On the reaction of 3-bromo-2-nitrobenzo[b]thiophene with some ortho-substituted anilines: An analysis of the products of reaction and of their NMR and MS properties

Article abstract of DOI:10.1016/S0040-4020(03)01060-3

The title reaction, carried out in DMF in the presence of triethylamine or potassium carbonate, has furnished the 'expected' 3-amino-2-nitrobenzo[b]thiophenes 2o together with the 'unexpected' 2-amino-3-nitrobenzo[b]thiophenes 3o, thus recalling the situation observed with other weak nucleophiles in the presence of non-nucleophilic bases. The effects (electronic as well as steric) of the ortho-substituent (OH, NH ₂, OMe, Me, Et, F, Cl and Br) on the course of the reaction have been investigated, determining their influence on yields and product ratios (2o/3o). An analysis of ¹³C NMR and MS spectra of 2o and 3o has been carried out. Ab initio computations on 2of, 2oi, 3of and 3oi at DFT level have furnished informations on their geometry and stability in the gas phase, thus allowing to assign a role to their stability on the course of the reaction as well as on some EI-MS results.

Full text of DOI:10.1016/S0040-4020(03)01060-3

Tetrahedron 59 (2003) 7189–7201
On the reaction of 3-bromo-2-nitrobenzo[b]thiophene with some
ortho-substituted anilines: an analysis of the products of reaction
and of their NMR and MS properties
Barbara Cosimelli,a Liliana Lamartina,b Camilla Zaira Lanza,c Domenico Spinelli,c
,
*
Raffaella Spisanic and Federica Vegnab
a
`
Dipartimento di Chimica Farmaceutica e Tossicologica, Universita degli Studi Federico II, Via D. Montesano 49, I-80131 Napoli, Italy
`
b
Dipartimento di Chimica e Tecnologie Farmaceutiche, Universita di Palermo, Via Archirafi 32, I-90123 Palermo, Italy
`
Dipartimento di Chimica Organica A. Mangini, Universita di Bologna, Via S. Donato 15, I-40127 Bologna, Italy
c
Received 7 May 2003; revised 9 June 2003; accepted 3 July 2003
Abstract—The title reaction, carried out in DMF in the presence of triethylamine or potassium carbonate, has furnished the ‘expected’
3-amino-2-nitrobenzo[b]thiophenes 2o together with the ‘unexpected’ 2-amino-3-nitrobenzo[b]thiophenes 3o, thus recalling the situation
observed with other weak nucleophiles in the presence of non-nucleophilic bases. The effects (electronic as well as steric) of the ortho-
substituent (OH, NH2, OMe, Me, Et, F, Cl and Br) on the course of the reaction have been investigated, determining their influence on yields
and product ratios (2o/3o). An analysis of 13C NMR and MS spectra of 2o and 3o has been carried out. Ab initio computations on 2of, 2oi, 3of
and 3oi at DFT level have furnished informations on their geometry and stability in the gas phase, thus allowing to assign a role to their
stability on the course of the reaction as well as on some EI-MS results.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
In the framework of our investigations on the nucleophilic
reactivity of halogenonitroheterocycles1 – 3 we have
observed that 3-bromo-2-nitrobenzo[b]thiophene (1) reacts
with weak nucleophiles (for example, anilines) in the
presence of non-nucleophilic bases (triethylamine or
potassium carbonate) giving beside the expected 2m and
2p also the unexpected 3m and 3p, deriving from a novel
kind of nucleophilic substitution with rearrangement
(Scheme 1).4–6 Moreover we have been interested for a
long time in the study of chemico-physical and biological
properties of nitroheterocyclic compounds showing
pharmacological activity.7 – 10 Actually, a number of
N-substituted 3-amino-2-nitrobenzo[b]thiophenes 2 show
analgesic, anti-exudative and/or anti-inflammatory
activities. In particular, some (2m and 2p) show high anti-
inflammatory activity11,12 (probably affecting the bio-
synthesis of the prostaglandines or of the autacoids related
to them) and present, at the same time, only a low mutagenic
activity (as shown by the Ames test)9 notwithstanding the
presence of the nitrogroup: thus, they can be indicated as
promising anti-inflammatory drugs. These facts further
Scheme 1.
strengthen the well recognised observation that the
Keywords: nucleophilic aromatic substitution; rearrangements in SNAr;
3-bromo-2-nitrobenzo[b]thiophene; 13C NMR; EI-MS.
thiophene ring is a structural element in a number of
pharmacologically active compounds (chemotherapeutics
as well as radiosensitizers).13,14
*
Corresponding author. Tel.: þ39-051245110; fax þ39-051244064;
e-mail: spinelli@alma.unibo.it
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)01060-3

7190
B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
Therefore, we have now extended our interest to the
reactivity of 1 with a series of ortho-substituted anilines (4o:
X¼OH, NH2, OMe, Me, Et, F, Cl, Br) with the two main
aims of (i) gaining further informations about the influence
of the structure of the nucleophile on the course of the
reaction, and then on the relevant 2o/3o ratios, and (ii)
attaining, through the synthesis of new amines characterised
by the benzo[b]thiophene skeleton, a better knowledge on
structure/activity relationships in thiophenes and ben-
zo[b]thiophenes.
mixing of the operating proximity effects (electronic and
steric), are always less nucleophilic and basic than the
corresponding para-substituted derivatives. This agrees
with the well known observation that para-substituted
anilines are usually more reactive as nucleophiles than
ortho-substituted anilines at least by a factor of ten (for
example, in substrates such as halogenonitrothiophenes,
which are relatively not very sensitive to steric effects),18
also because of the lower basicity of the latter (differences
ranging between 0.5 and 1.8 pKa units).19
Furthermore, following our investigation of the 13C NMR15
and MS16 properties of the corresponding 2m and 2p, we
herein report on both the 13C SCS (substituent-induced
chemical shift) and MS behaviour of 2o, in order to make a
comparison of substituent effects and an estimation of the
influence of proximity effects in such compounds. A
comparison of substituent effects in the two series of
ortho-substituted compounds obtained (2o and 3o) has also
been possible.
First of all it must be remarked that, in the experimental
conditions adopted, no nucleophilic substitution occurred
with ortho-substituted anilines containing strong electron-
withdrawing groups such as CF3 or NO2. According to the
general behaviour observed in the anilino-debromination
reaction,1–3 a relation between the reaction time and the
electronic effects of the substituent present in the aniline has
been observed: thus, reaction times vary from 30 min for
X¼ortho-OMe to ca. 3 h for X¼ortho-Cl. The case of the
reaction with ortho-bromoaniline appears interesting, as a
reaction time much longer (10 h) than that required by all
the other ortho-substituted anilines is associated with a low
global yield (20%), probably due to a significant time-
dependent decomposition of the reactants and/or products.
2. Results and discussion
2.1. Reactivity of 1 with ortho-substituted anilines:
synthesis of 2o and 3o
A comment on the trend of the 2o/3o ratios, calculated by 1H
NMR analysis of the final reaction mixtures is also possible,
e.g. as a dependence on the substituent electronic effects
(measured by the Hammett para-substituted constants):20
with regard to the resonance effects, in first approximation
an ortho-substituent can be assimilated to a para one. Thus,
anilines containing electron-repelling or donating sub-
stituents (X¼Et, Me, or OMe) furnish predominantly
(2o/3o, 3.2–8.1) or exclusively (X¼OH and NH2; note
that ortho-phenylenediamine is the strongest base/nucleo-
phile we have used) the relevant 2o, while ortho-chloro and
ortho-bromoaniline give a prevalence of the relevant 3o
(2o/3o, 0.7–0.9). Interestingly ortho-fluoroaniline, as fore-
seeable on the basis of the Hammett para-substituent
constant, shows an intermediate behaviour (2o/3o, 1.4).
The title reactivity and the influence of the electronic and
proximity effects of the ortho-substituent on the course of
the reaction have been evaluated by carrying out the
reaction between 1 and various 4o in DMF at reflux until
complete disappearance of 1, in the presence of triethyl-
amine. As we have previously pointed out4,6 the best results
(in terms of both yields and reaction times) are obtained
carrying out the reaction in DMF [a solvent with a high
hydrogen-bond donor (HBD) character]17 and using a 1:3:3
substrate to aniline to Et3N ratio. For the sake of comparison
with previous results11,12 the reactivity of 1 has been tested,
with two model ortho-substituted anilines (4ob–oc X¼NH2
and OMe), also in the absence of non-nucleophilic bases
obtaining, as expected,6 only the relevant 2o in excellent
yields (93 and 99%, respectively).
For the sake of comparison we have also studied the reaction
of 1 with para-chloroaniline in the presence of a non-
nucleophilic base to gain information on the effect of the
position of the substituent in the arylamino ring. We have
thus observed that para-chloroaniline reacts with 1 faster
The obtained data, collected in Table 1, allow some
considerations on both the reaction times necessary to
complete the reaction and the 2o/3o ratios. It must be
remembered that ortho-substituted anilines, because of the
Table 1. Reaction of 3-bromo-2-nitrobenzo[b]thiophene (1) with ortho-substituted anilines (4o) in DMF (1208C) in the presence of Et3N until disappearance of
starting product 1
Compounds
X
Time (min)
Yield (%)
%2o–%3o
Mixture separation
a
b
c
d
e
g
h
i
OH
NH2
OMe
Me
Et
60a
65
30
90
120
150
175
600
30
19
82
84
73
83
46
20
100:0
100:0
89:11
77:23
76:24
58:42
47:53
42:58
Crystallization from ethanol
Ethyl acetate–petroleum ether¼1:4 on SiO2
Fractional crystallization from ethanol
Ethyl acetate–petroleum ether¼1:8 on Al2O3
Ethyl acetate–petroleum ether¼1:9 on Al2O3
Ethyl acetate–petroleum ether¼1:6 on Al2O3
Ethyl acetate–petroleum ether¼1:7 on Al2O3
Ethyl acetate–petroleum ether¼1:7 on Al2O3
F
Clb
Br
a
Temperature 508C; see text and experimental.
b
In comparable experimental conditions the reaction with para-chloroaniline goes to completion into 80 min with high yield (90%) and a high 2/3 ratio (4.9).
2ph (see Ref. 12); 3ph, mp 2348C from ethanol, 1H NMR (d): 7.31 (1H, dd, J¼8.0, 7.3 Hz, H6), 7.47 (1H, dd, J¼8.0, 7.3 Hz, H5), 7.61 (m, 4H, H20, H30, H50,
H60), 7.77 (1H, d, J¼7.8 Hz, H7), 8.31 (1H, d, J¼8.0 Hz, H4), 11.24 (1H, s exch, NH); HRMS: found 304.00721, calcd 304.00733.

