Secondary steric effects in SNAr of thiophenes: A coordinate kinetic, thermodynamic, UV-VIS, crystallographic and ab initio study

Article abstract of DOI:10.1039/a604516b

The reactivity of some dinitro-benzene and -thiophene derivatives with piperidine and sodium benzenethiolate has been examined giving evidence that benzenes show large and thiophenes show small kinetic secondary steric effects, respectively. The acid diss

Full text of DOI:10.1039/a604516b

View Article Online / Journal Homepage / Table of Contents for this issue
Secondary steric effects in SNAr of thiophenes: a coordinate kinetic,
thermodynamic, UV–VIS, crystallographic and ab initio study
Giovanni Consiglio,a Vincenzo Frenna,b Angelo Mugnoli,c Renato Noto,b
Marcella Panic and Domenico Spinellia
a
Dipartimento di Chimica Organica ‘A. Mangini’, Via S. Donato 15, Bologna, I-40127, Italy
b Dipartimento di Chimica Organica ‘E. Paternò’, Via Archirafi 20, Palermo, I-90123, Italy
c Dipartimento di Chimica e Chimica Industriale, Via Dodecaneso 31, Genova, I-16146, Italy
The reactivity of some dinitro-benzene and -thiophene derivatives with piperidine and sodium
benzenethiolate has been examined giving evidence that benzenes show large and thiophenes show small
kinetic secondary steric effects, respectively. The acid dissociation reaction and the U V–VIS spectra of
some nitrothiophenamines have also been studied. The crystal structure and the absolute conformation of
2-iodo-3,5-dinitrothiophene and of 2-iodo-4-methyl-3,5-dinitrothiophene have been determined. For
different conformations of the analogous chlorodinitro derivatives in the thiophene and benzene series,
ab initio energy calculations have been performed. The results collected show that steric strain among
adjacent groups affects the benzene and thiophene compounds and the kinetic, thermodynamic and
spectroscopic properties of the molecules examined differently. D ifferences in geometry of the two
aromatic rings and then in rotation of nitro groups with respect to the rings themselves as well as
differences along reaction coordinates (essentially depending on hybridation changes) are used to explain
the above data.
Activated aromatic nucleophilic substitution in halogeno-
benzenes can be subject to secondary steric effects.1 In fact a
strong decrease of reactivity is observed when the activating
group (electron-withdrawinggroupssuch asNO2, CN, COR, etc.
in the ortho and/or para position with respect to the halogen)
cannot exert its total potential electronic effect because of steric
effects.
behaviour has been blamed on the peculiar geometry of five-
membered ring derivatives,4a which strongly lowers the steric
interactions between the substituents of the thiophene ring.
On comparing the piperidinodechlorination rates in ethanol
of 2,4-dinitrochlorobenzene 1 and of 3-chloro-2,6-dinitro-
toluene 2 (kH /kM e ca. 800 at 30 ЊC), Capon and Chapman 1b
showed a classical example of the secondary steric effect. This is
related to the torsion and/or to the out-of-plane bending of the
nitro groups determined by the intramolecular hindrance of the
methyl group inserted between the two nitro groups.
In order to confirm this view about the poor weight of
secondary steric effects in SN Ar reactions of thiophene deriv-
atives and to correlate molecular geometry with reactivity
we have carried out a crystal structure analysis of 2-iodo-4-
methyl-3,5-dinitrothiophene 4c (which should show the highest
steric interactions among compounds 4 on account of the size
of iodine) and of 2-iodo-3,5-dinitrothiophene 3c.
To extend the comparison between the behaviour of five-
and six-membered ring derivatives we have also collected the
following data: (a) further kinetic data in methanol on SN Ar
with different nucleophiles (piperidine and sodium ben-
zenethiolate) of some thiophene (3a and 4a, 5 and 6) and ben-
We have recently determined the crystal structure of 22 (that
of 1 was already known)3 and the results obtained have con-
firmed the importance of the steric inhibition of resonance in
connection with the observation of secondary steric effects in
SN Ar reactions.
On the other hand some of us have pointed out that in SN Ar
reactions of thiophene derivatives secondary steric effects are
not important.4 In fact 2-halogeno-3,5-dinitrothiophenes 3 and
+
zene (1 and 2, 7 and 8) derivatives; (b) pKBH measurements of
some thiopheneamines to provide evidence for the effect of
the alkyl group present at C(4) on the through-conjugation
between the amino group at C(2) and the nitro groups at
C(3) and/or at C(5); (c) ab initio calculations of com-
pounds 2, 4a with the aim of leaving out the effect of packing
forces on the molecular conformation; (d) UV–VIS spectra
of some model compounds (nitrosubstituted thiophenamines
9–12).
2-halogeno-4-methyl-3,5-dinitrothiophenes
4 show similar
reactivities with piperidine (kH /kM e ca. 2 at 20 ЊC).4a,c This
J. Chem. Soc., Perkin Trans. 2, 1997
309

View Article Online
extending the study of dinitrothiophenes to an anionic
nucleophile.
2,4-Dinitrochlorobenzene 1 and 3-chloro-2,6-dinitrotoluene
2 give with piperidine and with sodium benzenethiolate in
methanol comparable results concerning the occurrence of
kinetic secondary steric effects: in fact in both instances high
kH /kM e ratios have been measured (850 and 1500 at 20 ЊC,
respectively), confirming the results obtained by Capon and
Chapman on the reactivity with piperidine in ethanol (kH /kM e
960 at 20 ЊC).1b
An interesting confirmation of our previous findings con-
cerning the leaving group effect on the kinetic secondary steric
effect 4c has been obtained by comparing the reactivity of 2,4-
dinitrophenyl phenyl sulfone 7 and of 3-methyl-2,4-dinitro-
phenyl phenyl sulfone 8 with piperidine in methanol: in fact the
kH /kM e ratio obtained is significantly larger (12 800 versus 850 at
20 ЊC) than that observed with chlorine as leaving group, so
duplicating the situation observed in thiophene compounds
[compare previous points (i) and (ii)]. In contrast the same
comparison with the nucleophile sodium benzenethiolate fur-
nishes similar values of the kH /kM e ratios (930 versus 1500 at
20 ЊC) probably because now the early nature of the transition
states of the two reactions studied makes the change of leaving
group (from chlorine to phenylsulfonyl) unable to affect the
reactivity.
