement reactions in the synthesis of cyanophenoxazines and related compounds

Article abstract of DOI:10.1039/b008503k

Cyano-activated fluoro displacement reactions between 2,3- and 3,4-difluorobenzonitriles and the nucleophiles catechol, 2-aminophenol, 2-aminobenzenethiol and benzene-1,2-dithiol either in DMF at 130°C or in DMSO at rt, in the presence of potassium carbon

Full text of DOI:10.1039/b008503k

View Article Online / Journal Homepage / Table of Contents for this issue
Cyano-activated Ñuoro displacement reactions in the synthesis of
cyanophenoxazines and related compounds¤
Geo†rey C. Eastmond,* Thomas L. Gilchrist, Jerzy Paprotny and Alexander Steiner
Department of Chemistry, University of L iverpool, L iverpool, UK L 69 7ZD.
E-mail: eastm=liv.ac.uk
Received (in Cambridge, UK) 19th October 2000, Accepted 23rd December 2000
First published as an Advance Article on the web 22nd February 2001
Cyano-activated Ñuoro displacement reactions between 2,3- and 3,4-diÑuorobenzonitriles and the nucleophiles
catechol, 2-aminophenol, 2-aminobenzenethiol and benzene-1,2-dithiol either in DMF at 130 ¡C or in DMSO at rt,
in the presence of potassium carbonate, lead to new substituted heterocycles or other species in high yield. Catechol
at rt gives quantitative yields of cyanodibenzo[1,4]dioxines. 2-Aminophenol and 2-aminobenzenethiol at 130 ¡C
give cyanophenoxazines and cyanophenothiazines, respectively, while at rt 2-aminophenol yields
(aminophenoxy)cyanoÑuorobenzenes which can be converted into di†erent cyanophenoxazines on heating in the
presence or absence of base. The base-catalysed reactions involve a Smiles rearrangement. Benzene-1,2-dithiol
yields cyanothianthrenes (130 ¡C) or a mixture of cyanothianthrene and bis(cyanoÑuorophenylsulfanyl)benzenes
at rt.
bis(ether nitrile) 5a or 5b (Scheme 1, path B). In fact the rate of
Introduction
the intramolecular meta-Ñuoro displacement reaction, involv-
We recently reported a procedure for the simple and quanti-
tative synthesis of new family of cyano-substituted
ing cyclisation, was greatly accelerated relative to the pre-
viously described intermolecular meta-Ñuoro displacement2
and cyanodibenzo[1,4]dioxines 4 were formed in quantitative
yields.
a
dibenzo[1,4]dioxines1 involving a cyano-activated Ñuoro dis-
placement reaction between catechol 1a, or a catechol deriv-
ative, and either 2,3- or 3,4-diÑuorobenzonitrile 2a or 2b in an
aprotic solvent, DMF, in the presence of potassium carbonate
at 130 ¡C (Scheme 1). The reaction involved a combination of
either a facile ortho- or para-Ñuoro displacement reaction and
a, presumably, subsequent and less facile meta-Ñuoro displace-
ment reaction; previous work demonstrated that cyano-
activated Ñuoro displacement with 3-Ñuorobenzonitrile will
proceed in N-methylpyrrolidinone (NMP) but only at the
higher temperature of about 170 ¡C.2 Following the initial dis-
placement reaction, it is conceivable that reaction of the Ðrst-
formed product 3a or 3b could follow either of two routes, an
intramolecular cyclisation reaction (Scheme 1, path A) to form
a cyanodibenzo[1,4]dioxine 4a or 4b or an intermolecular
reaction with a second molecule of 2a or 2b to form a
Subsequently, as reported here, it was discovered that the
same Ñuoro displacement reactions will also proceed at
ambient temperatures if the solvent is changed to DMSO.
Thus, even at room temperature, compound 3 will cyclise
rather than the second hydroxyl group undergoing an inter-
molecular reaction to form a Ñuoro-substituted bis(ether
dinitrile) 5a, 5b, referred to here as an “open productÏ.
The potential of the cyano-activated Ñuoro displacement
reactions in the synthesis of other heterocycles has also been
explored. To this end, and to investigate the possibilities of
synthesizing phenoxazines, phenothiazines and thianthrenes
by Ñuoro displacement reactions, we treated 2,3- and 3,4-
diÑuorobenzonitrile (2a, 2b) with 2-aminophenol 1b, 2-
aminobenzenethiol 1c and benzene-1,2-dithiol 1d (Scheme 2).
Recently, Martinvingt-Mounir et al.3 also reported the use
of cyano-activated halogeno displacement reactions to form
cyanophenothiazines at ambient and elevated temperatures
and we take this opportunity to compare their observations
with ours.
¤ Electronic supplementary information (ESI) available: 13C NMR
data for cyanophenoxazines and related compounds, crystal struc-
tures of compounds 17 and 18. See http://www.rsc.org/suppdata/nj/
b0/b008503k/
Scheme 1
DOI: 10.1039/b008503k
New J. Chem., 2001, 25, 385È390
385
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2001

View Article Online
Scheme 2
coupled with 2-naphthol to give a red dye, the azo compound
10. Thus, it was concluded that the initial reaction at 20 ¡C
Results and discussion
Reactions were performed between equimolar quantities of
nucleophile 1 and either 2,3- or 3,4-diÑuorobenzonitrile in the
presence of potassium carbonate. Under one set of reaction
conditions the reagents were allowed to react at room tem-
perature in DMSO. Alternatively, reagents were treated under
reÑux in DMFÈtoluene at 130 ¡C.
