Z, E-Isomerization mechanism for N-arylthio-1,4-benzoquinonimines: DNMR and DFT investigations

Article abstract of DOI:10.1002/mrc.2254

Z, E-Isomerization has been investigated for the series of the N-arylthio-1,4-benzoquinonimines using a line shape analysis in the ¹H NMR spectra. Thermodynamic parameters and substituent effects have been analyzed for the isomerization process. It has been shown based on the DFT (B3LYP) calculations that the dynamic transformation for N-arylthio-1,4- benzoquinonimines should be considered as a combination of the two different processes, a rotation about the N-S bond and an inversion at nitrogen via the transition state with the linear C = N-S moiety. The free energies of activation for the isomerization (ΔG_{298 K}) measured experimentally depend on the substitution in the quinonimine moiety and phenyl ring and can be referred either to the inversion of the nitrogen atom or to the hindered rotation about the N-S bond. Copyright

Full text of DOI:10.1002/mrc.2254

Research Article
Received: 11 February 2008
Revised: 15 April 2008
Accepted: 21 April 2008
Published online in Wiley Interscience: 17 July 2008
(www.interscience.com) DOI 10.1002/mrc.2254
Z, E-Isomerization mechanism for
N-arylthio-1,4-benzoquinonimines: DNMR
and DFT investigations
V. V. Pirozhenko,a A. B. Rozhenko,a,b∗ A. P. Avdeenko,c S. A. Konovalovac
and A. A. Santalovac
Z, E-Isomerization has been investigated for the series of the N-arylthio-1,4-benzoquinonimines using a line shape analysis in
the 1H NMR spectra. Thermodynamic parameters and substituent effects have been analyzed for the isomerization process. It
has been shown based on the DFT (B3LYP) calculations that the dynamic transformation for N-arylthio-1,4-benzoquinonimines
should be considered as a combination of the two different processes, a rotation about the N–S bond and an inversion at
nitrogen via the transition state with the linear C N–S moiety. The free energies of activation for the isomerization (ꢀG
measured experimentally depend on the substitution in the quinonimine moiety and phenyl ring and can be referred either to
)
298 K
c
the inversion of the nitrogen atom or to the hindered rotation about the N–S bond. Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: isomerization; inversion; benzoquinonimines; variable-temperature 1H NMR; line shape analysis; DFT calculations; B3LYP
R
Introduction
1
N
A Z,E-isomerization is inherent to benzoquinonimines (I) as
representatives of the class of imino compounds.[1]
R
Y
2
C
R
R
1
Y
R
1
N
N
Y
2
N
O
O
R
N
2
A
B
Y
R
R
1
2
X
X
Y
N
I
D
Two mechanisms are usually considered for the intramolecular
isomerization of imines of general formula R1R2C N–Y: rotation
of the N-substituent about the C N bond with the nonplanar
transition state (transition state C in Scheme 1) and inversion at
the nitrogen atom over the planar transition state with the linear
Scheme 1. Possible isomerization mechanisms for imines R1R2 C N–Y.
moiety,[3–5] due to the easy analysis of the 1H NMR spectra where
the tert-butyl substituent resonances appeared as singlets. For
such cases, the activation ꢀGC energy values could be estimated
from the coalescence temperature TC using the known approxi-
mated expressions.[6] As a consequence, the only effect of the sub-
stituentsattheimino-nitrogenatom(oratthephenylringattached
C
N–Y moiety (transition state D), or a combination of both. If
states A and B are identical (R1 = R2), a degenerated isomeriza-
tion (or topomerization) is considered. The choice between the
two mentioned isomerization mechanisms is usually somewhat
arbitrary, and the interpretation is based on the influence of the
properties of the substituent at the C N bond on the thermody-
namic activation parameters, as well as the solvent effects.[1,2]
For benzoquinonimines (I) the measured activation barrier val-
ues, which determine the probability of such conformational
transformations, strongly depend on the type of the substituents
at the nitrogen atom and vary within 40–100 kJ/mol.[2] Therefore,
such processes can be investigated using the NMR spectroscopy.
Most of the published papers devoted to the Z,E-isomerization
in benzoquinonimines dealt only with the derivatives substituted
with the tert-butyl groups at positions 2 and 6 of the quinonimine
∗
Correspondence to: A. B. Rozhenko, Institute of Organic Chemistry, National
Academy of Sciences of Ukraine, 5 Murmanskaya Str. Kiev, 02660, Ukraine.