B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
7191
or in the presence of added bases (Et3N or K2CO3) the only
observed product has been 2oa: neither 3oa nor the relevant
ether 5 has been observed. Perhaps a fast Smiles-like
rearrangement21 would prevent the detection of any formed
5. The detected highest yield in 2oa has been observed at
room temperature (63%, no base added), the increase of
either temperature (508C) or concentration of the added
bases causing a significant yield reduction.
In conclusion the systematic study of the reaction of 1 with
several ortho-substituted anilines has shown a strong
dependence of the reactivity as well as of the 2o/3o ratio
on the proximity (steric and electronic) effects of the ortho-
substituent.
2.2. Analysis of 13C NMR data of 2o and 3o in DMSO-d6
and CDCl3 solutions
Anisotropy, resonance and inductive effects affect 13C
chemical shifts (c.s.’s) of carbon atoms of aromatic and
heteroaromatic rings: the superimposition of these effects22
usually prevents the occurrence of mono- (Hammett) or bi-
parametric (DSP, dual substituent parameter)23 linear free-
energy relationships (lfer’s) for the relevant SCS values,
when substituent and probe carbon atom are in the same
ring. In contrast, the effect of a substituent not directly
linked to the ring containing the probe carbon atom leads to
SCS values (possibly small) that can be correlated by means
of mono- or bi-parametric lfer’s.7,15,24,25
Scheme 2. Conditions and yields: (i) rt, 5 days (63%); (ii) 508C, 90 min
(30%); (iii) 508C, Et3N, 60 min (30%); (iv) 508C, K2CO3, 30 min (41%);
(v) 508C, K2CO3 (large excess), 15 min (25%).
than ortho-chloroaniline (reaction time, 80 min vs 3 h) and
gives a higher 2/3 ratio (4.9 vs 0.7). These results are well in
line with previsions: in fact, on going from ortho- to para-
chloroaniline both the basicity (DpKa ca. 1.3)19 and the
nucleophilicity (for example, in simple thiophene systems
the reactivity ratios with the two anilines are about 102)18
increase, thus largely favouring the normal SNAr pathway
with respect to the base-catalysed nucleophilic substitution
with rearrangement.
Thus, investigating the 13C NMR data of several 2m and 2p
in DMSO-d6 solutions we have evidenced the occurrence of
an alternate polarisation, which can involve carbons C-3,
C-2, C-3a, C-7a, C-4 and C-5 of the 2-nitrobenzo[b]thio-
phen-3-yl moiety. A DSP treatment of the 13C SCS indicates
a large and a low resonance contribution for aryl para- and
meta-substituents, respectively, while the inductive com-
ponent remains constant throughout.15
The reaction of 1 with ortho-aminophenol (4oa) has been
deeply investigated. This aniline is a bidentate nucleophile
which could, at least in principle, behave as a nitrogen- as
well as an oxygen-nucleophile. A choice between the two
different possibilities should be possible by changing the
experimental conditions: in the absence of bases the
aminogroup should be effective, while in the presence of
bases the ionised phenolic hydroxyl could be the nucleo-
philic species. Some complications could be dependent on
the fact that with a substrate such as 1 the added bases could
affect the course of the reaction with a weak amine
(Scheme 2).
However in compounds 2o and 3o some further compli-
cation could be expected considering the proximity effects
that the substituent could exert.
2.2.1. 13C NMR data in DMSO-d6. In Tables 2 and 3 13
C
SCS data of 2o and 3o in DMSO-d6 are collected.
Notwithstanding the absence of compounds containing
strong electron-withdrawing substituents (see above) a
significant set of substituents has been investigated (ranging
from OH, NH2 and OMe to halogens). A rough examination
Interestingly it must be remarked that operating at room
temperature as well as at 508C (at higher temperature, e.g.
808C, only tar formation has been observed) in the absence
Table 2. 13C NMR substituent-induced chemical shifts (SCSs) for 3-anilino-2-nitrobenzo[b]thiophenes (2o) in DMSO-d6
Compounds
X
C-2
C-3
C-3a
C-4
C-5
C-6
C-7
C-7a
C-10
C-20
C-30
C-40
C-50
C-60
f
H
124.42 142.71 129.35 126.19 124.59 130.72 123.78 137.49 139.26 124.19 129.18 126.21 129.18 124.19
a
b
c
d
e
g
h
i
OH
NH2
OMe
Me
Et
F
Cl
Br
DSCSa
22.53
22.65
21.89
22.55
22.53
20.22
20.75
20.98
2.65
2.70 20.21 20.40
2.40 20.51 0.25
1.95 20.29 20.57
2.47 20.57 20.38
2.72 20.76 20.05
0.13
0.18
0.15
0.24
0.12
0.50
0.54
0.44
0.54
0.30 20.03
0.42 20.02
0.08 213.90
0.52 216.87
0.05 212.67
28.07 212.74
22.27 213.67
29.51 216.86
2.66
29.73
3.09 213.01
3.23
3.85
2.59
3.18
3.44
3.51
4.12
4.52
4.52
0.33
0.32
0.35
0.20
0.33
0.30
0.42
0.07
0.16
0.19
2.56
2.03
2.38
2.40
2.89
3.20
3.20
28.44
22.09
22.09
24.13
20.62
20.33
13.01
0.46
0.57
21.69
22.37
10.76
16.30
1.80
0.22
0.18
0.16 21.03
0.16 20.25 212.10
32.24 212.90
0.82 20.23 20.81
0.89 20.41 20.71
0.29
0.23
0.32
0.05
0.07
0.82
22.88
21.57 23.13
16.87 35.37
6.02
0.98
4.06
2.72
0.92
1.28
20.92
a
DSCS range of substituent effect on chemical shifts.