Reactions were performed between 2,3-diÑuorobenzonitrile
2a and all four nucleophiles 1a–1d (Scheme 2). We had pre-
viously established that the ““high-temperatureÏÏ reaction at
130 ¡C between 1a and 2a in DMF gave 1-cyanodibenzo[1,4]-
dioxine 8a in 99% yield.1 Reactions have now been carried
out in DMSO at room temperature for periods of 24 h and 5
days. Again, only one crystalline product was isolated and, by
comparison with previous results, this was conÐrmed as 8a;
the yield after 24 h was 44% and after 5 days was 96%. Thus,
under both sets of reaction conditions only the cyclised
product (the dioxine) was obtained. There was no evidence for
the formation of the hydroxyphenyl ether 6a as an isolatable
intermediate; nor for a second intermolecular Ñuoro displace-
ment to form 5 (Scheme 1).
proceeded by nucleophilic displacement of Ñuoride ortho to
the nitrile of 2a to give 1-(2-aminophenoxy)-2-cyano-6-Ñuoro-
benzene 6b as the initial product.
Compound 6b, on reÑuxing for 2 h in DMF in the presence
of potassium carbonate, i.e. under similar reaction conditions
as used for the ““high-temperatureÏÏ displacement reaction,
gave 1-cyanophenoxazine 8b (mp 186È187 ¡C), identical with
the product of the ““high-temperatureÏÏ Ñuoro displacement
reaction.
By analogy with the low-temperature reaction in DMSO,
we presume that the initial displacement reaction at 130 ¡C is
a Ñuoro displacement by the hydroxyl of compound 1b to
form the intermediate 6b which is rapidly converted, by a
Smiles rearrangement, into a phenol 7b which then undergoes
cyclisation to 8b (Scheme 2).
The fact that the cyanophenoxazine produced was 1-
cyanophenoxazine was conÐrmed by hydrolysing it to the cor-
responding acid 11b and characterising that product.
Cyanophenoxazine 8b was hydrolysed by reÑuxing with pot-
assium hydroxide in ethylene glycol at 170 ¡C to form a
phenoxazine carboxylic acid. The designation of this carboxy-
phenoxazine as 1-carboxyphenoxazine, as opposed to the
alternative 4-carboxyphenoxazine, and hence the designation
of the new cyanophenoxazine produced here as 1-
cyanophenoxazine, is based on earlier studies. The carbo-
xyphenoxazine produced here had the same melting point as
the carboxyphenoxazine produced by Gilman and Moore,4
Blank and Baxter (247È248 ¡C),5 Antonio et al. (244 ¡C)6 and
Katritzky et al. (247È249 ¡C),7 the alternative 4-
Reactions with aminophenols
When 2-aminophenol 1b was used as the nucleophile the
results were more complex than those with catechol. Single
products, identiÐed by single peaks in GC, were isolated from
reactions both at room temperature (20 ¡C) and at 130 ¡C, but,
according to their retention times and the masses of their
molecular ions, the products were di†erent. Reaction for 1 h at
130 ¡C in DMF gave a yellow, Ñuorescent product (see ESI¤)
that analysed as cyanophenoxazine 8b or 9b (mp 182È184 ¡C;
X-ray data on the N-acetyl derivative as described below
conÐrm the product as 8b, see ESI¤) in high yield. In contrast,
reaction for 67 h at room temperature in DMSO gave in
88.5% yield a product which analysed as an (aminophen-
oxy)Ñuorobenzonitrile 6b; this was established by MS, FTIR
and NMR. Further proof that the product was a primary aro-
matic amine was provided by its solubility in HCl. In addi-
tion, the product was diazotized with sodium nitrite and
386
New J. Chem., 2001, 25, 385È390

View Article Online
Scheme 3
carboxyphenoxazine is reported to have a melting point of
presence of base were 8b and 14b, respectively, was obtained
by X-ray crystallography (see ESI¤). In order to remove any
possible ambiguity in the analysis of data from the cyano-
phenoxazines themselves, they were converted into their N-
acetyl derivatives 17 and 18.
181È182 ¡C.6 In addition, the product had the same 13C NMR
spectrum in CDCl as obtained and assigned by Katritzky et
3
al;7 the 13C NMR spectrum of the unsubstituted phenoxazine
was assigned by Ragg et al.8 (see ESI¤). Although the car-
boxyphenoxazine produced by Gilman and Moore was orig-
inally claimed to be 4-carboxyphenoxazine, Blank and Baxter
demonstrated the product was 1-carboxyphenoxazine.
This and later conversions of the cyanophenoxazines into
the corresponding carboxylic acids were not clean reactions.