Moreover the kinetic data obtained by measuring the reac-
tivities in benzenethiolate substitution reactions of thiophene
derivatives 3a and 4a, 5 and 6 in methanol have confirmed the
absence of kinetic secondary steric effects in SN Ar of thio-
phenes: in fact kH /kM e 3.8 and 5.1 at 20 ЊC have been obtained,
respectively, i.e. values similar to those measured in piperidino
substitution of the same thiophene compounds.4c
Results and discussion
Kinetic data
The influence of the nature of the leaving group and of
the steric interactions in the reaction area on the kinetic sec-
ondary steric effects in thiophene derivatives has been widely
studied.4c–d Moreover the effect of the size of the alkyl group
inserted at C(4) between the two activating nitro groups has
also been investigated.4e
All results pointed to the following: (i) by comparing the
reactivity of 2-halogeno-3,5-dinitrothiophenes
halogeno-4-methyl-3,5-dinitrothiophenes 4 with piperidine and
other amines kH /kM e ratios ranging from 1.1 to 1.9 have been
measured;4c (ii) by comparing the reactivity of 3,5-dinitro-2-
+
pKBH and UV–VIS data
+
The pKBH values measured for some 3-nitro-2-piperidyl-4-R-
or -5-R-thiophenes (9, 10: R = H, Me), 5-nitro-2-piperidyl-4-R-
thiophenes (11: R = H, Me, Et, Pri, But) and 3,5-dinitro-2-
piperidyl-4-R-thiophenes (12: R = H, Me, Pri) have been col-
3 and 2-
+
lected in Table 2. The obtained pKBH values were calculated by
the modified Hammett method 5a (Hammett acidity functions
method: HAFM, i.e. using H0ٞ as acidity function for tertiary
aromatic amines) and by the Marziano–Cox–Yates method 5b
(excess acidity method: EAM). The data obtained by the two
methods are nearly coincident.
phenylsulfonylthiophene
5 and of 3,5-dinitro-4-methyl-2-
phenylsulfonylthiophene 6 with piperidine and other amines
kH /kM e ratios of 5.6–16 have been measured, which shows in
this case the occurrence of small but significant kinetic second-
ary steric effects;4c (iii) kH /kM e ratios measured in the reactions
with various nucleophiles (some amines, sodium methoxide and
benzenethiolate) with both bromine and phenylsulfonyl as leav-
ing groups are practically independent of the nature of the
atom (or group) linked at C(5) (hydrogen, bromine or nitro
group), and this shows that the steric interactions are restricted
to those between the leaving group at C(2) (bromine or phenyl-
sulfonyl), the nitro group at C(3) and the methyl group at C(4)
and that the substituent at C(5) does not affect steric inter-
actions by an enhanced buttressing effect;4c (iv) the insertion at
C(4) between the two activating nitro groups of a bulky alkyl
group (e.g. the Pri group) causes an enhancement of the kinetic
The presence of an alkyl group causes in each case, as
expected on the basis of its electronic effects,6 a decrease in the
acidity of the protonated amines to an extent which depends
essentially on the relative position of nitro and alkyl groups,
+
and ranges from 0.4 to 6.6 units in pKBH values. Apparently,
this is in contrast with the observation that the electronic effects
of alkyl groups are very similar to each other. Indeed, the vari-
+
ations observed in pKBH values indicate either a very large
sensitivity of the dissociation reaction to the electronic effect
(e.g. considering the series of 5-nitro-2-piperidyl-4-R-
thiophenes 11, where R varies from hydrogen, to methyl, ethyl,
isopropyl and tert-butyl, this implies that the Hammett type
relationship would require a susceptibility constant greater than
10)† or the occurrence of effects different from the expected elec-
tronic ones (e.g. steric, solvation or polarization-type effects).
In order to evaluate the electronic effects exerted by a methyl
group, free from steric, solvation and/or polarization-type
i
secondary steric effects: in fact with some amines kH /kPr ratios
increase to ca. 7 and to 60–110 when the leaving groups are
bromine or phenylsulfonyl, respectively.4e However, these
reactivity ratios are much lower than those expected for six-
membered ring derivatives, indicating that in thiophenes the
strain between the four adjacent substituents is markedly lower
than in benzenes and is presumably at least in part relieved on
going from substrates to SN Ar transition states.4e
+
effects, we have compared the pKBH of 3-nitro-2-piper-
idylthiophene 10a and of 5-methyl-3-nitro-2-piperidylthio-
phene 10b: the introduction of the methyl group causes a
+
decrease of only 0.4 pKBH units. The decrease in acidity is
To confirm this point we have measured (see data in Table 1)
the reactivities in methanol of benzene derivatives 1, 2, 7 and 8
with piperidine and sodium benzenethiolate and of thiophene
derivatives 3a, 4a, 5 and 6 with sodium benzenethiolate, so
+
† This is not consistent with the literature data (e.g. for the pKBH of a
series of 5-substituted 3-nitro-2-piperidylthiophenes a ρϪ value of
around 4 has been calculated).5c
310
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Table 1 Rate constants and activation parameters for the reactions of compounds 1–8 with nucleophiles in methanol
k/l molϪ1sϪ1b
(at various temperatures)
∆H‡c/kJ
Ϫ∆S‡d/JKϪ1
kH /kM e
(at 20 ЊC)
Substrate Nucleophilea
molϪ1
molϪ1
1
2
PIP
PIP
2.01 × 10Ϫ3
(0.9)e
4.19 × 10Ϫ3 (10.08)
8.44 × 10Ϫ3
(20.05)
43.4
69.4
129
103
850
9.94 × 10Ϫ6
(20.05)
2.58 × 10Ϫ5 (30.07)
6.59 × 10Ϫ5
(40.10)
3a f
4a f
5f
PIP
PIP
PIP
PIP
PIP
0.866 (0.03)
0.494 (0.02)
2.43 (0.02)
0.257 (0.02)
1.88 (9.99)
3.74 (20.03)
1.93 (20.00)
6.80 (20.03)
0.878 (20.03)
46.0
42.7
31.8
38.9
49.4
76
93
1.9
7.7
1.05 (10.00)
4.14 (10.02)
0.485 (10.00)
2.96 × 10Ϫ3 (9.94)
118
115
118
6f
7
1.36 × 10Ϫ3
(0.01)g
6.46 × 10Ϫ3
(20.01)h
12 800
1500
8
PIP
5.70 × 10Ϫ7
(21.70)
1.14 × 10Ϫ6 (29.99)
2.38 × 10Ϫ6
(39.95)
57.3
169
1
2
NaBT
NaBT
3.51 (0.03)
7.31 (9.95)
15.8 (19.89)i
47.7
61.5
59
72
10.0 × 10Ϫ2
(19.88)
2.32 × 10Ϫ2 (29.99)
5.65 × 10Ϫ2
(39.95)
3ah
4a
5
NaBT
NaBT
NaBT
NaBT
NaBT
NaBT
34 200 (20)
9000 (20)
3.8
5.1
930
1 640 000 (20)j
320 000 (20)
2400 (20)
6
7h
8
0.573 (Ϫ0.02) 1.22 (9.93)
2.57 (19.95)
61.5
74
a PIP = piperidine; NaBT = sodium benzenethiolate. b The rate constants are accurate to within ±3%. c At 20 ЊC; the maximum error is 2.1 kJ molϪ1
.