7192
B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
Table 3. 13C NMR substituent-induced chemical shifts (SCS’s) for 2-anilino-3-nitrobenzo[b]thiophenes (3o) in DMSO-d6
Compounds
X
C-2
C-3
C-3a
C-4
C-5
C-6
C-7
C-7a
C-10
C-20
C-30
C-40
C-50
C-60
f
H
162.42 119.94 130.94 121.23 126.69 124.74 122.34 125.02 139.23 124.60 129.66 127.68 129.66 124.60
c
d
e
g
h
i
OMe
Me
Et
F
Cl
20.79
1.75 20.77
2.08 20.72
0.10 20.26
0.15
0.02
0.01
0.17
0.00 20.16
0.01 20.13
0.16
0.13
0.09
0.07
0.24 211.83
27.33 217.17
0.91 28.72 21.25
0.13
0.37
0.24
0.22
20.73
21.43
9.94
15.61
1.71
0.05
1.11 22.25
1.41 22.27
2.19 24.36
2.28 20.86
2.56 20.30
1.83
2.10
2.72
3.22
3.36
4.61
0.14
0.07 20.05
0.00 20.03 20.12 20.11 20.01 212.29
31.34 213.05
0.41 20.15
0.45 20.25
0.11
0.04
0.63
0.17
0.11
0.17
0.21
0.16
0.24
0.10
0.05
0.32
0.22
0.16
0.33
0.25
0.25
0.26
22.64
21.17 23.91
12.29 35.25
5.30
0.94
4.05
Br
DSCSa
2.87
0.87
21.22
2.56
8.72
a
DSCS range of substituent effect on chemical shifts.
of data evidences the absence of the expected substituent
effects on SCS: for example, in the 2o series both electron-
withdrawing and -repelling substituents cause shielding and
deshielding of C-2 and C-3 (the carbon carrying the anilino
moiety), respectively. Also in the case of 3o the SCS
variations can not be related to the electronic effects of the
substituents, essentially paralleling the situation observed
for 2o.
2.2.1.1. 13C NMR Data in CDCl3. In order to lower the
proximity effects caused by the solvent used we have also
collected 13C c.s. values of 2o and 3o in CDCl3 solutions
(data in Tables 5 and 6), that is in a solvent with a
completely different ability to behave as HBD solvent.17 In
this solvent, though, 2oa is so little soluble that we have not
recorded its spectrum.
In the 2o series electron-withdrawing and -repelling
substituents cause shielding and deshielding of C-3
(DSCS 3.27 ppm), respectively, while the usual inverted
effect can be evidenced for C-2 (DSCS 2.74 ppm). The
relevant SCS values give good mono-parametric corre-
lations (r 0.97 and 0.92, respectively; data in Table 4,
lines 1 and 2)26 vs. sp2, with susceptibility constants (r 2
26.13 and 5.12, respectively) of the same order of those
calculated for the para-isomer (2p).15 This result can be
related to the absence of the specific solute–solvent
In contrast, the 0cross-correlations for the ipso- (C-20), ortho-
(C-10 and C-3 ) and para-carbons (C-50) of the anilino
moiety of 2o gave good or excellent statistical results versus
the relevant SCS of the ortho-substituted anilines (s 0.84–
1.21, r 0.987–0.999; data in Table 4, lines 6–9) or of
monosubstituted benzenes (data not reported). The same
excellent statistical results have been obtained for the
corresponding carbon atoms of 3o (s 0.80–1.13, r 0.990–
0.999; data in Table 4, lines 10–13).
Table 4. Statistical data for the single-parameter analysis of SCS values of carbon atom of 2o and 3o and for the cross-correlation with anilines (4o)
Line
Solvent
Probe atom Series r^sr (or b^sb) Substituent constanta or SCS of probe atom Series
i^si
n
r
CL.%
1
2
3
4
5
6
7
8
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
DMSO-d6 C-10
DMSO-d6 C-20
DMSO-d6 C-30
DMSO-d6 C-50
DMSO-d6 C-10
DMSO-d6 C-20
DMSO-d6 C-30
DMSO-d6 C-50
C-2
C-3
C-7a
C-2
C-3
2o
2o
2o
3o
2o
2o
2o
2o
2o
3o
3o
3o
3o
5.12^0.71
26.13^0.56
20.96^0.30
29.40^1.09
0.74^0.04
1.21^0.07
0.84^0.05
0.98^0.02
1.16^0.03
1.13^0.07
0.80^0.05
0.98^0.02
1.12^0.04
sp2
20.17^0.14
0.07^0.11
20.15^0.06
0.48^0.19
20.12^0.06
1.06^0.52
3.11^0.95
0.76^0.17
0.64^0.18
1.34^0.47
2.42^1.00
0.72^0.16
0.42^0.14
8
8
8
6
6
8
8
8
8
7
7
7
7
0.924 99.9
0.974 99.9
0.781 99.0
0.982 99.9
0.995 99.9
0.987 99.9
0.989 99.9
0.999 99.9
0.997 99.9
0.990 99.9
0.990 99.9
0.999 99.9
0.997 99.9
sp2
sp2
sp2
C-2
C-1
C-2
C-3
C-5
C-1
C-2
C-3
C-5
3o
4o
4o
4o
4o
4o
4o
4o
4o
9
10
11
12
13
r, susceptibility constant for the single-parameter analysis; b, slope of the cross correlation; i, intercept; sr and sb standard deviations; n, number of points; r,
correlation coefficient; CL, confidence level.
a
Substituent constants from Ref. 20.
Table 5. 13C NMR substituent-induced chemical shifts (SCSs) for 3-anilino-2-nitrobenzo[b]thiophenes (2o) in CDCl3
Compounds
X
C-2
C-3
C-3a
C-4
C-5
C-6
C-7
C-7a
C-10
C-20
C-30
C-40
C-50
C-60
f
H
125.00 144.26 128.78 127.17 124.33 130.61 123.49 138.87 138.60 125.42 129.73 127.41 129.73 125.42
b
c
d
e
g
h
i
NH2
OMe
Me
Et
F
Cl
21.06
20.59
21.51
21.47
0.74 20.69
1.09 21.52
1.23 21.56
2.74 3.27
1.71
0.36
0.02 20.39
0.17 20.60 20.03
0.46
0.24 20.15
0.04 214.98
0.04 20.18 20.31 211.78
17.74 213.46
27.98 217.99
2.26 210.77
2.75
0.43
1.51
1.73
1.80
0.92
1.09
2.75
0.97
0.95
1.18
1.35
0.56
0.86
2.26
29.04
22.48
22.58
24.89
22.08
21.41
10.77
1.19 20.04 20.64
1.35 20.12 20.35
0.28
0.16
0.33
0.31 20.02
0.30 20.04
0.18 20.03
0.14 20.01
0.08
0.18
21.68
22.18
9.36
15.09
1.66
0.02
0.13 21.07
0.08 20.83
0.07 20.69
0.05
0.04 20.34 211.99
31.46 212.89
0.01 20.46
0.07 20.38
22.57
21.07 25.46
14.98 36.92
4.26
0.79
3.99
Br
DSCSa
0.29
1.07
0.49
0.28
0.25
0.64
21.98
a
DSCS range of substituent effect on chemical shifts.