Acid hydrolysis, with sulfuric acid at 140È150 ¡C, caused
decomposition of the phenoxazines. Alkaline hydrolysis with
KOH was slow and prolonged reaction gave black powders
from which no useful products could be extracted by subli-
mation. The most satisfactory hydrolysis procedure was a
rapid reaction with KOH in ethylene glycol at ca. 170 ¡C
when after 15 min the acids could be extracted; crude yields
were modest (ca. 70%) and attempts at further puriÐcation by
sublimation reduced the overall yields as the products decom-
posed or oligomerised. Thus, hydrolysis of cyanophenoxazines
was not found to be a good route to carboxyphenoxazines, in
contrast to hydrolysis of cyanodibenzo[1,4]dioxines which
gives good yields of carboxydibenzo[1,4]dioxines.
A corresponding series of reactions was carried out between
compound 1b and 3,4-diÑuorobenzonitrile 2b (Scheme 3). The
results were entirely consistent with those obtained for reac-
tion with 2a. Thus, Ñuoro displacement reaction at 130 ¡C
yielded a cyanophenoxazine, while reaction at room tem-
perature gave an amino ether. These results are consistent
with reactions proceeding by nucleophilic displacement of
Ñuoride para to the nitrile (in preference to that meta to the
nitrile) of 2b by the 2-aminophenoxide anion derived from 1b.
Thus, the cyanophenoxazine was assigned the structure 14b
(rather than 15b) and the aminoether 12b. The cyano-
phenoxazine was hydrolysed to the corresponding acid and
assigned the structure 16b; this acid is not known in the liter-
ature and independent conÐrmation of the structure was not
possible.
When compound 12b, produced by Ñuoro displacement in
DMSO at 20 ¡C, was heated in NMP/toluene at 170È180 ¡C
under nitrogen and in the absence of base in an attempt to
induce a thermal Smiles rearrangement, it was slowly trans-
formed into other products, including another cyano-
phenoxazine. Analysis of the reaction mixture, after 16 h, by
GC-MS showed the presence of three products. There was a
small amount of residual 12b (m/z 228), a second peak (m/z
228) and a third peak (m/z 208) with intensity ratio 1 : 14 : 38;
there was an extemely small fourth peak with m/z 208. The
mixture of solids isolated was separated by treatment with
sodium hydroxide (1 M) to extract any phenol. Another
product was isolated from the residue by vacuum sublimation.
The phenolic product analysed as 13b and was found to have
m/z 228 and a mp of 104È106 ¡C; the initial 12b had a mp of
89È90 ¡C and was not alkali soluble. Thus the phenolic
product was assigned the structure 13b. Further evidence for
the structure was obtained by acetylation with acetic anhy-
dride in pyridine; this gave a diacetyl derivative that was
assigned structure 19 on the basis of analytical and spectro-
scopic data.
Further proof that the cyanophenoxazines prepared directly
at 130 ¡C or by heating compound 6b or 12b at 130 ¡C in the
New J. Chem., 2001, 25, 385È390
387

View Article Online
The product that was isolated by sublimation was found to
have m/z 208; it analysed as a cyanophenoxazine and had a
mp of 212È214 ¡C, compared with 198È199 ¡C for 3-
cyanophenoxazine. The mixed melting point of the two cyano-
phenoxazines was 170 ¡C. The mixture of cyanophenoxazines
gave two peaks when analysed by GC-MS, each having m/z
208. This conÐrmed that the species were isomers. The very
small fourth peak present in the GC trace of the original reac-
tion mixture had the same retention characterisitics as those
of 14b, indicating that this cyanophenoxazine was formed as a
trace product in the absence of base. In view of the proof of
identity of 14b, it was concluded that the cyanophenoxazine
produced by heating 12b in the absence of base was 2-
cyanophenoxazine 15b. Acetylation of 15b gave a monoacetyl
derivative 20, the melting point of which was di†erent from
that of 10-acetyl-3-cyanophenoxazine 18.
GC analysis of samples taken during the course of the reac-
tion demonstrated that compounds 13b and 15b were formed
in a constant ratio, establishing that the species are formed
independently and not sequentially. The minute amount of 3-
cyanophenoxazine 14b that was detected in the mixture indi-
cates that cyclisation of the phenol 13b to 14b is a very minor
reaction in the absence of base.
resonances at about d 102, as expected for the structure 22.
There is precedence for the formation of dimeric species from
naphthalene-based phenoxazines in the presence of catalysts
such as copper(II) acetate9 and we attribute the product
obtained here to the analogous species 22; steric restrictions
may well preclude the formation of a dimer from 21.
Reactions with 2-aminobenzenethiol
A Ñuoro displacement reaction between 2,3-diÑuorobenzoni-
trile 2a and 2-aminobenzenethiol 1c in the presence of pot-
assium carbonate at room temperature gave 2-aminophenyl
2-cyano-6-Ñuorophenyl sulÐde 6c in high yield. Displacement
at 125 ¡C gave a mixture of two products: the aminophenyl
cyanophenyl sulÐde 6c and a cyanophenothiazine that was
tentatively assigned structure 8c by analogy with the literature
report.3 This reaction sequence would also be analogous to
that observed with 2-aminophenol. Martinvingt-Mounir et al.
performed similar displacement reactions with both 2,3- and
3,4-diÑuorobenzonitrile and 2-aminophenol but in the pres-
ence of sodium hydride.3 Our observations correspond to
theirs and when the results are taken together they indicate
that there is a very similar pattern of reactions between these
nitriles and either 2-aminophenol or 2-aminobenzenethiol.