E-mail: a rozhenko@ukr.net
a
b
c
Institute of Organic Chemistry, National Academy of Sciences of Ukraine, 5
Murmanskaya Str. Kiev, 02660, Ukraine
¨
¨
¨
Fakultat fu¨r Chemieder Universitat Bielefeld, Universitatstr. 25, 33615 Bielefeld,
Germany
Donbass State Engineering Academy, 72 Shkadinova Str., Kramatorsk, 84313,
Ukraine
c
Magn. Reson. Chem. 2008, 46, 811–817
Copyright ꢀ 2008 John Wiley & Sons, Ltd.

V. V. Pirozhenko et al.
to this atom) on the ꢀGC activation energy was studied previously,
to the best of our knowledge. On the basis of these investigations
it was established that for 2,6-di-tert-butyl-substituted N-aryl-,[2a]
N-aroyl-[4] and N-arylsulfonyl-1,4-benzoquinonimines,[2b,c,5] elec-
tron acceptor substituents at the para-position of the aromatic
ring reduced the activation energy for the Z, E-isomerization pro-
cess. At the same time, for the structures involving a bivalent
sulfur atom such substituents increased the activation barrier.[7]
Moreover, a similar trend was also observed for the imino systems
with a thioaryl moiety bound to the nitrogen atom, in particular
for Tol2C N–S–Ar,[8] but no explanation was given for such pe-
culiarity of the thioimino compounds. One of the possible versions
could be the change of the isomerization mechanism from the
inversion to rotation or to the more complicated process.[9]
the isomerization processes in benzoquinonimines were obtained
using the NMR spectrometers operating at frequencies of 100 MHz
or lower. This provided small differences in the chemical shifts
for the groups taking part on the exchange processes which
adversely affected the accuracy of the determination of the
thermodynamic parameters.[10]
We have investigated the Z, E-isomerization process in the wide
series of N-arylthio-1,4-benzoquinonimines (1–10) substituted
both in the phenyl ring and in the quinonimine moiety
using the 1H NMR line-shape analysis. To the best of our
knowledge, the influence of the substitution at the various
positions of the quinonimine part of the molecule on the Z, E-
isomerization processes in benzoquinonimines has previously not
beenestablished. TheexperimentalresultsarecollectedinTable 1.
As it has been already mentioned above, the most available
data in the literature on the Z, E-isomerization were obtained
using the approximated expressions,[6] which determined the
activation ꢀGC energy values based only on the coalescence
temperature TC of the singlet peaks. As a consequence, such
a comparison was not always reliable because the coalescence
for the different structures was achieved by the different
temperatures. This effected the −TꢀS contribution into the ꢀGC
expression. Hence, a satisfactory comparison was sometimes not
possible even for structurally very similar species. Additionally it
should be noted that most published thermodynamic data on
R1
R3
2
3
1
4
O
N
2'
3'
6
5
4'
1'
S
R
R2
R4
Table 1. Thermodynamic activation parameters for the Z, E-isomerization process for N-arylthio-1,4-benzoquinonimines
Structure
R1
R2
R3
R4
R
Solvent
ꢀG298K (kJ/mol)
ꢀH (kJ/mol)
ꢀS (J/mol·K)
na
ꢀT (K)
1a
H
H
H
H
H
H
H
H
OMe
H
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
CDBr3
74.7
74.1
73.7
76.3
75.1
75.7
76.0
77.2
78.2
78.6
79.2
79.9
80.7
79.7
73.3
70.0
71.2
72.3
70.9
71.1
70.2
68.4
60.4
62.4
66.6
75.5 0.8
73.3 0.5
70.7 0.7
78.5 0.7
68.1 0.7
72.1 0.7
69.8 1.0
74.9 0.4
81.3 0.5
86.0 1.3
83.0 0.6
82.8 0.6
84.1 0.9
83.8 1.0
75.4 0.5
66.7 0.6
68.0 0.7
69.7 0.5
67.3 0.4
70.4 0.7
69.5 0.6
66.0 0.8
55.6 0.6
60.9 0.4
64.4 0.7
2.7 2.3
−2.3 1.5
−9.9 1.5
7.5 1.8
15
14
13
12
20
14
21
23
24
20
19
29
32
27
19
16
22
15
12
10
10
12
33
23
18
314–419
316–397
314–418
314–418
331–395
323–400
325–395