d At 20 ЊC; the maximum error is 7 JKϪ1 molϪ1 e This datum compares well with a previous kinetic result (at 0 ЊC, k 1.95 × 10Ϫ3) of J. F. Bunnett, E.
.
W. Garbisch, Jr. and K. M. Pruitt, J. Am. Chem. Soc., 1957, 79, 385. f Data from ref. 4c. g This datum compares well with a previous kinetic result (at
0 ЊC, k 1.43 × 10Ϫ3) of AA. of note e. h Datum from ref. 10. i This datum compares well with a previous kinetic result (at 20 ЊC, k 14.4) of AA. of note
j
e. A value of 770.000 was previously reported 10 for kinetic experiments carried out with benzenethiol solutions in the presence of an excess of
sodium methoxide. The observed difference in k values is due to the fact that operating with an excess of methoxide ion some, probably reversible,
interaction occurs that affects absorbance and then the k value measured. Now the reaction substitution has been carried out with benzenethiolate
solutions in the presence of an excess of benzenethiol thus eliminating every secondary reaction.
+
Table 2 pKBH values of nitropiperidylthiophenes 9–12 in aqueous sulfuric acid
HAFM
EAM
ε (λmax/nm)a
+
+
Substrate
ϪpKBH (±spK
)
m(±sm)
ϪpKBH (±spK
)
m*(±sm)
9a = 10a b
9b
1.07 (0.03)
0.28 (0.03)
0.65 (0.01)
1.92
1.27 (0.03)
1.20 (0.03)
1.00 (0.08)
0.36 (0.01)
6.66
1.03 (0.02)
0.98 (0.04)
1.01 (0.01)
0.81
0.78 (0.04)
0.73 (0.04)
0.90 (0.06)
1.08 (0.01)
1.04
1.06 (0.03)
0.29 (0.02)
0.64 (0.01)
1.87
1.27 (0.03)
1.12 (0.03)
1.17 (0.11)
0.39 (0.01)
6.55
1.78 (0.07)
1.70 (0.04)
1.72 (0.04)
1.20
1.25 (0.02)
1.00 (0.03)
1.71 (0.18)
2.05 (0.01)
1.53
5 620(398)
3 700(404)
10b
6 020 (410)
31 600 (448)
34 900 (440)
34 700 (438)
34 700 (444)
32 900 (450)
15 800 (380)
17 800 (408)
16 600 (420)
11ab
11b
11c
11d
11e
12ab
12b
2.91 (0.20)
1.99 (0.09)
0.89 (0.05)
1.01 (0.04)
2.86 (0.15)
1.87 (0.09)
1.30 (0.07)
1.60 (0.08)
12c
a In methanol. b Values from ref. 5c.
higher in all other instances: e.g. in the series of 5-nitro-2-
a polarization-type process.‡ In principle all these effects are
possible, but taking into account the effects exerted by the same
alkyl groups on the piperidinodebromination reactions in
methanol of some 2-bromo-5-nitro-4-R-thiophenes 13 one can
+
piperidyl-4-R-thiophenes 11 it is 0.6–1.5 pKBH units on going
from hydrogen to methyl, ethyl, iso-propyl and tert-butyl.
+
The pKBH variations observed in compounds 11 can be
interpreted by assuming that alkyl groups cause either a torsion
and/or an out-of-plane bending of the nitro group (i.e. an elec-
tronic secondary steric effect) or a hindrance to the solvation of
the nitro group or an electronic effect, whose entity is related to
‡ The relevance of solvation effects in acid dissociation equilibria of
thiophenes 11 is witnessed by the observed meaningful variations in m*
values.
J. Chem. Soc., Perkin Trans. 2, 1997
311

View Article Online
make a choice among them.7 In fact, the rates of the nucleo-
philic substitutions in 13 increase with the size of the alkyl
group which behaves as a substituent activating the SN Ar. Thus,
the main effect of the alkyl group is not to hinder the coplanar-
ity of the activating nitro group with the thiophene ring, which
would cause a strong reduction in the reactivity. Moreover acid
dissociation constants of 11a–e and kinetic rates of piperidino-
debromination of 13a–e7 give a good linear free energy rela-
tionship (s 0.44 ± 0.04, i 0.02 ± 0.03, n 5, r 0.990) showing that
the alkyl groups exert similar effects on the dissociation equi-
libria and on the piperidinodebromination reactions. In a pre-
vious paper it has been proposed that the rate acceleration
determined by alkyl groups in 13 depends on the hindrance of
the orientation of solvent molecules near the nitro group,
which ‘should act mainly in the initial state rather than in the
transition state’,7 or on an electronic effect ‘resulting in the
stabilization of both positive and negative charge through a
polarization-type process’,7 and one can confidently assume
Fig. 1 UV–VIS spectra of compounds 12a–c in methanol
The UV–VIS spectra of piperidyl derivatives 9 and 12 agree
with this interpretation: in fact the introduction of the methyl
group causes in 9 a significant hypochromic effect [ε9a ca. 5600
at λmax(398 nm) and ε9b ca. 3700 at λmax(404 nm) in methanol].