B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
7193
C-60
Table 6. 13C NMR substituent-induced chemical shifts (SCSs) for 2-anilino-3-nitrobenzo[b]thiophenes (3o) in CDCl3
Compounds
X
C-2
C-3
C-3a
C-4
C-5
C-6
C-7
C-7a
C-10
C-20
C-30
C-40
C-50
f
H
160.92 121.75 131.06 122.64 127.06 125.22 121.49 125.28 138.28 122.49 129.98 127.32 129.98 122.49
c
d
e
g
h
i
OMe
Me
Et
F
Cl
22.05
1.72
2.08
20.86
22.01
21.65
4.13
0.26 20.24 20.12 20.06 20.11
0.06
0.11
0.11
0.11
0.06
0.08
0.11
0.16 210.70
28.04 218.62 20.32 29.14 22.88
1.20
1.17
0.34 20.04 20.02 20.07
0.28 20.16 20.09 20.10
0.25
0.19
21.11
21.74
10.42
16.31
32.47 213.18
1.73
0.00
0.93 22.51
1.26 22.58
0.90 24.95
1.16
1.45
0.44
0.65 20.03
0.71 20.23
0.59 20.17
0.15
0.05
0.02
0.31
0.18
0.13
0.11
0.27
0.14
0.14
0.13
0.25
0.11 211.55
0.08
0.00
0.25
22.62
21.24 24.81
11.55 37.28
4.74
0.64
3.90
0.16 21.97 20.74
0.70 21.28 20.19
Br
DSCSa
1.20
0.58
22.52
1.58
9.14
4.33
a
DSCS range of substituent effect on chemical shifts.
interactions that, in contrast, characterise the c.s.’s in
DMSO-d6.
and 3o are reported in Table 7. Chloro and bromo derivatives
are considered as effectively monoisotopic species, i.e. the
relative abundance of halogen-containing ions has been
calculated by summing up all isotopic contributions.
Concerning the 2o series the small DSCS values (0.3–
1.1 ppm) and the occurrence of some scattered behaviour
for the other carbons of the benzo[b]thiophene moiety
discourages the search for lfer’s, except for C-7a. In this
instance only a rough correlation has been observed by
means of a mono-parametric lfer treatment (r 2 20.96, r
0.78; data in Table 4, line 3). Interestingly a dissociation of
electronic effects in their components (inductive and
resonance contributions) strongly improves the statistical
results (rI 21.17 and r2R 20.29, R 0.945): this occurrence
can be related to the peculiar r2R /rI ratio observed (l ca.
0.25), unusually low for conjugated systems. As a matter of
fact the l value calculated recalls the figure observed in
meta-substituted compounds,15 indicating that anyhow
some significant steric hindrance operates, thus lowering
the resonance effects.
The molecular ion is always responsible for the base peak of
the mass spectra except for 2ob and 2oi (X¼NH2 and Br,
respectively). In the case of 2oi this should be related to the
presence of the feeble CAr–Br bond: as a matter of fact in
both 2oi and 3oi all the –Brz peaks ([M2NO2X]þz,
[M2HNO2X]þ, [M2OHX]þz) show high relative
abundance (see after).
The following fragmentation pathways have been observed:
(a) loss of NOz, (b) loss of NO2z and HNO2, (c) loss of OHz,
(d) loss of Xz and (e) formation of XC6H4Oþ. These
pathways are reported in Scheme 3 and strictly recall those
observed for the previously studied 2m and 2p.
Some considerations can be made on the results obtained.
For compounds of the 3o series a somewhat different
behaviour has been observed: only the C-2 c.s. shows the
expected substituent effects. As a matter of fact, electron-
withdrawing and -repelling substituents cause shielding and
deshielding on c.s. (DSCS 4.15 ppm), giving good mono-
parametric correlations (r 0.98; data in Table 4, line 4) vs.
s2p with a susceptibility constant (r 2 29.40) comparable
with that calculated for C-3 of 2o. Therefore a good cross-
correlation (s 0.74, r 0.995; data in Table 4, line 5) can be
evidenced between the SCS’s of C-3 of 2o and of C-2 of 3o,
that is between the SCS’s of the two carbon atoms directly
linked to the anilino moiety.
2.3.1. Loss of NOz. The nitro to nitrite isomerization
followed by loss of NOz is a well-established process in EI-
MS of nitro-aromatic or -heteroaromatic compounds. A
comparison between the two series (2o and 3o) gives an
interesting information: the [M2NO]þ peak is usually of
low intensity and more abundant in 3o than in the relevant
2o, except than for 2oc (X¼OMe), for which a significant
fragmentation occurs (30%).
2.3.2. Loss of NOz2 and HNO2. The peaks for the relevant
two ions are present in all the examined compounds but in
2oi (X¼Br). The sum of the two peaks is more abundant in
2o than in the corresponding 3o except for 2oi (X¼Br),
where all the peaks related to –Xz are prevalent, and once
more again for 2oc (X¼OMe). In the case of 2ob (X¼NH2)
[M2NO2]þ is the base peak. Considerations concerning the
HNO2 loss have been previously discussed and a 10H-
[1]benzothieno[3,2-b]indole structure for the formed cation
has been proposed.16
Anyhow, also in CDCl3 the cross-correlations of the SCS’s
for the ipso- (C-20), ortho- (C-10 and C-30) and para-carbons
(C-50) of the anilino moiety of 2o and 3o gave good or
excellent statistical results versus the relevant SCS’s of
ortho-substituted anilines or of monosubstituted benzenes
(data not reported), paralleling the situation observed in
DMSO.
The whole of the 13C NMR results furnishes a new
interesting example of lfer concerning ortho-substituted
compounds, which strictly recalls that observed in the case
of ortho-substituted 5-nitro-3-thiophenecarboxanilides
previously studied by us.7
The loss of Xz together with NO2z and HNO2 will be
discussed hereinafter.
2.3.3. Loss of OHz. The loss of OHz causes the formation of
[M2OH]þ and [M22OH]þz: the second being more
abundant than the first one. In some cases the [M22OH]þz
ion is responsible for a peak with a relative abundance
higher than 33% (3od, of–oh: X¼Me, H, F, Cl; 2od:
X¼Me).
2.3. Analysis of MS of 2o and 3o
The main peaks of the 70 eV mass spectra of compounds 2o

7194
B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
Scheme 3. Main fragmentation pathways of compounds 2o and 3o.
The formation of these two ions reflects a typical ortho-
effect observed in the mass spectra of several classes of
ortho-nitro-aromatic (or -heteroaromatic) amines: well
known examples are those of ortho-nitrodiarylamines,27–29
ortho-nitroanilines,29–31 ortho-nitroaminopyridines32 and
aryl(2-nitrobenzo[b]thiophene-3-yl)amines.16 Studies per-
formed by using substrates labelled with deuterium at the
nitrogen of the amino group as well as at the phenyl ring or
at the alkyl group linked to the aminogroup evidenced a
significant preference of OHz loss from the hydrogen of the
amino group, the preference being enhanced in the
metastable decomposition. Anyhow the loss of two OHz
gave the same final ion16,28 for which a [1]benzothieno[2,3-
b]quinoxaline structure has been proposed.16
2.3.4. Loss of Xz. The formation of several ions deriving
from this loss has been observed, the most common being
[M2NO2X]þz, [M2HNO2X]þ, [M2OHX]þz. The direct
occurrence of significant [M2X]þ peaks has been observed
only in the case of 2oi (5.5%) and 3oi (ca. 1%), according
with the peculiar weakness of the CAr–Br bonds.
The formation of the [M2NO2X]þz ion by further
fragmentation from the [M2NO2]þ one competes with
formation of the [M2HNO2]þz ion. The relative abundance
of the [M2HNO2]þz and [M2NO2X]þz ions might be
related to the CAr–X bond strength: in the fluoro and amino
derivatives the formation of the [M2HNO2]þz ion largely
prevails, while in the chloro, bromo and ethyl derivatives
the formation of the [M2NO2X]þz ion is overwhelming.
Methyl and methoxy derivatives show non-homogeneous
behaviours: prevalent formation of [M2HNO2]þz and
[M2NO2X]þz has been observed for 2oc and 3oc
(X¼OMe), respectively, while in 2od and 3od (X¼Me) a
slight preference of [M2NO2X]þz and similar quantities of
the two ions could be detected, respectively. Concerning the
[M2HNO2X]þ ion, its formation, always observed but for
3oc, is particularly important in chloro and bromo
derivatives (31–51%).

B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
7195
In the formation of the [M2OHX]þz ions a situation similar
to that seen in the case of the [M2NO2X]þz ions has been
observed.
compare formation of M2NOþ2 and of M2HNO2þz): this can
depend, at least in part, on the higher stability of 3of with
respect to 2of (DE 16.8 kJ mol21, see after). The higher
stability of 3o with respect to 2o causes even larger effects in
the case of the 2oi/3oi couple: for example the molecular ion
(45%) of 2oi is not responsible for the base peak, in contrast
with what happens with 3oi (molecular ion 100%). The
behaviour of 2ob can be related to the same factor.
2.3.5. Formation of [XC6H4O]1. The formation of
significant amounts of this ion has been recently observed
in compounds bearing a strong electron-repelling or
donating substituent in the para-position of the anilino
moiety.16 Its formation results from an oxygen migration
from the ortho-nitro group to other parts of the molecule and
characterises several EI-induced fragmentations. Replace-
ments of an aromatic or heteroaromatic nitrogen by oxygen
from other sources such as carbonyl groups have been
similarly observed.33–35 The driving force for such
rearrangements appears to be the stabilisation of the
ortho- or para-quinoid structure.16,33
2.4. Ab initio calculations of the stability and the
geometry of some 2o and 3o compounds
The course of the reaction of 1 with weak nucleophiles in
the presence of non-nucleophilic bases with formation of an
unexpected substitution product (3) together with the
expected one (2) raises an intriguing question. Because
bromine is a good leaving group, the reaction of
nitroactivated bromoaromatic (or heteroaromatic) substrates
would give only the expected product of reaction (that is the
ipso-substitution product);36 then the question is: why does
1 show this unexpected behaviour? That is, which is the
driving force that induces this unexpected behaviour? In an
attempt to reply to this question we have carried out an ab
initio computation to gain information on the geometry and
the stability of 2of and 3of (the parent compounds of the two
series) in the gas phase. For the sake of comparison we have
also carried out the same calculations on the 2oi and 3oi
couple, that is on the anilino compounds containing a bulky
and electron-withdrawing substituent linked to the phenyl
ring with a feeble CAr-Br bond (i.e. factors which could
As a matter of fact significant amounts of these ions have
been observed when methyl, ethyl, methoxy and amino
groups are present.
The whole of the obtained results (loss of NOz, NO2z , OHz
and 2OHz, Xz and the formation of [ArO]þ) strongly
supports the results of the EI-induced fragmentation
evidenced in 2m and 2p compounds.
Interestingly, ab-initio computations can help in under-
standing some EI-MS results. If we look at the behaviour of
2of and 3of we can observe that fragmentations of the
former are higher than those of the latter (for example,
Figure 1. Optimised geometries for 2of and 3of (conformers I and III).
Figure 2. Optimised geometries for 2oi and 3oi (conformers V and VI).