331–393
329–405
323–376
334–418
336–416
339–397
311–374
315–385
305–399
311–391
321–391
311–400
297–334
287–334
283–328
268–339
268–346
280–356
1b
1c
H
H
H
H
Cl
1d
2cb
3a
H
H
H
H
NO2
Cl
Me
Me
Me
Me
i-Pr
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
Cl
H
H
H
−23.4 1.8
−11.9 2.0
−20.8 2.8
−7.7 1.0
10.4 1.3
24.9 3.6
13.0 1.6
10.1 1.5
11.8 2.6
13.5 2.9
7.0 1.4
Me
Me
Me
i-Pr
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
H
H
H
OMe
Cl
3c
H
H
3d
4cc
5a
H
H
NO2
Cl
H
H
H
H
OMe
Cl
5c
H
H
5dd
5d
5d
6cb
7a
H
H
NO2
NO2
NO2
Cl
H
H
H
H
C6D5CD3
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
CDCl3
H
H
Cl
Cl
H
H
OMe
Cl
−11.1 1.6
−10.9 1.9
−8.8 1.4
−12.2 1.3
−2.3 2.0
−2.4 1.9
−8.1 2.6
−16.1 2.0
−5.1 1.4
−7.4 2.1
7c
Cl
Cl
H
H
7d
8ce
9a
Cl
Cl
H
H
NO2
Cl
Br
Br
H
H
H
H
Cl
Cl
Cl
Me
Me
Me
Cl
Cl
Cl
Me
Me
Me
OMe
Cl
9c
H
H
CDCl3
9d
10a
10c
10d
H
H
NO2
OMe
Cl
CDCl3
H
H
CDCl3
H
H
CDCl3
H
H
NO2
CDCl3
a Number of points used for the calculation of thermodynamic parameters.
b For compounds 2c, 6c, the thermodynamic characteristics for the isomerization processes are listed for the transformation from the more stable
state (E) into the less stable one (Z). The E-/Z-isomer ratios for these species are 0.63/0.37 and 0.52/0.48, respectively (298 K).
c Data from Ref. [2c]
d Data from Ref. [11]
e Published data [2c]: ꢀG298K = 70.4 kJ/mol, ꢀH = 65.9 kJ/mol, ꢀS = −15.1 J/mol · K, ꢀT = 305–355 K (200 MHz).
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2008, 46, 811–817

Z, E-Isomerization mechanism for N-arylthio-1,4-benzoquinonimines
Results and Discussion
mechanism for N-arylthio-1,4-benzoquinonimines. In the case of
the rotational mechanism the substituents at positions 3 and 5
would enlarge the activation energy due to the sterical effects.[1a]
It was shown previously[7,8] that the electron acceptor sub-
stituents at the para-position of the phenyl ring increased
activation energies for the isomerization of N-arylthio-substituted
compounds. However, our results indicate that this trend holds
for all investigated series with only exception for the 3,5-
dichlorosubstituted thioquinonimines 9a–d where the opposite
trend has been found. Indeed, it is evident from Table 1 that
within 9a–d the electron attracting substituents in the phenyl
ring reduce the activation energy for the Z, E-isomerization. The
reason for such a phenomenon could not be identified by simple
terms. Again, a more complicated mechanism could be proposed
involving along with inversion a contribution of rotation also. In
order to elucidate the isomerization mechanism for thioquinon-
imines and to interpret the unique behavior of 9a–d we have
carried out quantum chemistry investigations for the series of the
N-arylthio-1,4-benzoquinonimines at the DFT (B3LYP/6–31+G∗)
level of theory.
As one can see, the activation ꢀG298K energy values for the unsub-
stituted and 2,6-disubstituted N-arylthio-1,4-benzoquinonimines
raise with increasing electron acceptor properties of the sub-
stituents in the phenyl ring. While this holds true for all types of the
investigated 2,6-disubstituted N-arylthio-1,4-benzoquinonimines,
one can observe such an increase in the case of the unsubsti-
tuted compounds only if the strongly electron-attracting nitro
group is attached to the para-position of the phenyl ring (com-
pound 1d). Our results demonstrate also an essential dependence
of the isomerization barrier energies on the properties of the
substituents in the quinonimine moiety. The alkyl substitution
at positions 2 and 6 (compounds 2–5) increase the activation
energy. Moreover, the latter increases with the raising donor prop-
erties of the substituents (from Me to tert-Bu) and their number.