Similar results were obtained when compound 6b was
heated in the absence of base. A second cyanophenoxazine
was produced with a melting point di†erent to that of the
cyanophenoxazine produced in the presence of base and
shown to be 8b. Thus this second cyanophenoxazine was
assigned the structure 4-cyanophenoxazine, 9b. Similarly, a
phenolic product that was isolated in the absence of base was
assigned structure 7b.
In summary, the results obtained from Ñuoro displacement
reactions between diÑuorobenzonitriles and 2-aminophenol
demonstrate that, at 130 ¡C and in the presence of a base,
cyanophenoxazines are formed readily. The structures of these
compounds indicate that they result from a Smiles rearrange-
ment in an intermediate. Reactions carried out at room tem-
perature in the presence of a base yield 2-aminophenyl
cyanoÑuorophenyl ethers. These ethers, on heating to 130 ¡C
in the presence of a base, undergo Smiles rearrrangement and
are converted into the same phenoxazines as are obtained by
direct reaction of the two precursors. However when the
ethers are heated in the absence of a base two di†erent pro-
ducts are isolated from each: a phenoxazine that results from
cyclisation without rearrangement and a 2-hydroxydiphenyl-
amine that results from Smiles rearrangement. We conclude
that in the absence of an added base. Smiles rearrangement is
very slow and this allows a competing intramolecular dis-
placement of the meta Ñuoro substituent by the amino group
to take place, giving the cyanophenoxazines 9b and 15b. Thus,
all four possible cyanophenoxazines can be produced by an
appropriate choice of reagents and conditions. The 2-
hydroxyphenyl ethers 7b and 13b apparently cyclise very
slowly, if at all, in the absence of a base.
The synthesis of cyanophenoxazines was extended to 2-
amino-3-hydroxynaphthalene via reactions at 130 ¡C. By
analogy with 2-aminophenol, reaction with 2,3-diÑuorobenzo-
nitrile gave a cyanophenoxazine which was assigned the struc-
ture 21; the 13C NMR spectrum consisted of 17 distinct lines,
as expected. In contrast, when the reaction was performed
with 3,4-diÑuorobenzonitrile a high melting point, single
product, for which elemental analysis data were consistent
with a cyanophenoxazine, was isolated in 96% yield. MS data,
however, showed this product to have a mass of 514, corre-
sponding to a dimeric species, such as 22. The integrated
intensity of the single NH peak in the 1H NMR spectrum was
approximately one nineteenth of the total integrated intensity.
The 13C spectrum showed 33 distinct lines as opposed to the
34 expected, one line however was of high intensity and we
attributed that to two carbons; there were two distinct CN
Reactions with benzene-1,2-dithiol
Benzene-1,2-dithiol 1d was allowed to react with the diÑuoro-
benzonitriles 2a and 2b in DMSO at room temperature in the
presence of potassium carbonate. The product mixtures were
analysed by GS-MS but neither showed the presence of a
thioether corresponding to a single displacement of Ñuoride
(m/z 261). The mixture obtained from the reaction with 2,3-
diÑuorobenzonitrile 2a contained a component with m/z 241,
consistent with 1-cyanothianthrene 8d. A pure specimen of the
compound was isolated by vacuum distillation as a colorless
crystalline solid with a strong yellow Ñuorescence. A second
component was isolated that showed m/z 380; this was tenta-
tively assigned the structure 23. This is the product that would
result from double intramolecular displacement of the ortho
Ñuoride by the dithiol, a reaction that corresponds to path B
in Scheme 1 but which was not observed with catechol. A very
similar reaction pattern was found with 3,4-diÑuorobenzo-
nitrile 2b. Although 2-cyanothianthrene 14d was detected in
the reaction mixture by GC-MS and by its strong Ñuorescence
it was not isolated pure. Again, a 2 : 1 adduct, which was
assigned structure 24, was also formed.
In contrast, each reaction gave a single substance in high
yield when they were carried out in DMF under reÑux. These
substances were identiÐed as 1-cyanothianthrene 8d and 2-
cyanothianthrene 14d, respectively. Thus, the reactions take a
di†erent course from those with catechol at room temperature
but the same course at higher temperature, the double dis-
placement of Ñuoride providing an efficient route to these new
thianthrene derivatives.