7196
B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
˚
Table 8. Bond lengths (A) for I, III, V and VI calculated using DZVP basis
set in the gas phase and for VII39
one. In contrast, conformer IV of 3of, with the phenyl
group directed towards the nitro group, is much less stable
(DE 41.9 kJ mol21) as it can not form any hydrogen bond;
moreover the phenylamino and the thienyl group are not
coplanar because of the repulsion between the phenyl and
the nitro group (dihedral angle between thienyl and
phenylamino rings 428, with the nitro group rotated of 118
and moreover 78 out of the plane with respect to the thienyl
ring). Looking at 2of we observe that it cannot be planar
because of steric requirements: in the more stable conformer
(I), with the proton of the phenylamino group [N(H)]
directed towards one of the two oxygens of the nitro group
I
III
V
VI
VII
S/C-2
1.758
1.399
1.408
1.753
1.414
1.757
1.396
1.412
1.756
1.411
1.754
1.355
C-2/C-3
C-2/NO2
C-3/NO2
C-3/NH
C-2/NH
C-3/C-3a
C-3a/C-4
C-4/C-5
C-5/C-6
C-6/C-7
C-7/C-7a
C-3a/C-7a
1.414
1.418
1.368
1.372
1.351
1.454
1.410
1.395
1.404
1.396
1.395
1.412
1.358
1.454
1.411
1.395
1.404
1.396
1.395
1.411
1.458
1.412
1.389
1.410
1.391
1.402
1.418
1.456
1.412
1.389
1.410
1.391
1.402
1.418
1.446
1.423
1.368
1.397
1.379
1.396
1.399
˚
(a hydrogen bond weaker than in III is present: 1.922 A),
the coplanarity is prevented by the repulsive interactions
between the C–4(H) and the C–20(H) protons, that force the
phenylamino ring to be rotated with respect to the thienyl
one of about 358. Therefore every attempt to calculate the
energy content for a stable coplanar structure (see above) of
2of fails, because a planar structure would requires a big
deviation of bond angles from their normal values, making
this conformation energetically and then geometrically
‘impossible’. In conformer I the distance between the two
Table 9. Bond angles (8) for I, III, V and VI calculated using a DZVP basis
set in the gas phase and for VII39
I
III
V
VI
VII
S/C-2/C-3
S/C-2/NO2
S/C-2/NH
115.3
118.0
111.1
124.2
115.3
118.0
110.6
124.3
112.4
a
˚
protons above is 2.889 A, i.e. higher than the contact
distance between two protons, leading to a lower stability of
b
C-2/C-3/NH
C-2/C-3/NO2
C-2/C-3/C-3a
C-3/C-3a/C-7a
C-3a/C-4/C-5
C-4/C-5/C-6
C-5/C-6/C-7
C-6/C-7/C-7a
C-7/C-7a/C-3a
C-7a/C-3a/C-4
C-7a/S/C-2
123.4
123.6
the conformer I with respect to III (DE 16.8 kJ mol21).
122.5
113.9
111.2
119.5
121.4
120.1
118.2
122.8
117.9
92.0
122.6
114.1
111.0
119.5
121.5
120.0
118.1
123.1
117.8
92.2
110.0
111.9
119.7
120.7
120.7
118.7
121.4
118.8
89.2
110.1
111.9
119.7
120.7
120.7
118.7
121.4
118.9
89.3
112.4
112.9
117.9
123.1
119.6
118.5
122.1
118.7
91.7
In conformer II, the least stable among the four considered,
the interaction between the phenylamino ring and the nitro
group prevents molecular coplanarity (see above, the
phenylamino ring appear rotated with respect to the thienyl
one of about 448), and the repulsive interactions between the
C–4(H) and the N(H) protons appear less important
˚
(distance between the two protons 2.151 A). In this case,
a
as in IV, no hydrogen bond is present. Surely the absence in
II and IV of a more or less strong hydrogen bond causes the
highest energy content (that is the lowest stability); their
energy contents are similar (DE 4.6 kJ mol1) and both
conformers are non-planar.
S/C-2/CH3 118.48.
C-2/C-3/CH3 123.88.
b
affect the 2o/3o ratios as well as the course of the EI-MS
fragmentations).
In conclusion III represents the most stable conformer
(DE 16.8, 41.9 and 46.5 kJ mol21 with reference to I, IV
and II, respectively), because of the occurrence of the
strongest hydrogen bond and of the best conjugative
interactions between phenylamino and nitro groups. This
second factor could be responsible for the length of the C-2/
2.4.1. Geometry and stability of 2of, 3of, 2oi and 3oi. Ab
initio computations have been carried out at the density
functional theory (DFT) level37,38 in order to consider
electronic correlations. The results obtained are collected in
Figures 1–2 and in Tables 8 and 9, where bond lengths and
angles for the most stable conformers I, III, V and VI are
reported. For both 2of and 3of we have addressed our
attention to the study of conformers (possibly planar) with
two opposite geometries: the first one with the proton of the
phenylamino group [N(H)], the second with the phenyl
directed towards the nitro group, respectively. Let us now
consider the results obtained for the optimised structures.
˚
˚
C-3 bond of III (1.414 A) being higher than in I (1.399 A),
˚
IV (1.393 A) and II (1.391 A) and at the same time of the
˚
˚
length of the CTh–N(H) bond of III (1.351 A) being lower
˚
than in I (1.368 A), IV (1.370 A) and II (1.377 A).
˚
˚
The other bonds show only small variations in the different
conformers and their values are consistent with those
determined by X-rays measurements for the 5-bromo-2,3-
dimethylbenzo[b]thiophene (VII),39 which is again a 2,3-
disubstituted benzo[b]thiophene as I–VI. For the sake of
comparison, also the experimentally determined39 bond
lengths and angles of VII are collected in Tables 8 and 9,
respectively.
First of all it must be remarked that in all the studied
conformers the individual five- and six-membered rings are
by their selves planar, but the benzo[b]thiophene system as a
whole always shows a (small) deviation from planarity (not
higher than 28).
An interesting outcome is that only in conformer III of 3of
the phenylamino and the thienyl groups are coplanar: III
On the basis of the results obtained with 2of and 3of we have
taken into account, for the 2oi and 3oi couple, only the two
most stable conformers, that is those with the proton of the
arylamino group [N(H)] directed towards one of the two
˚
shows a strong hydrogen bond (1.784 A) between the proton
of the phenylamino group [N(H)] and one of the two
oxygens of the nitro group and comes out as the most stable