In contrast, replacing hydrogen at positions 2 and 6 by halogen
(compounds 6–8) reduces the activation energy for isomeriza-
tion. The measured barrier values decrease consecutively from
the unsubstituted (1c) to the monosubstituted (6c) and disubsti-
tuted species (7c). The solvent polarity does not practically affect
the measured activation energies (Table 1, compound 5d). At the
same time, one should note the barriers raising in proton-donor
solvents, such as bromoform, which is in line with the inversion
isomerization mechanism for the investigated objects.[12]
Quantum Chemistry Calculations
To the best of our knowledge, no quantum chemistry calculations
have been reported for quinonimines up to date. The theoretically
investigated structures together with the ZPE-corrected total
energies for all calculated N-arylthio-1,4-benzoquinonimines, as
well as the relative energies are listed in Table 2. A complete
According to our data, the activation energies for iso-
merization decrease for 3,5-dimethylsubstituted N-arylthio-1,4-
benzoquinonimines (Table 1, 10a–d), as compared to all other
congeners. This fact also agrees with the inversion isomerization
Table 2. Calculated total (E)a and relative (ꢀE) energy values for N-arylthio-1,4-benzoquinonimines (1–10)
Structure
EA (a.u.)
EB (a.u.)
EC (a.u.)
ED (a.u.)
ꢀEAB (kJ/mol)
ꢀEAD (kJ/mol)
1a (syn)b
(anti)
1b
−1105.152461
−1105.152560
−990.658619
−1195.166138
−1183.740201
−1183.740214
−1069.246581
−1273.755056
−1419.443164
−1419.443084
−1304.949604
−1509.457971
−2024.356067
−2024.356108
−1909.861816
−2114.368505
−2024.350393
−2024.350266
−1909.856111
−2114.362812
−1183.730212
−1183.730323
−1069.236104
−1273.743892
−1105.125906
−1105.125813
−990.631540
−1195.137803
−1183.712801
−1183.712426
−1069.218723
−1273.725779
−1419.415591
−1419.415646
−1304.921479
−1509.428626
−2024.330347
−2024.330352
−1909.835613
−2114.340867
−2024.332029c
−1105.146050c
−1105.135480
−1105.135523
−990.640778
−1195.149509
−1183.723882
−1183.723903
−1069.229750
−1273.739480
−1419.426618
−1419.426650
−1304.932665
−1509.442606
−2024.338514
−2024.338606
−1909.843472
−2114.351123
−2024.322699
−2024.322726
−1909.827346
−2114.335432
−1183.708028
−1183.707931
−1069.214564
−1273.723888
69.7
70.2
71.1
74.4
71.9
73.0
73.1
76.9
72.4
72.0
73.8
77.0
67.5
67.6
68.8
72.6
47.9
44.5
44.8
46.8
43.7
42.8
42.8
44.2
40.9
43.4
43.1
44.5
40.3
46.1
46.0
48.2
45.6
72.7
72.3
75.5
71.9
58.2
58.8
56.6
52.5
−990.650407
−1195.154069
−1183.733157c
1d
3a (syn)b
(anti)
3b
−1069.237667
−1273.742123
−1419.435405c
3d
5a (syn)b
(anti)
5b
−1304.939990
−1509.444672
−2024.350241c
5d
7a (syn)b
(anti)
7b
−1909.854242
−2114.357091
−2024.334448c
7d
9a(syn)b
(anti)
9b
−1909.836483
−2114.340855
−1183.710021c
−1909.838763
−2114.343004
−1183.714495c
51.5
57.6
53.0
9d
10a(syn)b
(anti)
10b
−1069.214731
−1273.719441
−1069.218951
−1273.723344
56.1
64.2
10d
a Here the only ZPE-corrected total energy values are shown. The calculated data are presented in more detail in Supporting Information.
b Two different conformations with syn- and anti-orientation of the OMe group relatively to the S–N bond.
c Only one isomer exists.
c
Magn. Reson. Chem. 2008, 46, 811–817
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

V. V. Pirozhenko et al.
(a)
(b)
Figure 2. MOLDEN plots for the calculated transition-state structures
5d and 10d for the inversion reaction. The most important geometry
parameters are (bond lengths in Å, bond angles in degrees): C O 1.234(a),
1.235(b), C N 1.290(a), 1.283(b), N–S 1.563(a), 1.574(b), S–C1 1.814(a),
1.836(b), ∠C1NS 178.7◦(a), 178.2◦(b), ∠NSC1ꢁ 110.1◦(a), 108.3◦(b).