388
New J. Chem., 2001, 25, 385È390

View Article Online
cyanobenzo[b]phenoxazine 21 (2.27 g, 88%). A portion was
sublimed to give an analytical specimen as a pale green crys-
talline solid, mp 299È300 ¡C (decomp.); IR (KBr): 3314, 2220,
Experimental
For General Methods see the preceding paper.10
1525, 1499, 1320 and 740 cm~1; 1H NMR (d -DMSO): d 6.71
Crystallography
6
(dd, 1 H, J \ 8.0, 7.9), 7.00 (dd, 1 H, J \ 7.9, 1.3), 7.14 (dd, 1
Crystal data were collected on a Stoe-IPDS di†ractometer at
213 K using Mo-Ka radiation (j \ 0.710 73 A). Full-matrix
least squares reÐnements on F2 using all data (SHELX 97).11
H, J \ 7.8, 1.8), 7.15 (s, 1 H), 7.21 (s, 1 H), 7.27 (m, 3 H), 7.54
(d, 1 H, J \ 7.5), 7.58 (d, 1 H, J \ 7.8 Hz) and 9.31 (s, 1 H); for
13C NMR see ESI.¤ Calc. for C
H
N O: C, 79.06; H, 3.90;
17 10
2
17:
C
P2 /c,
H
N O , M \ 250.27, monoclinic, space group
N, 10.85%. Found: C, 78.79; H, 3.86; N, 10.92%. HRMS:
15 10
2 2
a \ 8.010(2),
b \ 17.598(3),
c \ 8.826(2)
A,
calc. for C
H
N O m/z 258.079 32, found 258.078 86.
1
17 10
2
b \ 105.42(3), V \ 1199.4(5) A
Ž
3, Z \ 4, k(Mo-Ka) \ 0.094
mm~1, R1[I [ 2p(I)] \ 0.065, wR2 (all data) \ 0.194, 4437
3-Cyano-12-(3-cyanobenzo[b]phenoxazin-11-yl)benzo-
measured and 1630 independent reÑections, R \ 0.115. 18:
[b]phenoxazine 22. Reaction between 3-amino-2-naphthol
and 3,4-diÑuorobenzonitrile 2b, conducted as for the synthesis
of 1-cyanophenoxazine, gave a crude product (96%) as a
colourless crystalline solid, mp [300 ¡C. A pure sample was
obtained by slow vacuum sublimation at 250 ¡C and 0.5
mmHg; some decomposition was found when attempts were
made to recrystallise the product from THF. GC-MS showed
the presence of only one component with m/z 514; 1H NMR
int
C
H
N O , M \ 250.27, monoclinic, space group P2 /n,
15 10
2
2
1
a \ 11.872(2), b \ 7.8232(12), c \ 12.943(2) A, b \ 99.18(2)¡,
V \ 1186.7(3)
AŽ 3, Z \ 4, k(Mo-Ka) \ 0.095 mm~1,
R1[I [ 2p(I)] \ 0.113, wR2 (all data) \ 0.297, 6426 measured
and 1642 independent reÑections, R \ 0.102. The crystals of
int
compound 18 were of poor quality leading to a rather high R
factor. While the overall connectivity is not in doubt, the
quality of the reÐnement does not allow a detailed discussion
of structural parameters.
CCDC reference number 440/254. See http://www.rsc.org/
suppdata/nj/b0/b008503k/ for crystallographic Ðles in .cif
format.
(d -DMSO): d 6.15 (d, 1 H, J \ 8.3), 6.44 (s, 1 H), 6.86 (d, 1 H,
6
J \ 8.3), 7.18È7.55 (m, 12 H), 7.73 (d, 1 H, J \ 7.8), 7.86 (d, 1
H, J \ 8.3 Hz) and 9.57 (s, 1 H); 13C NMR (d -DMSO): d
6
102.22, 102.76, 108.46, 111.21, 111.52, 112.22, 113.30, 114.44,
118.18, 118.26, 118.74, 118.84, 120.63, 124.77, 124.98, 125.30,
126.42 (2C), 126.78, 126.94, 127.97, 129.15, 129.25, 130.06,
130.17, 130.24, 130.45, 130.95, 134.48, 135.42, 142.04, 143.19,
1-Cyanophenoxazine 8b. 2-Aminophenol 1b (1.09 g, 10
mmol) was dissolved in DMF (20 mL) with potassium carbon-
ate (3 g) and toluene (12 mL) and the mixture heated under
reÑux for 0.5 h under a Ñow of nitrogen to remove water with
the aid of a DeanÈStark trap. The Ñask was cooled to about
80 ¡C when 2,3-diÑuorobenzonitrile 2a (1.39 g, 10 mmol) was
added and the mixture heated under reÑux for 4 h after which
the toluene was distilled o† and replaced with DMF. The
143.33 and 143.37. Calc. for C
N, 10.89%. Found: C, 78.79; H, 3.86; N, 10.92%.
H
N O : C, 79.37; H, 3.53;
34 18
4 2
2-Aminophenyl 2-cyano-6-Ñuorophenyl ether 6b. 2-
Aminophenol 1b (2.18 g, 20 mmol) was dissolved in anhydrous
DMSO (20 mL) and the solution deoxygenated by blowing
nitrogen through for 5 min. Then 2,3-diÑuorobenzonitrile 2a
(2.78 g, 20 mmol) was added followed by anhydrous pot-
assium carbonate (5 g). The reaction mixture was stirred in a
closed vessel for 72 h; then poured into iceÈwater when the
solid product that precipitated was Ðltered o† and thoroughly
washed with water and dried. This gave the ether 6b (4.33 g,
95%) as the only product. An analytical specimen was
obtained by vacuum distillation as a pale pink solid, mp 95È
96 ¡C; IR (KBr): 3430, 3351, 2233, 1623, 1505 and 753 cm~1;
mixture was heated at 150 ¡C for 4.5
cyanophenoxazine 8b (1.98 g, 95%) as a bright yellow, highly
Ñuorescent, crystalline solid, mp 182È184 ¡C; IR (KBr): 3310,
h to give 1-
2223, 1510, 1473, 1300 and 732 cm~1; 1H NMR (d -DMSO):
6
d 5.86 (s, 1 H), 6.50 (d, 1 H, J \ 7.5), 6.62 (dd, 1 H), 6.66 (d, 1
H), 6.76 (d, 1 H), 6.80 (dd, 1 H), 6.91 (d, 1 H, J \ 7.1 Hz) and
7.74 (dd, 1 H); for 13C NMR see ESI.¤ Calc. for C H N O:
13
8 2
C, 74.98; H, 3.87; N, 13.45%. Found: C, 75.04; H, 3.83; N,
13.38%. HRMS: calc. for C H N O m/z 208.063 65, found
208.063 65.