B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
7197
oxygens of the nitro group. Once more conformer VI (3oi) is
more stable (DE 11.4 kJ mol21) than V (2oi): it shows a
The ab initio computations herein, providing information
about the ‘relative’ stability of the two isomers, better
enlighten the driving forces of the reaction. The two
energetically favoured conformers of 2of and 3of show a
˚
strong hydrogen bond (1.804 A), the arylamino and the
thienyl groups are coplanar and a good arrangement is
achieved for bromine, which seems able to interact with the
proton of the arylamino group [N(H)]. As matter of fact the
significantly different energy content (DE 16.8 kJ mol21
)
and the formation from D/E of the ‘unexpected’ isomer
would be preferred with respect to that of the less stable
‘expected’ one. Moreover the different 2o/3o ratios as a
function of the strength of the weak nucleophile used (that
is, the more efficient the nucleophile, the higher the ratio)
seem to confirm that the partition between the two reaction
pathways depends on the nature of the nucleophile: good
nucleophiles do not require base-catalysis and then favour
the normal SNAr pathway.6
˚
calculated distance in the NH…Br system is 2.506 A, while
˚
the sum of the Van der Waals radii is not lower than 3.1 A.
The conformer V (2oi) is less stable, has a hydrogen bond
˚
weaker than VI (1.954 A), and, as in I, the arylamino and
the thienyl groups are not coplanar, the arylamino ring being
rotated with respect to the thienyl one of about 398. Again
the more stable VI has a C-2/C-3 bond longer than V (1.411
˚
and 1.396 A, respectively) and a shorter CTh–N(H) bond
˚
(1.358 and 1.372 A, respectively).
2.4.2. Insight into the reaction mechanism. Ab initio
computations can be of help in understanding the course of
the reaction of 1 with weak nucleophiles. Recently, we have
deeply discussed some data on the reactivity of 1 with
several nucleophiles in different experimental conditions:6
the proposed mechanisms are collected in Scheme 4. We
have pointed out that the peculiar energetic situation of
benzocondensed heterocycles, in which the heterocyclic
ring can be hardly considered aromatic, is probably
responsible for the occurrence of two independent reaction
pathways: 1) the normal SNAr reaction (the exclusive
reaction pathway with strong nucleophiles or with weak
nucleophiles in the absence of added non-nucleophilic
bases; presumably one of the reaction pathways with weak
nucleophiles also in the presence of added non-nucleophilic
bases), which gives only the ‘expected’ substitution
product; 2) a base-catalysed reaction pathway,1–3,40,41 that
occurs only in the presence of non-nucleophilic bases and
whose contribution increases with their concentration,
whereby effective deprotonation of A leads to B, and
hence to the anionic intermediate C: the latter evolves
towards cyclic intermediates eventually furnishing the
‘unexpected’ substitution products and, less probably, also
the ‘expected’ one.
3. Conclusions
A coordinated synthetic, spectrometric (13C NMR and EI-
MS) and theoretical study on the reactivity of 1 with some
ortho-substituted anilines and on the relevant reaction
products has been carried out.
The whole of reactivity results has allowed to ascertain
the influence of the nature of the substituent in the aniline
on the course of the nucleophilic displacement (reactivity
as well as 2o/3o ratio). An accurate analysis of 13C and
EI-MS spectra of the obtained 2o and 3o has also been
carried out.
Finally, ab initio computations at DFT level in the gas phase
have furnished useful informations on the different
geometry and stability of the 2o and 3o isomers and of
their conformers. The results obtained have given further
support to the reaction mechanism proposed by us,4–6
ascribing a role to the different stability of the two isomers
2o and 3o (for example, 3of is 16.8 kJ mol21 more stable
than 2of) on the course of the reaction. Such a stability
difference could also be responsible of some EI-MS results.
Scheme 4. Reaction pathways proposed for the formation of the substitution products.

Table 10. Melting point, crystallisation solvent, colour, IR and HRMS of 2oa-oe, og-oi and 3oc-oe, og-oi
Compounds
X
2o: mp (8C); crystallisation solvent; colour; IR
n (cm21 a
HRMS
3o: mp (8C); crystallisation solvent colour; IR
n (cm21
HRMS
)
)
a
OH
224–226; ethanol; red; 3384.3, 3278.9,
3014.2, 1575.6, 1402.7
205–206; ethanol; red; 3486.8, 3393.4,
3248.1, 3009.4, 1570.6, 1539.6, 1482.5,
1404.7, 1311.3
Found: 286.04085, calcd: 286.04121
Found: 285.05698, calcd: 285.05720
–
–
–
–
b
NH2
c
OMe
Me
Et
141–141.5; ethanol; red; 3268.9, 3009.4,
1575.9, 1539.6, 1407.7, 1311.3
Found: 300.05709, calcd: 300.05686
Found: 284.06221, calcd: 284.06195
Found: 298.07767, calcd: 298.07760
Found: 288.03668, calcd: 288.03688
Found: 304.00764b, calcd: 304.00733
212–213; ethanol; yellow; 3258.5, 3019.8,
1555.2, 1498.1, 1384.0
Found: 300.05725, calcd: 300.05686
Found: 284.06226, calcd: 284.06195
Found: 298.07782, calcd: 298.07760
Found: 288.03715, calcd: 288.03688
Found: 304.00716b, calcd: 304.00733
d
e
160.5–161; ethanol; orange; 3248.3, 3009.6,
1570.8, 1544.9, 1482.6, 1404.8, 1316.3
121–122; cyclohexane; orange; 3248.1, 3009.4,
1575.9, 1544.8, 1482.5, 1404.7, 1316.5
174–175; ethanol; orange; 3265.4, 3011.7,
1575.8, 1542.9, 1491.3, 1402.0, 1317.5
192–193; ethanol; pale orange; 3268.9,
3018.8, 1575.9, 1539.6, 1472.2, 1404.7,
1320.0
116–117; cyclohexane; yellow; 3248.3, 3009.6,
1550.1, 1477.4, 1441.1, 1384.0, 1332.1
147–148; cyclohexane; yellow; 3248.1, 3009.4,
1555.8, 1472.2, 1446.2, 1384.0, 1337.3
173–174; ethanol; yellow, 3227.4, 3009.4,
1560.4, 1456.6, 1441.8, 1384.0, 1337.3
161.5–162.5; ethanol; garnet red; 3248.1,
3019.8, 1550.0, 1441.0, 1384.0, 1332.1
g
h
F
Cl
i
Br
196.5–197.5; ethanol; pale orange; 3268.9,
3009.4, 1575.9, 1539.0, 1472.2, 1404.7,
1319.7
Found: 347.95653c, calcd: 347.95681
158–158.5; ethanol; garnet red; 3258.5,
3009.4, 1550.0, 1461.6, 1441.0, 1384.0,
1337.3
Found: 347.95699c, calcd: 347.95681
a
In CHCl3 solution.
35Cl isotope.
79Br isotope.
b
c