The spectra of 12 are very interesting: the insertion of alkyl
groups causes both a large bathochromic (in methanol λmax
shifts from 380 nm in 12a to 408 and 420 nm, respectively, in
12b and 12c) and a hyperchromic (in methanol εmax increases
from 15 000 in 12a to 17 800 and 16 600 in 12b and in 12c,
respectively) effect. These effects can be easily understood by
taking into account the spectra in methanol of 3-nitro-2-
piperidylthiophene (10a) and of 5-nitro-2-piperidylthiophene
(11a) for which the absorption band can be assigned to
[ N+᎐C(2) → C(3)᎐NO Ϫ] and to [ N+᎐C(2) → C(5)᎐
+
that the same effect determines pKBH values variation.
According to this interpretation no effect of 4-alkyl groups on
the UV–VIS spectra of 5-nitro-2-piperidyl-4-R-thiophenes
11a–e has been observed (see data in Table 2).
The situation appears different for 3-nitro-2-piperidyl-4-R-
thiophenes 9 and 3,5-dinitro-2-piperidyl-4-R-thiophenes 12. In
fact the introduction of the methyl group at C(4) in 9 causes a
᎐
᎐
᎐
᎐
2
NO2Ϫ] electronic transitions, respectively (see Fig. 1). In 12a the
absorption band seems to be affected more by the [ N + ᎐C(2)
᎐
᎐
→ C(3)᎐NO Ϫ] electronic transition than by the [ N+ ᎐C(2)
᎐
2
→ C(5) = NO2Ϫ] electronic transition, probably because
of the through-conjugation which favours electronic inter-
actions between substituents at C(2) and C(3) (hyper-ortho
effect),8 as indicated by the wavelength and the molar extinction
coefficient observed for the maximum. The insertion of an alkyl
group between the two nitro groups as in 12b and 12c causes a
significant variation in the spectra, which can be easily
accounted for by assuming that the alkyl groups hinder the
coplanarity of the nitrogroup at C(3): thus making more rele-
vant the electronic transition [ N+᎐C(2) → C(5)᎐NO Ϫ]with
+
decrease in pKBH values of 0.8 units, and the insertion at C(4)
between the two nitro groups of a methyl or of an isopropyl
group causes a decrease of 3.7 and 4.7 units, respectively, in 12.
+
Note that 12c shows a pKBH value (Ϫ1.99) similar to that of
2-piperidyl-5-nitrothiophene (Ϫ1.92).
In principle in 9 and in 12 the introduction of alkyl groups
can also affect the coplanarity, the solvation of the nitro
group(s) and/or a polarization-type process as in 11, but both
kinetic and spectroscopic data seem to indicate the occurrence
of a small secondary steric effect. In fact considering the effect
of alkyl groups on the rates of piperidinodebromination in
methanol of 2-bromo-3-nitro-4-R-thiophenes (14: R = H,
Me)4c and of 2-bromo-3,5-dinitro-4-R-thiophenes (3b, 4b, 15)4e
one can observe only a small decrease in the reactivity on intro-
duction of the alkyl groups (in 14: kH /kM e ca. 2; in 3b and 15:
᎐
᎐
2
respect to [ N+᎐C(2) → C(3)=NO Ϫ]. The proposed torsion
᎐
2
and/or out-of-plane bending of the nitro group at C(3) deter-
mined by the insertion of an alkyl group between the two nitro
groups at C(3) and at C(5) is strongly supported by the com-
parison of the crystal structures of compounds 3c and 4c and
by the comparison of ‘ab initio’ results for compounds 2 and
4a.
i
kH /kPr ca. 7).
Crystal structure analysis of 3c and 4c and ab initio calculations
for 2 and 4a
To gain further information on the structure of thiophene
derivatives and to make a comparison with the previously stud-
ied benzene derivatives 13 and 22 we considered it useful to
determine the crystal structure of compounds 3c and 4c.
An ORTEP view9 of 3c and 4c with the atom numbering
scheme is shown in Figs. 2 and 3. Hydrogen atoms have been
named according to the numbering of the bonded carbon atom.
Values of bond lengths and bond angles are in the normal
range and can be compared with those obtained in previous
crystal structure determinations of dinitrothiophene derivatives
(e.g. 2-chloro-3,5-dinitrothiophene and 3,5-dinitro-2-phenyl-
sulfonylthiophene;10 2,5-dimethyl-3,4-dinitrothiophene).11
The thiophene ring is planar within experimental errors in
both compounds; Table 3 summarizes the deviations from
planarity and the dihedral angles between least-squares planes.
312
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Table 3 Least-squares planes
assume smaller values (129.8Њ and 121.4Њ, respectively) the
strain is relieved by a rotation of the 3-nitro group, the torsion
angle C(2)᎐C(3)᎐N(1)–O(1) being Ϫ44.7(6)Њ.
Highest deviation/Å
Whereas in both structures the 5-nitro group is substantially
coplanar with the thiophene ring, the behaviour of the 3-nitro
group is different (see Table 3), as can be foreseen on the basis
of steric strain from adjacent substituents; this feature can have
influence when considering the SN Ar reactivity, owing to the
fact that the activating effect of the nitro group reaches its max-
imum in the case of coplanarity with the ring. On the same
basis one can predict, in full agreement with experiment, that
the reactivity should be even lower for the corresponding
halogeno-benzene derivatives, where the higher hindrance
between vicinal groups should lead to a larger rotation for the
ortho-nitro group. This must be true e.g. in 2-nitrohalogeno-
benzenes and, to a larger extent, in 2-nitro-3-halogenotoluenes.