In order to obtain crystals suitable for X-ray analysis (see
ESI¤) the compound was further characterised as its 10-acetyl
derivative 17 by acetylation with acetic anhydride in pyridine
with 4-dimethylaminopyridine as catalyst; mp 188È189 ¡C.
13
8 2
1H NMR (d -DMSO): d 5.20 (s, 2 H), 6.44È6.47 (m, 2 H), 6.80
6
(ddd, 1 H, J \ 8.6, 1.5, 0.5 Hz), 6.84È6.89 (m, 1 H), 7.42È7.47
(m, 1 H) and 7.73È7.80 (m, 2 H); 13C NMR (d -DMSO): d
6
107.86, 114.16, 114.64, 115.60, 122.77, 124.19, 126.53, 129.71,
138.27, 144.17, 144.94, 152.60 and 155.09. Calc. for
Calc. for C
C, 71.74; H, 3.99; N, 11.13%.
H
N O : C, 71.99; H, 4.02; N, 11.19%. Found:
C
H FN O: C, 68.41; H, 3.97; N, 12.27%. Found: C, 68.35;
15 10
2 2
13
H, 3.92; N, 12.33%. HRMS: calc. for C H FN O m/z
13
9
2
9
2
228.069 89, found 228.069 59.
3-Cyanophenoxazine 14b. Reaction between 2-aminophenol
1b and 3,4-diÑuorobenzonitrile 2b was conducted as for
the synthesis of 1-cyanophenoxazine. This gave 3-
cyanophenoxazine 14b (1.98 g, 95%) as a yellow highly Ñuo-
rescent crystalline solid, mp 198È199 ¡C; IR (KBr): 3314, 2220,
2-Aminophenyl 4-cyano-2-Ñuorophenyl ether 12b. Reaction
between 2-aminophenol (20 mmol) and 3,4-diÑuorobenzoni-
trile 2b (20 mmol) as described above gave the ether 12b (95%)
as a pinkÈwhite solid. An analytical sample, obtained by
vacuum distillation, had mp 89È90 ¡C; IR (KBr): 3479, 3377,
1525, 1499, 1320 and 740 cm~1; 1H NMR (d -DMSO): d 6.49
6
(m, 2 H), 6.64 (m, 2 H), 6.77 (ddd, 1 H, J \ 7.8, 7.6, 2.2), 6.97
2235, 1504, 1314 and 741 cm~1; 1H NMR (d -DMSO): d 5.20
(d, 1 H, J \ 1.7), 7.16 (dd, 1 H, J \ 8.2, 1.8 Hz) and 8.90 (s, 1
H); for 13C NMR see ESI.¤ Calc. for C H N O: C, 74.98;
6
(s, 2 H), 6.44È6.47 (m, 2 H), 6.60 (td, 1 H, J \ 8.5, 1.6), 6.78 (t,
13
8 2
1 H, J \ 8.5), 6.89 (dd, 1 H, J \ 8.0, 1.5), 6.91 (1 H, dd,
J \ 8.0, 1.4), 7.03 (td, 1 H, J \ 7.6, 1.4), 7.59 (ddd, 1 H,
J \ 8.5, 2.01, 1.1) and 7.960 (dd, 1 H, J \ 11.0, 2.0 Hz); 13C
H, 3.87; N, 13.45%. Found: C, 74.83; H, 3.87; N, 13.52%.
HRMS: calc. for
208.06365.
The 10-acetyl derivative 18 had mp 147È148.5 ¡C and was
characterised by X-ray di†raction (see ESI¤). Calc. for
C H N O m/z 208.063 65, Found
13
8 2
NMR (d -DMSO): d 104.43, 116.39, 117.36, 117.83, 120.51,
6
120.84, 126.51, 130.09, 139.24, 140.65, 149.67, 150.22 and
152.69. Calc. for C H FN O: C, 68.41; H, 3.97; N, 12.27%.
C
H
N O : C, 71.99; H, 4.02; N, 11.19%. Found: C, 71.68;
13
9
2
15 10
2 2
Found: C, 68.52; H, 3.95; N, 12.35%. HRMS: calc. for
H FN O m/z 228.069 89, found 228.070 03.
H, 4.03; N, 11.25%.