(a)
(b)
Figure 1. MOLDEN plots for the equilibrium structures of N-4-nitrophenyl-
2,6-bis(tert-butyl)-1,4-benzoquinonimine, 5d (a) and of N-4-nitrophenyl-
3,5-dimethyl-1,4-benzoquinonimine, 10d (b). The most important cal-
culated geometry parameters are as follows (bond lengths in Å, bond
angles in degrees, in parentheses the experimentally found parameters for
5d[13]) : C O 1.233 [1.227(4)] (a), 1.234 (b), C N 1.309 [1.302(5)] (a), 1.305
(b), N–S 1.684 [1.663(3)] (a), 1.672 (b), S–C1 1.779 [1.761(4)] (a), 1.783 (b),
∠C1NS 121.3 [119.6(3)] (a), 128.1 (b), ∠NSC1ꢁ 100.7 [100.10(17)] (a), 100.7
(b), ∠C1NSC1ꢁ 180.0 (a), 180.0 (b).
via rotation. The equilibrium structures for the transition states
are characterized by an almost linear configuration of the C N–S
triad (as an example, the transition state for the isomerization for
5d is shown in Fig. 2(a)). Our attempts to localize transition states
for the rotation about the C N bond (starting the optimization
from the structure where the C N–S triad is nonlinear and the
N–S bond is orthogonal to the quinonimine ring) have led to the
transition state structures discussed above. This means that the
former conformations do not correspond to the real transition
states but are the saddle points of the higher order. Most probably
the isomerization of the N-arylthio-1,4-benzoquinonimines via
rotation does not proceed.
The quantum chemistry investigation of the Z, E-isomerization
of thioquinonimines demonstrates that the process considered
has a more complicated character than the simple transformation
of one conformation into another via the nitrogen inversion.
Our calculations predict the second local minima in energy
for N-arylthio-1,4-benzoquinonimines as the products of the
isomerization reaction and additional transition states, which are
consistent with the rotation about the N–S bond. A full description
of the isomerization in the series of the studied thioquinonimines
is demonstrated in Fig. 3 for N-phenylthio-1,4-benzoquinonimine
(1b) as an example. An initial structure A converts to the second
local minimum in energy (C) proceeding via the transition state
(B). Then, C transforms to Aꢁ via rotation about the N–S bond.
The second transition state D exists on the path from C to Aꢁ.
data set can be found in the Supporting Information. The
structures have been calculated corresponding to the ground
states and transition states. For all series, the species have been
studied, substituted with the OMe, H, NO2 groups at the para-
position of the phenyl moiety. Such a choice of substituents
was determined by our wish to reproduce the effects of the
π-donor and π-acceptor groups and to compare the calculated
structural and thermodynamic characteristics with those found in
the experiment.
The MOLDEN view for the equilibrium structure of N-(4ꢁ-
nitrophenyl)-2,6-bis(tert-butyl)-1,4-benzoquinonimine (5d) is pre-
sented in Fig. 1(a) and the calculated geometry parameters are
comparedwiththosepreviouslyfoundwiththesinglecrystalX-ray
diffraction method.[13] A good agreement between these two data
sets is an evidence of the proper choice of the calculation level.
Our calculation results support the suggestion that the
isomerization proceeds via the inversional mechanism and not
A
B
C
D
A′
Figure 3. Optimized structures for the ground states (A, Aꢁ), second local minimum (C) and transition states (B, D) for N-phenyl-1,4-benzoquinonimine
1b.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2008, 46, 811–817

Z, E-Isomerization mechanism for N-arylthio-1,4-benzoquinonimines
80
70
60
50
40
30
20
10
0
80
74.4
D 71.9
70
B
60
57.6
C 52.0
B
50
43.7
40
D
31.7
C
30
20
10
A
A′
A
A′
0
Figure 4. Graphical representation of the dynamic processes and relative energies for different states of 1d (left) and 9d (right).
Thus, a studied degenerated isomerization really consists of the
inversional and rotational parts, but these contribute to the two
separate processes. In Table 2, the ꢀEAB values determine the
activation barriers for the inversion and ꢀEAD magnitudes are
the activation energies for the rotation about the N–S bond.