In fact in the crystals of 2,4-dinitrochlorobenzene (alpha
form) the 2-nitro group is rotated by ca. 40Њ with respect to the
benzene ring3 and in 3-chloro-2,6-dinitrotoluene the 2-nitro
group is rotated by nearly 80Њ.2
Compound 3c Compound 4c
(A) Thiophene ring
(B) 3-Nitro group
(C) 5-Nitro group
0.011(7)
0.012(7)
0.013(7)
0.006(4)
0.002(4)
0.003(4)
Dihedral angles (Њ)
Compound 3c Compound 4c
A–B
A–C
0.8(3)
3.5(3)
44.2(2)
2.0(2)
These values however could be affected by packing forces in
the crystal phase. Indeed, a recent study13 of crystal structures
retrieved from the Cambridge Structural Database14 has shown
that, besides the steric hindrance of adjacent substituents, pack-
ing forces can play a noticeable role in modifying the rotation
angle of a nitro group bonded to an aromatic six-membered
carbon ring. For nitro groups having 0, 1 and 2 adjacent non-H
substituents the following ranges of rotation angles have been
found: 0–28Њ; 5–49Њ for methyl (and methylene), 14–43Њ for
chlorine; and 56–80Њ for carbon and halogen substituents,
respectively. According to De Ridder and Schenk,13 this
behaviour ‘. . . makes it almost impossible to predict the rota-
tion angle of a nitro group for one or two given adjacent sub-
stituents’. In the present case therefore an independent study of
the molecular conformation appeared desirable; we deemed it
worthwhile to evaluate the change in energy for the isolated
molecule of 4c (or of the corresponding 2-chloro-derivative, 4a)
and for the isolated molecule of 3-chloro-2,6-dinitrotoluene, 2,
when rotating the 3- and the 2-nitro groups, respectively. We
adopted the ab initio methods using the GAUSSIAN92 pro-
gram 15 at the HF level and the 6-31G* basis set, with d-type
polarization functions on all non-hydrogen atoms. In order to
use this basis set we had to consider the chlorine- (4a), instead
of the iodine-derivative (4c). Single point energy calculations
were performed for different fixed geometries, each of them
corresponding to a different torsion angle of the considered
nitro group.
Fig. 2 ORTEP view of the structure of 3c in the crystal. Thermal
ellipsoids are drawn at the 50% probability level; the hydrogen atom,
treated as isotropic, is on an arbitrary scale.
We followed the same protocol previously described.10 The
initial geometry was obtained from the experimental results,
with the ring idealized to an exactly planar conformation. The
C(2)–I bond distance was substituted, in 4a, by the correspond-
ing experimental10 C(2)–C1 value of 1.683 Å. For the 5-nitro
group in 4a the same conformation was assumed as found in 4c,
and for the 6-nitro group in 2 the experimental orientation 2 was
adopted; these were held fixed in subsequent calculations. Start-
ing with the 3-nitro group (in 4a) and the 2-nitro group (in 2)
coplanar to the ring, their torsion angle was decreased until an
increase in energy was found. Values of the torsion angles and
the resulting differences of energy with respect to the planar
conformation are summarized in Table 4.
As a result, the conformation of minimum energy for com-
pound 4a is fairly similar to that found experimentally for 4c;
the energy differences with respect to the planar molecule lie,
however, within 15 kJ molϪ1, possibly not precluding the nitro
group to reach coplanarity with the thiophene ring when the
reactants are approaching the transition state.10 This is in
agreement with the large SN Ar reactivity of 4c and 4a. On the
other hand for compound 2 the energy well reaches 91 kJ molϪ1
for a torsion of the 2-nitro group as high as Ϫ80Њ in the min-
imum energy conformation (again in agreement with the
Fig. 3 ORTEP view of the structure of 4c in the crystal. Thermal
ellipsoids are drawn at the 50% probability level; hydrogen atoms,
treated as isotropic, are on an arbitrary scale.
In both structures one short Iؒ ؒ ؒO intermolecular contact
has been found, namely: in 3c, I with O(2) in 2.5 Ϫ x, 1 Ϫ y,
Ϫ0.5 + z (3.04 Å); in 4c, I with O(1) in Ϫ0.5 + x, 1.5 Ϫ y,
2 Ϫ z (3.08 Å). This is however not very unusual and even
shorter Iؒ ؒ ؒO intermolecular distances can be seen [e.g. 2.92 Å
in 3-iodo-2-(methylsulfinyl)thiophene].12 All other non-bonded
distances are normal. Atoms I and O(1) are involved even in
rather short 1ؒ ؒ ؒ5 intramolecular contacts (3.143 and 3.243 Å
in compound 3c and 4c, respectively). In 3c the strain is relieved
by the large I᎐C(2)᎐C(3) (132.3Њ) and C(2)᎐C(3)᎐N(1) (124.9Њ)
bond angles; in 4c, where the corresponding bond angles
J. Chem. Soc., Perkin Trans. 2, 1997
313

View Article Online
Table 4 Energy differences (kJ molϪ1) with respect to the planar con-
formation as a function of the nitro group torsion angle
Energy difference/kJ molϪ1
Torsion angle (Њ) Ϫ20
Compound 4a Ϫ9.3 Ϫ13.5 Ϫ14.9 Ϫ13.2
Ϫ30
Ϫ40
Ϫ50
Ϫ60
Ϫ9.5
Torsion angle (Њ) Ϫ20
Ϫ40
Ϫ60
Ϫ70
Ϫ80
Ϫ85
Compound 2 Ϫ33.6 Ϫ73.2 Ϫ89.0 Ϫ90.7 Ϫ90.8 Ϫ90.7
experimental result);2 this barrier to planarity prevents the 2-
nitro group from exerting its activating effect, as confirmed
from reactivity measurements.