B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
Table 11. 1H NMR spectral data of compounds 2o in DMSO-d6
7199
Compounds
X
1H NMR
a
OH
6.93 (1H, dd, J¼7.8, 7.5 Hz, H50); 7.00 (1H, d, J¼8.0 Hz, H30); 7.05 (1H, d, J¼8.5 Hz, H4); 7.17 (1H, dd, J¼8.5, 7.0 Hz,
H5); 7.28 (1H, dd, J¼8.0, 7.5 Hz, H40); 7.34 (1H, d, J¼7.8 Hz, H60); 7.57 (1H, dd, J¼8.1, 7.0 Hz, H6); 7.92 (1H, d,
J¼8.1 Hz, H7); 10.00 (1H, s exch, OH); 10.31 (1H, s exch, NH).
b
c
NH2
OMe
Me
5.40 (2H, s exch, NH2); 6.62 (1H, dd, J¼7.4, 7.4 Hz, H50); 6.85 (1H, d, J¼8.2 Hz, H4); 6.94 (1H, d, J¼8.5 Hz, H30); 7.11–
7.20 (3H, m, H5, H40, H60); 7.56 (1H, dd, J¼8.2, 7.4 Hz, H6); 7.90 (1H, d, J¼8.2 Hz, H7); 10.19 (1H, s exch, NH).
3.70 (3H, s, OMe); 7.02–7.09 (2H, m, H4, H50); 7.10–7.22 (2H, m, H5, H30); 7.39–7.45 (2H, m, H40, H60); 7.58 (1H, dd,
J¼7.8, 7.1 Hz, H6); 7.94 (1H, d, J¼7.8 Hz, H7); 10.29 (1H, s exch, NH).
d
e
2.23 (3H, s, Me); 6.63 (1H, d, J¼8.3 Hz, H4); 7.11 (1H, dd, J¼8.3, 7.8 Hz, H5); 7.31–7.46 (4H, m, H30, H40, H50, H60);
7.56 (1H, dd, J¼7.8, 7.8 Hz, H6); 7.93 (1H, d, J¼8.0 Hz, H7); 10.51 (1H, s exch, NH)
Et
1.10 (3H, t, J¼7.5 Hz, CH3); 2.63 (2H, q, J¼7.5 Hz, CH2); 6.62 (1H, d, J¼8.4 Hz, H4); 7.10 (1H, dd, J¼8.4, 7.3 Hz, H5);
7.31–7.37 (2H, m, H50, H60); 7.44–7.47 (2H, m, H30, H40); 7.55 (1H, dd, J¼8.0, 7.3 Hz, H6); 7.93 (1H, d, J¼8.0 Hz, H7);
10.56 (1H, s exch, NH).
g
h
i
F
7.12 (1H, d, J¼8.3 Hz, H4); 7.23–7.47 (4H, m, H5, H40, H50, H60); 7.52 (1H, dd, J¼8.0, 8.0 Hz, H30); 7.61 (1H, dd, J¼8.0,
7.7 Hz, H6); 7.98 (1H, d, J¼8.0 Hz, H7); 10.32 (1H, s exch, NH).
Cl
Br
6.77 (1H, d, J¼8.0 Hz, H4); 7.20 (1H, dd, J¼8.0, 7.7 Hz, H5); 7.45–7.49 (2H, m, H40, H50); 7.54–7.69 (3H, m, H6, H30,
H60); 7.97 (1H, d, J¼8.0 Hz, H7); 10.39 (1H, s exch, NH).
6.77 (1H, d, J¼8.0 Hz, H4); 7.19 (1H, dd, J¼8.5, 8.0 Hz, H5); 7.40–7.62 (4H, m, H6, H40, H50, H60); 7.84 (1H, d,
J¼8.0 Hz, H30); 7.97 (1H, d, J¼8.5 Hz, H7); 10.48 (1H, s exch, NH).
Table 12. 1H NMR spectral data of compounds 3o in DMSO-d6
Compounds
X
1H NMR
c
d
e
g
h
i
OMe
Me
Et
3.89 (3H, s, OMe); 7.13 (1H, dd, J¼7.8, 7.0 Hz, H50); 7.25–7.44 (3H, m, H6, H30, H40); 7.49 (1H, dd, J¼7.8, 7.8 Hz, H5);
7.62 (1H, d, J¼7.7 Hz, H60); 7.81 (1H, d, J¼7.3 Hz, H7); 8.33 (1H, d, J¼7.8 Hz, H4); 11.33 (1H, s exch, NH).
2.29 (3H, s, Me); 7.28 (1H, dd, J¼8.0, 7.8 Hz, H6); 7.37–7.53 (5H, m, H5, H30, H40, H50, H60); 7.72 (1H, d, J¼8.0 Hz, H7);
8.31 (1H, d, J¼7.8 Hz, H4); 11.14 (1H, s exch, NH).
1.14 (3H, t, J¼7.4 Hz, CH3); 2.65 (2H, q, J¼7.4 Hz, CH2); 7.28 (1H, dd, J¼7.7, 7.2 Hz, H6); 7.37–7.52 (5H, m, H5, H30,
H40, H50, H60); 7.72 (1H, d, J¼7.2 Hz, H7); 8.30 (1H, d, J¼7.6 Hz, H4); 11.17 (1H, s exch, NH)
7.31 (1H, dd, J¼7.8, 7.6 Hz, H6); 7.36–7.54 (4H, m, H5, H40, H50, H60); 7.70 (1H, dd, J¼8.2, 7.7 Hz, H30); 7.77 (1H, d,
J¼7.6 Hz, H7); 8.31 (1H, d, J¼7.8 Hz, H4); 11.11 (1H, s exch, NH).
F
Cl
7.32 (1H, dd, J¼7.8, 7.3 Hz, H6); 7.46–7.58 (3H, m, H5, H40, H50); 7.73–7.79 (3H, m, H-7, H30, H60); 8.32 (1H, d,
J¼7.5 Hz, H4); 11.28 (1H, s exch, NH)
Br
7.31 (1H, dd, J¼8.4, 7.3 Hz, H6); 7.40–7.50 (2H, m, H5, H40); 7.59 (1H, dd, J¼8.7, 7.6 Hz, H50); 7.73–7.79 (2H, m, H7,
H60); 7.89 (1H, d, J¼7.9 Hz, H30); 8.32 (1H, d, J¼7.9 Hz, H4); 11.26 (1H, s exch, NH)
4. Experimental
the precipitate filtered, dried under vacuum and purified
by crystallisation or flash-chromatography on silica gel.
Melting points were determined using a Kofler apparatus
and are uncorrected. NMR spectra were recorded on a
Varian Gemini 300 Instrument in the Fourier transform
mode at 21^0.58C in DMSO-d6 and in CDCl3. 1H
(300.07 MHz, 0.02M) and 13C (75.43 MHz, 0.1 M) chemi-
cal shifts (d) are in ppm relative to TMS as secondary
internal reference. High resolution mass spectra (HRMS)
were recorded on a VG70 70E apparatus. IR spectra were
obtained with a Perkin–Elmer Spectrum RX IFT-IR System
in CHCl3 solution. Solvents were removed under reduced
pressure. TLC: precoated silica gel (Merck F254) or
aluminium oxide neutral (Merck art. 5550) plates. Flash
chromatography: silica gel 60 (ICN Silica 32–63) or
aluminium oxide neutral (Merck art. 1097). All new
compounds gave good C, H, N and S analysis. The substrate
(1) was obtained in 98% purity according to a literature
procedure.42
Reaction times, yields, isomeric ratios and separation
method are reported in Table 1. Crystallisation solvent,
melting point, colour, IR and HRMS are reported in
Table 10 and 1H NMR in DMSO-d6 are collected in
Tables 11 and 12, respectively.
4.2. Procedure for the reactions of 1 with anilines 4ob–
oc in the absence of Et3N
The appropriate aniline (4.65 mmol) was added to a solution
of 1 (0.400 g, 1.55 mmol) in DMF (4.5 mL) and the mixture
kept at 1208C until disappearance of 1. The mixture was
then cooled to room temperature, poured into ice/water and
the precipitate filtered, dried under vacuum and crystallised
from EtOH to give 2o as the exclusive product (2ob: 1 h;
93%. 2oc: 30 min; 99%).
4.3. Reactions of 1 with ortho-Aminophenol (4oa)
4.1. General Procedure for the reactions of 1 with
Anilines 4ob–oi
Method i. ortho-Aminophenol (4.65 mmol) was added to a
solution of 1 (0.400 g, 1.55 mmol) in DMF (4.5 mL) and the
mixture kept at room temperature until disappearance of 1
(5 days). The mixture was then poured into ice/water and the
precipitate filtered, dried under vacuum and purified by
crystallisation from ethanol to give 63% of 2oa. Physical
data are reported in Tables 10–12.
The appropriate aniline (4.65 mmol) and Et3N
(4.65 mmol) were added to a solution of 1 (0.400 g,
1.55 mmol) in DMF (4.5 mL) and the mixture kept at
1208C until disappearance of 1. The mixture was then
cooled to room temperature, poured into ice/water and