As one can see, the theoretical data are generally in a good
agreement with the experimentally found activation barriers for
isomerization. In accordance with the experiment, the inversion
barrier energies for all 2,6-disubstituted benzoquinonimines raise
with increasing electron acceptor properties of the substituents
in the aromatic moiety. Theory also reproduces well a trend of
the substituent influence on the isomerization barrier height. This
becomes obviously higher for the 2,6-dialkylsubstituted species
3,5andlowerinthecaseofthe2,6-dichloro-substitutedcongeners
7 compared to the unsubstituted compound 1.
At the same time, theoretical calculations predict lower
activation energies ꢀEAB for the inversion reaction for 3,5-
disubstituted thiobenzoquinonimines 9,10. The transition state
structure for the representative of this series, 10d, is shown in
Fig. 2(b). The obvious difference in comparison with 5d is the
orthogonal orientation of the phenyl ring relative to the C1NS
plane but this structural feature is probably not responsible for
the low activation energies found in the experiment for 10a–d.
For structure 10d corresponding to the main minimum in energy
(Fig. 1(b)) the C1NS bond angle (128.1◦) is significantly larger
than those calculated for 5d (121.3◦) and 3d (121.8◦) due to
the steric effect of the methyl substituents at positions 3,5. This
trend is even more pronounced for the 3,5-dichloro-substituted
thiobenzoquinonimines 9 (for instance, ∠C1NS = 130.5◦ for 9d).
This destabilization of the ground states for 9,10 diminishes the
calculated activation energies for inversion for these structures. At
the same time, the calculations suggest for series 9,10 increasing
barriers of rotation about the N–S bond (larger ꢀEAD values,
Table 2). For the unsubstituted (1) and 2,6-dialkylsubstituted
species (3,5), the activation barriers for rotation are definitely
lower than those for the inversion (by approx. 25–39 kJ/mol).
In contrast, for the 3,5-dimethylsubstituted benzoquinonimines
10, these magnitudes are already commensurable, and in the
case of 3,5-dichlorosubstituted congeners (9) the ꢀEAD values
exceed the ꢀEAB magnitudes. This means that in the case of 9, the
rotation about the N–S bond and not the inversion determines
the experimentally measured activation energies. Schematically it
can be represented as shown in Fig. 4.
Our calculations predict that the corresponding activation
energy on rotation about the S–N bond will decrease with in-
creasing electron-withdrawing properties of the para-substituent,
i.e. the lowest barrier on rotation is inherent to the compounds
with nitro group in the para-position of the phenyl moiety (9d).
This fact is in line with the experiment (Table 1). Therefore, a
unique behavior of ꢀG298K values found for series 9 reflects a
dominating contribution of the rotation about the N–S bond into
the experimentally measured activation energies (Fig. 4(b)).
The calculated energy barriers for C–D–Aꢁ transformation, i.e.
for the rotation about the N–S bond for the unsubstituted and
2,6-disubstituted derivatives are in the range of 40–50 kJ/mol, and
thereforecanbemeasuredusingtheNMRspectroscopy. However,
by the NMR investigation of 5d at low temperatures (at −80 ◦C
in toluene-d8 and at −60 ◦C in chloroform-d) only remarkable
broadening of the signals, but no separation for resonances
corresponding to the conformations A and C has been observed.
This is in line with the significant differences in total energies for
these two local minima predicted by theory. Due to the negligibly
low population of the intermediate C calculated for all compounds
of interest, the above-mentioned dynamic processes cannot be
investigated with the NMR spectroscopy under used experimental
conditions.
Conclusions
The theoretical and experimental data make it possible to affirm
that the isomerization of the N-arylthio-1,4-benzoquinonimines
should be considered as a combination of the two separate
dynamic processes, the rotation about the N–S bond and the
inversion via transition state B with the linear arrangement of
the C N–S triad. The experimentally measured free activation
energy for the isomerization process can be determined either
by inversion stage or by the rotation about the N–S bond.
A good agreement has been found between the experimental
ꢀG298K values and theoretically calculated differences in total
energies of the ground states and transition states (ꢀE). A trend
of the substituent influence on the activation barrier energies is
reproducedbytheoryaswell.Combiningthevariable-temperature
c
Magn. Reson. Chem. 2008, 46, 811–817
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

V. V. Pirozhenko et al.
NMR approach with the quantum chemical calculations is
obviously a good way to determine contributions from the
different dynamic processes into the experimentally measured
free-activation energy values.
A hybrid functional B3LYP[24,25] was chosen, in combination with
the 6–31+G∗[26] basis sets. The transition states were localized
using the following routine. First the guessed structure for the
transition state was optimized as a Z-matrix with a frozen (approx.