Conclusion
The results allow the following conclusions to be drawn. In
thiophenes4 the SN Ar reactivity with neutral and anionic
reagents is scarcely affected by secondary steric effects at
variance with what happens in benzenes.1 Considering the
results of the crystal structure analysis for thiophenes 3c and
4c and of the molecular conformation for benzene 2 and
thiophene 4a it appears evident that the differences in the
geometry of the two aromatic rings, benzene and thiophene,
play a relevant role in the occurrence of kinetic secondary
steric effects. In thiophenes the steric strain between the four
adjacent substituents is relieved by the rotation of the 3-nitro
group (in 4c: ca 45Њ) and the energy difference with respect to
lization from methanol (mp 179–180 ЊC; exact mass measure-
ment: found 322.0253, C13H10N2O6S requires 322.0260). δH -
(CD3)2SO 8.43 [AB-system, 2 H, H(5) and H(6)], 8.02–7.96 [m,
2 H, H(2Ј) and H(6Ј)], 7.84–7.79 [m, 1 H, H(4Ј)], 7.74–7.69 [m,
2 H, H(3Ј) and H(5Ј)], 2.31 (s, 3 H, CH3). (J in Hz).
Compounds 19–21 were prepared and purified according to
the general method reported.20 Compound 19 was crystallized
from methanol, mp 123 ЊC. UV–VIS spectrum in methanol: log
ε (λmax) 3.90 (316 nm). Exact mass measurements: found
290.0367, C13H10N2O4S requires 290.0361. δH (CD3)2SO 8.11 [d,
1 H, H(5), J 8.8], 7.70–7.45 [m, 5 H, H(2Ј)/H(6Ј)], 7.13 [d, 1 H,
H(6), J 8.8], 2.42 (s, 3 H, CH3). Compound 20 was crystallized
from ethanol–dioxane, mp 158 ЊC. UV–VIS spectrum in metha-
nol: log ε (λmax) 4.22 (360 nm). Exact mass measurements:
found 281.9772, C10H6N2O4S2 requires 281.9769. δH [(CD3)2SO]
8.56 [s, 1 H, H(4)], 7.85–7.64 [m, 5 H, H(2Ј)/H(6Ј)]. Compound
21 was crystallized from methanol–dioxane, mp 185 ЊC. UV–
VIS spectrum in methanol: log ε (λmax) 4.16 (362 nm). Exact
mass measurements: found 295.9920, C11H8N2O4S2 requires
295.9926. δH [(CD3)2SO] 7.86–7.63 [m, 5 H, H(2Ј)/H(6Ј)], 2.81 (s,
CH3).
the planar structure is relatively low (in 4a: ca. 15 kJ molϪ1
)
and seems to be significantly relieved, as a consequence of
the hybridation change at C(2) (from sp 2 to sp 3), on going
from substrates to SN Ar transition states thus determining the
absence4 of kinetic effects or the occurrence4 of small effects.
In contrast, in benzene 2 the same steric strain causes both a
larger rotation of the 2-nitro group (ca. 80Њ) and a higher
energy difference with respect to the planar structure (ca. 91
kJ molϪ1) which cannot be significantly relieved on going
from the substrate to the SN Ar transition state, thus deter-
mining the occurrence of a large kinetic secondary steric
effect.1
+
Concerning pKBH values and UV–VIS data of thiophenes
the situation can be different in derivatives containing four
adjacent substituents. For example, on going from 12a to 12b
and 12c the rotation of the 3-nitro group due to steric strain has
a large effect both on the acid dissociation reaction and on the
UV–VIS spectra. In fact for the acid dissociation reaction,
where no change in hybridation at C(2) occurs, large variations
All new products gave correct analyses (C, H, N, S).
Kinetic measurements
The kinetics for piperidine- and benzenethiolate-substitution
were followed spectrometrically as previously described.16,20
The concentrations used were 10Ϫ5–10Ϫ3  for substrates, 0.2–1
 for piperidine and 10Ϫ4–10Ϫ2  for sodium benzenethiolate
(in presence of 1–3 times excess of benzenethiol). Excess of
benzenethiol had no discernible effect on SN Ar reaction rate.21
A DU-6 Beckman instrument equipped with a thermostatted
cell compartment was used except in the case of benzenethi-
olate substitution of 3a, 4a, 5–7 for which, owing to the high
+
of the relevant pKBH are observed because of the insertion
+
between the two nitro groups of a methyl (∆pKBH ca. 3.7) or of
+
a bulkier isopropyl group (∆pKBH ca. 4.7). The same insertion
causes large bathochromic and hyperchromic effects on UV–
VIS spectra of compounds 12.
Thus, the kinetic, thermodynamic and UV–VIS data of the
thiophene studied can be easily interpreted in the light of crys-
tal structure analysis and ab initio calculations and the reasons
why an alkyl group adjacent to nitro group(s) can exert a small
or a large effect on some chemical and physical properties of
thiophenes can be well understood.
reaction rates,
a
Durrum D-110 stopped-flow spectro-
photometer was used.
+
pKBH measurements
+
Acid dissociation constants KBH were determined as previ-
ously described.5c Data treatments (according to HAFM and
EAM methods) and corrections were made as previously
described.5c
Experimental
Materials
Compounds 1,1b 2,1b 3a–c,16 4a,17 4b,4a 4c,17 5,16 6,16 7,18 9a,16
9b,4b 10b,7 11a,16 11b–e,7 12a,16 12b,4a 12c,4e 13a,16 13b–e,7
14a,16 14b,4b 15,4e 16,1b 17,1b 18,19 methanol,20 piperidine16 and
benzenethiol20 were prepared and/or purified according to the
methods reported.
Crystallographic measurements
After taking preliminary Laue photographs, crystals of 3c and
4c were mounted on a CAD4 diffractometer; data were col-
lected using graphite-monochromated Mo-Kα radiation. Cell
parameters were obtained by least-squares refinement on dif-
fractometer angles for 25 automatically centred reflections,
λ = 0.7107 Å.