C
13
9
2
1-Cyanobenzo[b]phenoxazine 21. Reaction between 3-
amino-2-naphthol and 2,3-diÑuorobenzonitrile 2a, conducted
as for the synthesis of 1-cyanophenoxazine, gave 1-
Thermal reaction of 2-aminophenyl 4-cyano-2-Ñuorophenyl
ether 12b in the absence of base. The ether 12b (0.5 g) was
New J. Chem., 2001, 25, 385È390
389

View Article Online
heated under reÑux in NMP (10 mL) and toluene (5 mL)
2-Aminophenyl 2-cyano-6-Ñuorophenyl sulÐde 6c. A solution
of 2-aminobenzenethiol 1c (0.37 g, 3.0 mmol) and 2,3-diÑuoro-
benzonitrile 2a (0.42 g, 3.1 mmol) in DMSO (20 mL) was
stirred at rt with K CO (3.0 g) for 67 h to give the sulÐde 8c
under N at 170È180 ¡C for 16 h. The solution was poured
2
into iceÈwater; the solid was Ðltered o†, washed three times
with water and dried (0.45 g). GC-MS showed the presence of
three components in the ratio 1 : 14 : 38. The Ðrst (m/z 228)
was identiÐed as the starting ether 12b, the second also had
m/z 228 and the third m/z 208. The components were separat-
ed, Ðrst by extracting the product with 1 M NaOH, then with
5 M HCl. The residue (fraction 3) was puriÐed by sublimation.
The acid soluble fraction (9 mg) was identiÐed as the starting
ether 12b. The alkaline-soluble fraction (m/z 228) had mp 104È
106 ¡C; IR (KBr): 3336, 3300È3200br, 1619 and 1525 cm~1;
this was tentatively identiÐed as the phenol 13b. This was
characterized as its diacetyl derivative 19, mp 117È118 ¡C.
2
3
(0.61 g, 84%), mp 96È97 ¡C (lit.,3 98 ¡C); IR (KBr): 3415 and
3339 cm~1. Calc. for C H FN S: C, 63.91; H, 3.71; N,
13
9
2
11.46%. Found: C, 63.83; H, 3.69; N, 11.50%.
Reaction of 2-aminobenzenethiol and 2,3-diÑuorobenzonitrile
on heating with base. The preceding reaction was repeated but
with DMF as the solvent at 125 ¡C for 50 min. This gave a
solid (0.64 g). GC-MS showed the presence of two com-
ponents in a ratio of 1 : 5.5: (i) the sulÐde 6c (m/z 244) identi-
Ðed by comparison with an authentic specimen and (ii) a
substance (m/z 224) tentatively identiÐed as 1-cyano-
phenothiazine 8c. The reaction mixture was not investigated
further.
Calc. for C
H
FN O : C, 65.38; H, 4.19; N, 8.97%. Found:
17 13
2 3
C, 65.18; H, 4.19; N, 9.08%. The third fraction (m/z 208) was
identiÐed as 2-cyanophenoxazine 15b, mp 212È214 ¡C, mixed
mp with 3-cyanophenoxazine 14b 170 ¡C; IR (KBr): 3352,
2219, 1582, 1494, 1313 and 733 cm~1. Calc. for C H N O:
1-Cyanothianthrene 8d. A solution of benzene-1,2-dithiol 1d
13
8 2
(0.50 g, 3.5 mmol) in DMF (15 mL) containing K CO (1.3 g)
C, 74.98; H, 3.87; N, 13.45%. Found: C, 74.80; H, 3.84; N,
2
3
was heated under reÑux under N and 2,3-diÑuorobenzonitrile
13.39%. The 10-acetyl derivative 20 had mp 106È108 ¡C. Calc.
2
2a (0.48 g, 3.5 mmol) in DMF (5 mL) added dropwise during
for C
H
N O : C, 71.99; H, 4.02; N, 11.19%. Found: C,
15 10
2 2
1.5 h. After a further 0.5 h the reaction mixture was poured
into ice to give a colourless precipitate (0.81 g, 95%). This was
puriÐed by sublimation at 0.5 mmHg to yield 1-
cyanothianthrene 8d (0.68 g, 81%), mp 128È129 ¡C; IR (KBr):
2228, 1442, 1402 and 747 cm~1. Calc. for C H NS : C,
71.71; H, 4.03; N, 11.26%.
Thermal reaction of 2-aminophenyl 2-cyano-6-Ñuorophenyl
ether 6b in the absence of base. The ether 6b (1.0 g) was heated
under reÑux in NMP (10 mL) and toluene (5 mL) under N at
13
7
2
2
64.69; H, 2.92; N, 5.80%. Found: C, 64.88; H, 2.91; N, 5.87%.
170È180 ¡C for 30 h. The solution was poured into iceÈwater;
GC-MS: m/z 241 (M`).
the solid was Ðltered o†, washed three times with water and
dried. GC-MS showed the presence of three components.
These were separated by extracting the mixture with 1 M
NaOH, then with 5 M HCl. The acid soluble fraction (9 mg)
was identiÐed as the starting amine 6b. The alkali soluble frac-
tion (9 mg) was assigned the structure 7b; IR (KBr): 3372,
3300È3000br, 2230, 1654, 1599, 1515 and 1426 cm~1; m/z 228.