7200
B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
Barbieri, C. L.; Spinelli, D.; Lamartina, L. Chem.-Biochem.
Interact. 1993, 86, 229–254.
Method ii. Operating as above, but at 508C, after 90 min
compound 2oa was obtained in 30% yield.
8. Hrelia, P.; Vigagni, F.; Maffei, F.; Fimognari, C.; Lamartina,
L.; Spinelli, D.; Juric, P.; Guerra, M. C.; Cantelli-Forti, G.
Mutagenesis 1995, 10, 171–177.
Method iii. ortho-Aminophenol (4.65 mmol) and triethyl-
amine (4.65 mmol) were added to a solution of 1 (0.400 g,
1.55 mmol) in DMF (4.5 mL) and the mixture kept at 508C
until disappearance of 1 (1 h). The mixture was then cooled
to room temperature, poured into ice/water and the
precipitate filtered, dried under vacuum and purified by
crystallisation from ethanol to give 2oa (30%).
9. Hrelia, P.; Fimognari, C.; Maffei, F.; Spinelli, D.; Lamartina,
`
L.; Sarva, M. C.; Cantelli-Forti, G. Chem.-Biochem. Interact.
1999, 118, 99–111.
10. Cosimelli, B.; Spinelli, D.; Costanzo, F.; Tonelli, D.;
`
Lamartina, L.; Sarva, M. C.; Seeber, R. Tetrahedron 2001,
57, 1857–1860.
Method iv. Operating as above, but with K2CO3 (4.65 mmol
in 0.5 mL of H2O), after 30 min compound 2oa was
obtained in 41% yield.
11. Guerrera, F.; Salerno, L.; Siracusa, M. A.; Ronsisvalle, G.;
Caruso, A.; Leone, M. G.; Felice, A. Il Farmaco 1990, 45,
29–38.
`
12. Guerrera, F.; Salerno, L.; Sarva, M. C.; Siracusa, M. A.;
Method v. Operating as above, but with K2CO3 (15.55 mmol
in 2.3 mL of H2O), after 15 min compound 2oa was
obtained in 25% yield.
Lamartina, L.; Caruso, A.; Cutuli, V. Il Farmaco 1992, 47,
1149–1160.
13. Press, J. B. Biologically Active Thiophene Derivatives
Revisited: 1983–1988. In The Chemistry of Heterocyclic
Compounds, Thiophene and Its Derivatives, Part 4;
Gronowitz, S., Ed.; Wiley: New York, 1991; pp 397–502.
14. Threadgill, M. D.; Webb, P.; O’Neill, P.; Naylor, M. A.;
Stephens, M. A.; Stratford, I. J.; Cole, S.; Adams, G. E.;
Fielden, E. M. F. J. Med. Chem. 1991, 34, 2112–2120.
4.4. Computational methods
All DFT molecular structures were fully optimised with the
gradient method available in the GAUSSIAN 9837,38
program package and rigorously characterised as minima
according to the number of imaginary mode by applying a
second-order derivative calculation. DFT level geometry
were performed using the non-local hybrid Becke’s three-
parameter exchange functional denoted as B3LYP,37,38
using the DZVP38 basis set which is a local spin density
(LSD)-optimised basis set of double-z quality in the valence
shell plus polarisation functions.
`
15. Lamartina, L.; Spinelli, D.; Guerrera, F.; Sarva, M. C. Magn.
Reson. Chem. 1995, 33, 883–888.
16. Avellone, G.; Ceraulo, L.; Lamartina, L.; Cosimelli, B.;
Spinelli, D.; Zuppiroli, L. Rapid Commun. Mass Spectrom.
1999, 13, 1360–1365, and references cited therein.
17. Reichardt, C. Solvents and Solvent Effects in Organic
Chemistry; 3rd ed. Wiley-VCH: Weinheim, 2003.
18. Consiglio, G.; Noto, R.; Spinelli, D.; Arnone, C. J. Chem. Soc.,
Perkin Trans. 2 1979, 219–221.
Further computational details including Cartesians coordi-
nates and computed total energies of optimised structures
are available by request to Corresponding Author.
19. E.g., see Albert, A.; Serjeant, E. P. The Determination of
Ionization Constants, A Laboratory Manual; 3rd ed. Chapman
& Hall: London, 1994; pp 153–157.
20. Exner, O. Correlation Analysis of Chemical Data; Plenum:
New York, 1988; pp 61–62 and 143–144.
Acknowledgements
21. Terrier, F. Nucleophilic Aromatic Rearrangements: The
Influence of the Nitro Group; VCH: New York, 1991;
pp 226–250, and references cited therein.
Investigation supported by the Universities of Palermo and
Bologna (funds for selected research topics).
22. Stothers, J. B. Carbon-13 NMR Spectroscopy; Academic: New
York, 1972.
23. Ehrenson, S.; Brownlee, R. T. C.; Taft, R. W. Prog. Phys. Org.
Chem. 1973, 10, 1.
References
24. Suezawa, H.; Yuzuri, T.; Hirote, M.; Ito, Y.; Hamada, Y. Bull.
Chem. Soc. Jpn 1990, 63, 328–334.
1. Spinelli, D.; Consiglio, G.; Dell’Erba, C.; Novi, M.
Nucleophilic Substitution in Thiophene Derivatives. In The
Chemistry of Heterocyclic Compounds. Thiophene and Its
Derivatives Part 4; Gronowitz, S., Ed.; Wiley: New York,
1991; pp 295–396.
25. Lamartina, L.; Bonfiglio, G.; Spinelli, D.; De Maria, P. Gazz.
Chim. Ital. 1997, 127, 331–338, and references cited therein.
26. Of course a DSP treatment of data causes an improvement of
the statistical results, but it is ‘small’ (for C-2: rI 6.38 and rR2
3.81 with R 0.955; for C-3: rI 27.70 and r2R 25.47 with R
0.98) and appears statistically not-important. Anyhow by
examining the results obtained the following observations can
be drawn: (i) the two (inductive and resonance) susceptibility
constants show the same sign, (ii) a significant prevalence of
inductive over resonance effects (see the comments on the
SCS of the C-7a probe below in the text) can be evidenced (l
0.60 and 0.71, respectively) in contrast with the results
observed for 2p.
2. Consiglio, G.; Spinelli, D.; Dell’Erba, C.; Novi, M.; Petrillo,
G. Gazz. Chim. Ital. 1997, 127, 753–769.
3. Consiglio, G.; Frenna, V.; Guernelli, S.; Macaluso, G.;
Spinelli, D. J. Chem. Soc., Perkin Trans. 2 2002, 965–970,
See also pp 971–975, and references cited therein.
4. Guerrera, F.; Salerno, L.; Lamartina, L.; Spinelli, D. J. Chem.
Soc., Perkin Trans. 1 1995, 1243–1246.
5. Spinelli, D.; Zanirato, P.; Di Miceli, E.; Lamartina, L.;
Guerrera, F. J. Org. Chem. 1997, 62, 4921–4923.
6. Cosimelli, B.; Lamartina, L.; Spinelli, D. Tetrahedron 2001,
57, 8903–8911.
27. Gawinecki, R.; Rasala, D.; Bak, T. Org. Mass Spectrom. 1992,
27, 39–43.
7. Hrelia, P.; Vigagni, F.; Morotti, M.; Cantelli-Forti, G.;

B. Cosimelli et al. / Tetrahedron 59 (2003) 7189–7201
7201
¨
28. Budzikiewicz, H.; Rullkotter, J.; Schiebel, H. M. Org. Mass
Spectrom. 1972, 6, 251–264.
Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzales, C.;
Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.;
Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian98, revision A.6; Guassian, Inc.:
Pittsburgh, PA, 1998.
29. Ramana, D. V.; Vairamani, M. Org. Mass Spectrom. 1977, 12,
166–168.
30. McLukey, S. A.; Glish, G. L. Org. Mass Spectrom. 1987, 22,
224–228, and references cited therein.
31. Mallen, D. N. B.; Smith, J. M. Org. Mass Spectrom. 1979, 14,
171–176.
32. Gawinecki, R.; Rasala, D. Org. Mass Spectrom. 1991, 26,
100–104.
33. Ceraulo, L.; De Maria, P.; Ferrugia, M.; Foti, S.; Saletti, R.;
Spinelli, D. J. Mass Spectrom. 1995, 30, 257–261, and
references cited therein.
38. Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Can.
J. Chem. 1992, 70, 560–571. UniChem Dgauss, version 2.3.1;
Cray Research, Inc., 1994.
34. Ceraulo, L.; Agozzino, P.; Ferrugia, M.; Rainieri, R. Trends
Heterocicl. Chem. 1995, 4, 115–134, and references cited
therein.
39. Hogg, J. H. C.; Sutherland, H. H. Acta Crystallogr. 1974, B30,
2058–2059.
35. Agozzino, P.; Ceraulo, L.; Ferrugia, M.; Rainieri, R. J. Mass
Spectrom. 1996, 31, 1141–1442, and references cited therein.
36. As pointed out by Terrier (Ref. 21, p 1) a ‘major characteristic
of SNAr substitions is that they occur without rearrangement’.
37. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J.R.; Zakrewski, V. G.;
Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.;
40. See Ref. 21, pp 46–58, and references cited therein.
41. Consiglio, G.; Frenna, V.; Spinelli, D. Base Catalysis in
Aromatic Nucleophilic Substitutions: Current Views 1. Topics
in Heterocyclic Systems—Synthesis Reactions and Properties;
Attanasi, O. A., Spinelli, D., Eds.; Research Signpost:
Trivandrum, India, 1966; Vol. 1, pp 169–186.
42. Boswell, D. E.; Brennan, J. A.; Landis, P. S.; Rodewald, P. G.
J. Heterocycl. Chem. 1968, 5, 69–75.