Compound 8 was prepared by refluxing 7 with sodium
benzenesulfinate in ethanol for several hours. The precipitate
obtained after cooling was filtered off and purified by crystal-
314
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Crystal data
and the National Research Council, CNR, for financial
support.
Compound 3c, C4HIN2O4S, M = 300.03. Orthorhombic,
a = 6.395(6), b = 10.009(4), c = 12.565(4) Å, V = 804.3 Å3, space
group P212121, Z = 4, Dc = 2.478 g cmϪ3. Crystal dimensions
References
0.30 × 0.15 × 0.03 mm, µ = 42.2 cmϪ1
.
1 (a) H. Lindemann and A. Pabst, Annalen, 1928, 462, 24; (b) B. Capon
and N. B. Chapman, J. Chem. Soc., 1957, 600 and references therein.
2 A. Mugnoli, D. Spinelli, G. Consiglio and R. Noto, Cryst. Struct.
Commun., 1981, 10, 983.
3 K. J. Watson, Nature, 1960, 188, 1102; E. M. Gopalakrishna, Acta
Crystallogr., Sect. A, 1969, 25, S150; H. Takazawa, S. Ohba and
Y. Saito, Acta Crystallogr., Sect. B, 1989, 45, 432; S. Sharma,
N. Weiden and A. Weiss, J. Chem. Phys., 1989, 90, 483; A. Wilkins,
R. W. H. Small and J. T. Gleghorn, Acta Crystallogr., Sect. B, 1990,
46, 823. In the last paper a less stable beta form, with a rotation
angle as high as 69Њ, is also described.
4 (a) D. Spinelli, C. Dell’Erba and G. Guanti, J. Heterocycl. Chem.,
1968, 5, 323; (b) D. Spinelli, G. Consiglio and A. Corrao, J. Chem.
Soc., Perkin Trans. 2, 1972, 1866; (c) D. Spinelli, G. Consiglio,
R. Noto and A. Corrao, J. Chem. Soc., Perkin Trans. 2, 1974, 1632;
(d) D. Spinelli and G. Consiglio, J. Chem. Soc., Perkin Trans. 2,
1975, 1388; (e) C. Arnone, G. Consiglio, S. Gronowitz, B. Maltes-
son, A.-B. Hörnfeldt, R. Noto and D. Spinelli, Chemica Scripta,
1978–1979, 13, 130; (f) G. Consiglio, C. Arnone, D. Spinelli,
R. Noto, V. Frenna and S. Fisichella, J. Chem. Soc., Perkin Trans. 2,
1985, 519; (g) G. Consiglio, C. Arnone, D. Spinelli, R. Noto,
V. Frenna, S. Fisichella and F. A. Bottino, J. Chem. Soc., Perkin
Trans. 2, 1985, 523.
Compound 4c, C5H3IN2O4S, M = 314.05. Orthorhombic,
a = 7.173(2), b = 9.276(3), c = 13.550(6) Å, V = 901.6 Å 3, space
group P212121, Z = 4, Dc = 2.314 g cmϪ3. Spherical crystal, diam-
eter 0.25 mm, µ = 37.7 cmϪ1
.
Data collection and processing
Compound 3c, ω/2θ scan mode, scan width 1.5Њ, scan speed
0.7–6.7Њ minϪ1
; 1090 independent reflections measured
(3.0 р θ р 27.5Њ), absorption correction 22 (max., min. transmis-
sion factors = 0.98, 0.52).
Compound 4c, ω/θ scan mode, scan width 1.2Њ, scan speed
0.9–20.0Њ minϪ1
; 1209 independent reflections measured
(2.5 р θ р 27.5Њ), spherical absorption correction (max., min.
transmission factors = 0.51, 0.50).
During the data collection the centring of three reflections
was repeated periodically for both structures, to test the crystal
orientation, and one reflection was monitored every 60 min to
check the crystal stability; no crystal decay was observed.
Structure determination and refinement
5 (a) C. H. Rochester, Acidity Functions, Academic Press, London,
1970; E. M. Arnett and G. W. Mach, J. Am. Chem. Soc., 1964, 86,
2671; K. Yates, H. Wai, G. Welch and R. A. McClelland, J. Am.
Chem. Soc., 1973, 95, 5960; (b) N. C. Marziano, G. M. Cimino and
R. Passerini, J. Chem. Soc., Perkin Trans. 2, 1975, 341; R. Cox and
K. Yates, J. Am. Chem. Soc., 1978, 100, 3861; (c) P. De Maria,
R. Noto, G. Consiglio and D. Spinelli, J. Chem. Soc., Perkin Trans.
2, 1989, 791.
Both structures were solved by the Patterson method. The
refinement was accomplished on F 2 data by means of isotropic,
and then anisotropic full-matrix least-squares process with the
program SHELXL 93.23 For compound 3c the atom H(4) was
located in the calculated position at a distance of 0.93 Å from
C(4), and allowed to ride on the bonded carbon atom. For
compound 4c the methyl hydrogen atoms were located by
means of a circular difference synthesis calculated for the
methyl group at a fixed C᎐H distance (0.96 Å) and a fixed
C᎐C᎐H angle (109.47Њ); both the rotation angle and a common
U(H) thermal factor were refined.
6 σm Values for alkyl groups range between Ϫ0.06 and Ϫ0.09 (e.g. see
O. Exner, Correlation Analysis of Chemical Data, p. 61, Plenum
Press, New York, 1988).
7 G. Consiglio, D. Spinelli, S. Gronowitz, A.-B. Hörnfeldt, B. Maltes-
son and R. Noto, J. Chem. Soc., Perkin Trans. 2, 1982, 625.
8 S. Gronowitz, Arkiv Kemi, 1958, 13, 295; D. Spinelli, G. Guanti and
C. Dell’Erba, J. Chem. Soc., Perkin Trans. 2, 1972, 441; S. Grono-
witz, I. Johnson and A.-B. Hörnfeldt, Chemica Scripta, 1975, 7,
76, 211; D. Spinelli, R. Noto and G. Consiglio, J. Chem. Soc., Perkin
Trans. 2, 1976, 747; D. Spinelli, R. Noto, G. Consiglio and A.