180◦) C N–S angle. Then the search of the transition state was
performed using option Opt = (CalcFC,TS) with the optimization
of all structural parameters. This simple approach can generally
be useful for the localization of the low-energy transition states
(within few kcal/mol). The vibrational analyses were performed
for all the local minima in energy (no imaginary vibrations)
and transition states (one imaginary vibration) using the level
of theory mentioned above and calculating analytically first
and second derivatives (Supporting Information). The MOLDEN
program[27] was used for the graphical presentation of the
optimized structures.
Experimental
The 1H (299.95 MHz) NMR spectra were recorded with Varian VXR-
300 NMR spectrometer. The thermodynamic parameters for the
isomerization processes (ꢀG298K, ꢀH, ꢀS) were calculated using
the Eyring equation[6] based on the rate constants determined
by the comparison of the experimental and simulated spectra at
the different temperatures. The complete set of the simulated
and experimental spectra for 3d as well as the rate constant
sets for all experimentally studied compounds are collected
in the Supporting Information. The temperature intervals and
the number of data points used for the calculation of the
thermodynamic parameters are listed in Table 1. The simulation of
the theoretical spectra was carried out using the WINDNMR[14] and
DNMR[15] program sets and our own modification of the DNMR3
program.[16] The rate constants were estimated either by the visual
comparison of the theoretical and experimental spectra (in the
case of the WINDNMR and DNMR3 program packets) or using
the least-squares-fit of experimental versus simulated spectra (for
the DNMR program packets) with the following visual comparison
of the experimental and simulated spectra. For the more precise
calculation of the activation energy values the line shape analyses
were carried out both for the quinonimine proton signals and
alkyl group resonances. For the correct determination of the rate
constants in the fast exchange area, the temperature dependence
of the chemical shift values was taken into account for the
signals involved in the exchange processes. For this purpose the
spectra were additionally recorded at the temperatures 20–30 K
lower than the range shown in Table 1. The changes in the
population differences of the corresponding isomers with the
increasing temperature were considered for the nondegenerated
isomerization processes (2c, 6c in Table 1). The estimated errors
for the activation energy ꢀG determination did not exceed 1%.
The temperature was determined with a precision of 1 K. The
spectra were recorded with a digital resolution 0.10–0.20 Hz per
point. The choice of the solvents was determined by the necessary
temperature range.
Supporting information
Supporting information may be found in the online version of this
article.
Acknowledgements
The authors thank Professor Dr. U. Manthe and Professor
Dr. W. W. Schoeller, University of Bielefeld (Germany) for the
access to the computer cluster and GAUSSIAN-03 program set.
References
[1] (a) H. Kessler, Tetrahedron 1974, 30, 1861; (b) G. E. Hall,
W. J. Middleton, J. D. Roberts, J. Am. Chem. Soc. 1971, 93, 4778;
(c) F. Kerek, G. Ostrogovich, J. Chem. Soc. B. 1971, 541.
[2] (a) H. Kessler, Angew. Chem. 1967, 79, 997; (b) A. V. Prosyanik,
N. Y. u. Kol’tsov, V. A. Romanchenko, V. V. Belov, K. S. Burmistrov,
S. V. Loban, Zhurn. Org. Khim. (Russ.) 1987, 23, 375;
(c) V. V. Pirozhenko, Y. u. P. Egorov, Ukr. Khim. Zhurn. 1992, 58, 567.
[3] A. Rieker, H. Kessler, Tetrahedron 1967, 23, 3723.
[4] V. V. Pirozhenko, A. P. Avdeenko, Zhurn. Org. Khim. (Russ.) 1995, 31,
1686; Russ. J. Org. Chem. 1995, 31, 1514.
[5] V. V. Belov, S. V. Loban, K. S. Burmistrov, A. V. Prosyanik, Zhurn. Org.
Khim. (Russ.) 1983, 19, 825.
[6] I. Sandstrom, Dynamic NMR Spectroscopy, Academic Press: New
York, London, 1982.
[7] V. V. Pirozhenko, A. P. Avdeenko, A. L. Yucina, A conformational
stability of N-arylthio-1,4-benzoquinonmonoimines. In Abstracts of
XIX Ukranian Conference of Organic Chemistry, 2001, p 122.
[8] C. Brown, B. T. Grayson, R. F. Hudson, J. Chem. Soc., Perkin Trans. 2
1979, 427.