The residue (0.5 g) was puriÐed by vacuum sublimation to
give 4-cyanophenoxazine 9b, mp 148È150 ¡C; IR (KBr): 3332,
2230, 1581, 1490, 1311 and 712 cm~1. Calc. for C H N O:
2-Cyanothianthrene 14d. A solution of benzene-1,2-dithiol
1d (0.45 g, 3.1 mmol) and 3,4-diÑuorobenzonitrile 2b (0.45 g,
3.2 mmol) in DMF (15 mL) containing K CO (1.0 g) was
2
3
heated under reÑux under N for 1.5 h. The reaction mixture
2
was poured into ice to give a colourless precipitate (0.70 g,
92%). This was puriÐed by sublimation at 0.5 mmHg to yield
2-cyanothianthrene 14d (0.62 g, 83%), mp 133È134 ¡C; IR
(KBr): 2230, 1446, 820 and 735 cm~1. Calc. for C H NS : C,
13
7
2
13
8 2
64.69; H, 2.92; N, 5.80%. Found: C, 64.75; H, 2.88; N, 5.83%.
C, 74.98; H, 3.87; N, 13.45%. Found: C, 74.93; H, 3.83; N,
13.40%.
GC-MS: m/z 241 (M`).
Acknowledgements
Base induced cyclisation of ether 6b. The ether 6b (0.1 g) was
heated under reÑux in DMF with K CO (0.9 g) for 2 h. The
The authors wish to thank EPSRC for Ðnancial support and
the help given by Mr S. Apter (elemental analysis, FTIR), Mr
A. Mills and Ms M. McCarron (GC and MS), Dr P. Leonard
(recording NMR spectra) and Mr J. Bickley (X-ray
di†raction).
2
3
product was precipitated into water, Ðltered o† and dried to
give 1-cyanophenoxazine 8b, mp 186È187 ¡C.
Phenoxazine-1-carboxylic acid 11b. 1-Cyanophenoxazine 8b
(50 mg) was heated under reÑux in ethane-1,2-diol with KOH
for 15 min. The crude acid was isolated (70%) after an acidic
work-up. A pure sample was obtained by vacuum sublimation
at 250 ¡C and had mp 243È245 ¡C (lit.4,5,7 247È248 ¡C); IR
References
1
2
3
G. C. Eastmond and J. Paprotny, Chem. L ett., 1999, 479.
G. C. Eastmond and J. Paprotny, J. Mater. Chem., 1997, 7, 1321.
C. Martinvingt-Mounir, Y. Mettey and J.-M. Vierfond, J. Hetero-
cycl. Chem., 1998, 35, 843.
(KBr): 3353, 1665, 1498 and 1278 cm~1; 1H NMR (d -
6
DMSO): d 6.61 (dd, 1 H, J \ 8.1, 7.9), 6.64È6.72 (m, 2 H),
6.73È6.80 (m, 2 H), 6.82 (dd, 1 H, J \ 7.8, 1.4), 7.32 (dd, 1 H,
J \ 8.1, 1.4 Hz), 8.95 (s, 1 H) and 13.22 (s, 1 H); for 13C NMR
see ESI.¤ Calc. for C H NO : C, 68.71; H, 3.99; N, 6.16%.
4
5
6
H. Gilman and L. O. Moore, J. Am. Chem. Soc., 1958, 80, 2195.
B. Blank and L. L. Baxter, J. Med. Chem., 1968, 11, 807.
Y. Antonio, P. Barrera, O. Contreras, F. Franco, E. Galeazzi, J.
Garcia, R. Greenhouse, A. Guzman, E. Velarde and J. M.
Muchowski, J. Org. Chem., 1989, 54, 2159.
13
9
3
Found: C, 68.30; H, 3.87; N, 5.95%. HRMS: calc. for
H NO m/z 227.058 23, found 227.058 32.
C
13
9
3
7
8
9
A. R. Katritzky, L. M. Vasques de Miguel and G. W. Rewcastle,
Heterocycles, 1987, 26, 3135.
Phenoxazine-3-carboxylic acid 16b. Hydrolysis of 3-
E. Ragg, G. Fronza, R. Mondelli and G. Scapini, J. Chem. Soc.,
Perkin T rans. 2, 1983, 1289.
J. A. Van Allen, G. A. Reynolds and D. P. Maier, J. Org. Chem.,
1969, 34, 1691.
cyanophenoxazine 14b as in the preceding preparation gave
phenoxazine-3-carboxylic acid 16b, mp 254È256 ¡C after
vacuum sublimation; IR (KBr): 3419, 1671, 1583, 1457 and
1310 cm~1. Calc. for C H NO : C, 68.71; H, 3.99; N, 6.16%.
10 G. C. Eastmond, J. Paprotny, A. Steiner and L. Swanson, New J.
Chem., 2001, 25, 379.
11 G. M.Sheldrick, SHELX 97, Program for crystal structure reÐne-
ment, University of Gottingen, Germany, 1997.
13
9
3
Found: C, 68.74; H, 3.97; N, 6.17%. HRMS: calc. for
C
H NO m/z 227.058 23, found 227.057 88.
13
9
3
390
New J. Chem., 2001, 25, 385È390