Storace, J. Chem. Soc., Perkin Trans. 2, 1976, 1805; G. Consiglio,
S. Gronowitz, A.-B. Hörnfeldt, R. Noto, K. Petterson and
D. Spinelli, Chemica Scripta, 1978–1979, 13, 20; G. Consiglio,
S. Gronowitz, A.-B. Hörnfeldt, R. Noto and D. Spinelli, Chemica
Scripta, 1980, 15, 49; G. Consiglio, D. Spinelli, S. Fisichella,
G. Alberghina and R. Noto, J. Chem. Soc., Perkin Trans. 2, 1980,
1153; R. Noto, S. Buscemi, G. Consiglio and D. Spinelli, J. Hetero-
cycl. Chem., 1981, 18, 735; G. Consiglio, D. Spinelli, S. Gronowitz,
A.-B. Hörnfeldt and R. Noto, Chemica Scripta, 1982, 19, 46.
9 C. K. Johnson, ORTEP, Report ORNL-3794, Oak Ridge National
Laboratory, Tennessee, 1965.
10 A. Mugnoli, D. Spinelli, G. Consiglio and R. Noto, J. Heterocycl.
Chem., 1988, 25, 177.
11 A. Mugnoli, M. Novi, G. Guanti and C. Dell’Erba, Cryst. Struct.
Commun., 1978, 7, 703.
12 U. Folli, D. Iarossi, A. Mucci, A. Musatti, M. Nardelli, L. Schenetti
and F. Taddei, J. Mol. Struct., 1991, 246, 99.
13 D. J. A. De Ridder and H. Schenk, Acta Crystallogr., Sect. B, 1995,
51, 221.
For both structures the refinement converged well, with a
final maximum shift-to-esd ratio of 0.001. The weighting
2
scheme was w = 1/[σ2(Fo ) + (aP)2 + bP], with a,b = 0.054,
0.00 and 0.033, 0.14 for 3c and 4c, respectively, with
2
2
P = (Fo2 + 2Fc )/3, and σ(Fo ) from the counting statistics. For
compound 4c, an extinction parameter 23 was refined to a final
value of X = 0.027(1). Final reliability factors were
2
R1 = 0.0297 on 910 Fo > 4σ(Fo), wR2 = 0.0806 on all 1090 Fo
data (110 parameters) for 3c, and R1 = 0.0201 on 1036
2
Fo > 4σ(Fo), wR2 = 0.0559 on all 1209 Fo data (121 param-
eters) for 4c. The values of the goodness of fit were 1.041 and
1.065 for the two structures. The final difference Fourier map
showed some residual peaks near to the position of the iodine
atom, the two highest values being 1.15 and 0.78 e ÅϪ3 for 3c,
and 0.54 and 0.38 e ÅϪ3 for 4c. The Flack parameter 24 was
0.00(5) and Ϫ0.09(3) for 3c and 4c, respectively. Thus the abso-
lute structure and, in the present cases, the absolute conform-
ation 25 have been established for both compounds.
The scattering factors were taken from SHELXL 93. All
geometry calculations were performed with the program
PARST 93.26 Ranges of esds on bond lengths and bond angles
are 0.007–0.011 and 0.4–0.8 for 3c, and 0.004–0.007 Å and 0.2–
0.5Њ for 4c, respectively. Atomic coordinates, bond lengths and
angles, and thermal parameters have been deposited at the
Cambridge Crystallographic Data Centre (CCDC). For details
of the deposition scheme, see ‘Instructions for Authors’,
J. Chem. Soc., Perkin Trans. 2, 1997, Issue 1. Any request to the
CCDC for this material should quote the full literature citation
and the reference number 188/38. Tables of anisotropic thermal
parameters and of the observed and calculated structure fac-
tors are available from an author (A.M.) on request.
14 F. H. Allen, O. Kennard and R. Taylor, Acc. Chem. Res., 1983, 16,
146.
15 M. J. Frisch, G. W. Trucks, M. Head-Gordon, P. M. W. Gill,
M. W. Wong, J. B. Foresman, B. G. Johnson, H. B. Schlegel,
M. A. Robb, E. S. Replogle, R. Gomperts, J. L. Andres, K. Rag-
havachari, J. S. Binkley, C. Gonzalez, R. L. Martin, D. J. Fox, D. J.
Defrees, J. Baker, J. J. P. Stewart and J. A. Pople, GAUSSIAN92,
Revision C.4. Gaussian, Inc., Pittsburgh, PA, 1992.
16 D. Spinelli, C. Dell’Erba and A. Salvemini, Ann. Chim. (Rome),
1962, 52, 1156 and references therein.
17 D. Spinelli, G. Consiglio, R. Noto and A. Corrao, J. Chem. Soc.,
Perkin Trans. 2, 1974, 1632.
18 R. W. Bost, J. O. Turner and R. Norton, J. Am. Chem. Soc., 1932, 54,
1985
19 D. Spinelli, C. Dell’Erba and G. Guanti, Ann. Chim. (Rome), 1965,
55, 1252.
Acknowledgements
We wish to thank the Italian Ministry, MURST (40% funds)
J. Chem. Soc., Perkin Trans. 2, 1997
315

View Article Online
20 L. Chierici, C. Dell’Erba and D. Spinelli, Ann. Chim. (Rome), 1965,
55, 1069.
21 J. F. Bunnett and N. Sbarbati Nudelmann, J. Org. Chem., 1969, 34,
2038.
24 H. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876.
25 P. J. Jones, Acta Crystallogr., Sect. A, 1986, 42, 57.
26 M. Nardelli, Comput. Chem., 1983, 7, 95.
22 A. C. T. North, D. C. Phillips and F. S. Mathews, Acta Crystallogr.,
Sect. A, 1968, 24, 351.
23 G. M. Sheldrick, SHELXL93, Program for Refinement of Crystal
Structures, University of Göttingen, 1993.
Paper 6/04516B
Received 28th June 1996
Accepted 20th August 1996
316
J. Chem. Soc., Perkin Trans. 2, 1997