N-Arylthio-1,4-benzoquinonimines (1a–10d)
To the suspension of 2 mmol of p-aminophenol in 20 ml of dry
diethylether, 6 mmol of the corresponding arylsulfene chloride
was added by intensive stirring and cooling to −10 ◦C. The
equivalent quantity of triethylamine was added dropwise to the
reaction mixture. Triethylamine hydrochloride was filtered off;
the filtrate was evaporated in vacuo. The residue was washed
with methanol and glacial acetic acid and recrystallized from
glacial acetic acid or heptane. The analytical and spectral data for
quinonimines were published earlier: (1b, 3b, 7b),[17,18] (1c),[19]
(1a–c),[20] (1d, 3a–3d, 4c, 5c–d, 10a–d),[21] (2c, 3a, 5a),[11,13] (6c,
7a–d, 8c, 9a–d).[22]
[9] D. Y. Curtin, E. J. Grubbs, C. G. McCarty, J. Am. Chem. Soc. 1966, 88,
2775.
[10] D. Kost, E. Carlsson, M. Raban, J. Chem. Soc. Chem. Commun. 1971,
655.
[11] A. P. Avdeenko, V. V. Pirozhenko, I. V. Konovalova, S. A. Yucina,
A. L. Stanovsky, Zhurn. Org. Khim. (Russ.). 2004, 40, 1340; Russ. J.
Org. Chem. 2004, 40, 1291.
[12] (a) H. Kessler, D. Leibfritz, Tetrahedron 1969, 25, 5127; (b) H. Kessler,
D. Leibfritz, Tetrahedron 1970, 26, 1805.
[13] A. P. Avdeenko, V. V. Pirozhenko, S. A. Konovalova, A. A. Santalova,
A. V. Vakulenko, Arkivoc 2005, 8, 60.
[14] H. J. Reich, WINDNMR: NMR Spectrum Calculation, Version 7.1.10.
Department of Chemistry, University of Wisconsin, Madison 2005,
www.chem.wisc.edu/areas/reich/plt/windnmr.htm.
[15] V. V. Pirozhenko, Spectroscopic investigations of the electronic
structure and conformational stability of some quinonimine
derivatives. PhD Thesis, Kiev, 1992, 174 p.
[16] D. A. Kleier, G. Binch, Quantum Chemistry Program Exchange, 1973,
N 165.
[17] D. N. Kramer, R. N. Gamson, J. Org. Chem. 1959, 24, 1154.
Calculation Details
The geometries for all structures were optimized using the
GAUSSIAN-03 program set[23] within the DFT approximation.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2008, 46, 811–817

Z, E-Isomerization mechanism for N-arylthio-1,4-benzoquinonimines
[18] D. H. R. Barton, I. A. Blair, P. D. Magnus, R. K. Norris, J. Chem. Soc.,
Perkin Trans. 1 1973, 1031.
[19] J. E. Baldwin, R. K. Norris, J. Org. Chem. 1981, 46, 697.
[20] V. V. Pirozhenko, K. S. Burmistrov, V. V. Belov, V. I. Nichvoloda, Ukr.
Khim. Zhurn. 1992, 58, 567.
[21] V. V. Pirozhenko, A. P. Avdeenko, A. L. Yucina, S. A. Konovalova,
Zhurn. Org. Khim. (Russ.). 2004, 40, 1169; Russ. J. Org. Chem. 2004,
40, 1121.
[22] V. V. Pirozhenko, A. P. Avdeenko, A. L. Yucina, S. A. Konovalova,
Zhurn. Org. Khim. (Russ.). 2002, 38, 1192; Russ. J. Org. Chem. 2002,
38, 1142.
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian
03, Revision B.03, Gaussian: Pittsburgh, 2003.
[24] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[25] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[23] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
[26] (a) R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971, 54,
724; (b) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972,
56, 2257; (c) P. C. Hariharan, J. A. Pople, Mol. Phys. 1974, 27, 209;
(d) P. C. Hariharan, J. A. Pople, Thermochim. Acta 1973, 28, 213.
[27] MOLDEN
Program,
Ver.
3.8
(2005)
(http://www.cmbi.ru.nl/molden/molden/html);
G. Schaftenaar,
J. H. Noordik, J. Comput.-Aided Mol. Design, 2000 14, 123.
c
Magn. Reson. Chem. 2008, 46, 811–817
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc