2D-NMR, X-ray crystallography and theoretical studies of the reaction mechanism for the synthesis of 1,5-benzodiazepines from dehydroacetic acid derivatives and o-phenylenediamines

Article abstract of DOI:10.1016/j.molstruc.2013.12.082

The synthesis of 1,5-benzodiazepines by the reaction of o-phenylenediamines (o-PDAs) with dehydroacetic acid DHAA [3-acetyl-4-hydroxy-6-methyl-2H-pyran-2- one] or conjugate analogues is largely reported in the literature, but still with uncontrolled stereochemistry. In this work, a comprehensive mechanistic study on the formation of some synthesized 1,5-benzodiazepine models following different organic routes is established based on liquid-state 2D NMR, single-crystal X-ray diffraction and theoretical calculations allowing the classification of two prototropic forms A (enaminopyran-2,4-dione) and B (imino-4-hydroxypyran-2-one). Evidences are presented to show that most of the reported 1,5-benzodiazepine structures arising from DHAA and derivatives preferentially adopt the (E)-enaminopyran-2,4-diones A.

Full text of DOI:10.1016/j.molstruc.2013.12.082

Journal of Molecular Structure 1061 (2014) 97–103
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
2D-NMR, X-ray crystallography and theoretical studies of the reaction
mechanism for the synthesis of 1,5-benzodiazepines from dehydroacetic
acid derivatives and o-phenylenediamines
c
Amal Rabahi a, Safouane M. Hamdi b, Yahia Rachedi a, Maamar Hamdi a, , Oualid Talhi ,
⇑
Filipe A. Almeida Paz d, Artur S.M. Silva c, , Balegroune Fadila e, Hamadène Malika e, Taïbi Kamel e
⇑
a Laboratoire de Chimie Organique Appliquée (groupe hétérocycles), Faculté de Chimie, Université des Sciences et de la Technologie Houari Boumediène BP 32,
El-Alia, Bab-Ezzouar, 16111 Alger, Algeria
b CHU Toulouse Laboratoire de Biochimie, IFB, Toulouse F-31000, France
c QOPNA, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
d CICECO, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
e Laboratoire de Crystallographie-Thermodynamique, Faculté de Chimie, Université des Sciences et de la Technologie Houari Boumediène BP 32, El-Alia,
Bab-Ezzouar, 16111 Alger, Algeria
g r a p h i c a l a b s t r a c t
h i g h l i g h t s
ꢀ
ꢀ
ꢀ
ꢀ
1,5-Benzodiazepines from o-
phenylenediamines and
dehydroacetic acid is discussed.
Difficulties of characterizing the
prototropy of 1,5-benzodiazepines
were discussed.
The presence of the (E)-amino-en-one
tautomer of 1,5-benzodiazepines A
was confirmed.
Theoretical calculations, NMR data
and especially single-crystal X-ray
diffraction were used.
X
X
X
OH
O
O
H
H
H
O
O
N
O
O
N
O
R
O
O
N
O
NH
R
NH
R
o-PDAs
NH
R
O
O
o-PDAs
DHAA
A
amino-en-one
B imino-en-ol
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 28 October 2013
Received in revised form 24 December 2013
Accepted 28 December 2013
Available online 8 January 2014
The synthesis of 1,5-benzodiazepines by the reaction of o-phenylenediamines (o-PDAs) with dehydroace-
tic acid DHAA [3-acetyl-4-hydroxy-6-methyl-2H-pyran-2-one] or conjugate analogues is largely reported
in the literature, but still with uncontrolled stereochemistry. In this work, a comprehensive mechanistic
study on the formation of some synthesized 1,5-benzodiazepine models following different organic
routes is established based on liquid-state 2D NMR, single-crystal X-ray diffraction and theoretical calcu-
lations allowing the classification of two prototropic forms A (enaminopyran-2,4-dione) and B (imino-4-
hydroxypyran-2-one). Evidences are presented to show that most of the reported 1,5-benzodiazepine
structures arising from DHAA and derivatives preferentially adopt the (E)-enaminopyran-2,4-diones A.
Ó 2014 Elsevier B.V. All rights reserved.
Keywords:
Benzodiazepines
Dehydroacetic acid
(E/Z)-Diastereomers
2D-NMR
X-ray crystallography
Theoretical chemistry
1. Introduction
Benzodiazepines are a class of outstanding organic compounds
mainly due to their application as psychoactive drugs, which are
used in the treatment of anxiety, insomnia, psychomotor agitation,
convulsions, spasms, or in the context of alcohol withdrawal syn-
drome. Benzodiazepines act on neurotransmitters in neurons of
⇑
Corresponding authors. Tel.: +213 21247311 (M. Hamdi), tel.: +351 234370714
(A.S.M. Silva).
E-mail addresses: artur.silva@ua.pt (A.S.M. Silva), prhamdi@gmail.com
(H. Malika).
0022-2860/$ - see front matter Ó 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2013.12.082

98
A. Rabahi et al. / Journal of Molecular Structure 1061 (2014) 97–103
the central nervous system by increasing their inhibitory activity.
They are also used to induce a state of sedation or for their
hypnotic, anxiolytic, antiepileptic, muscle relaxant and amnesic
activities. Because of such biological importance, the synthesis of
benzodiazepine has received much attention in the field of medic-
inal chemistry [1]. In this context, our interest is especially focused
on 1,5-benzodiazepines obtained from the reaction of o-phenylen-
ediamines (o-PDAs) with 3-acetyl-4-hydroxy-6-methyl-2H-pyran-
2-one (DHAA) or conjugate derivatives, such as 3-[(2E)-3-(4-aryl/
styryl)prop-2-enoyl]-4-hydroxy-6-methyl-2H-pyran-2-ones 1a,b
which have been thoroughly studied. Among these benzodiaze-
pines various forms can be sorted from the recent literature
descriptions [2–11]. Accordingly, we distinguish the two tauto-
meric forms, including enaminopyran-2,4-dione A and imino-4-
hydroxypyran-2-one B, which can be synthetically accessible
through different reaction pathways (a and b) (Scheme 1) by cou-
pling DHAA or conjugate analogues 1a,b with o-PDAs, for instance,
in the presence of catalytic amounts of acetic acid [2–5]. Form B
was reported to result from the reaction of o-PDA with DHAA to
form the imino intermediate 2 followed by the addition of
aromatic aldehydes, giving rise 3 and finally to the diazepine cyc-
lisation that was carried out under conventional heating conditions
in the presence of catalytic amounts of trifluoroacetic acid (TFA)
(Scheme 1, pathway b) [6]. Investigations such as those conducted
by Prakash et al. [2] have also reported the synthesis of benzodi-
azepines B, underlining the same physicochemical characteristics
as those described by Fodili et al. [6], and undoubtedly confirming
that the 4-hydroxy-6-methyl-2-pyrone scaffold is conserved in the
resulting benzodiazepine scaffolds, further supported by the work
performed by Chergui et al. [7]. Other synthetic attempts have,
however, shown that the reaction of DHAA with o-PDA in refluxing
alcohols (R0OH) may cause a pyrone ring opening through the
amino attack at position 6 of DHAA, and diazepine ring closure at
position 4 affording esters of 2-{(Z)-4-methyl-1H-benzo[b][1,4]
diazepin-2(3H)-ylidene}acetic acid 5, which were classified as
another interesting form of 1,5-benzodiazepines but received far
less attention [8] (Scheme 1). Claramunt et al. [3] have recently
reported the synthesis of benzodiazepines A by adopting exactly
the same procedures that allowed the access to benzodiazepines
B [2], these findings reveal the difficulties faced in differentiating
between benzodiazepines of form A or B in solution using NMR.
To date, Kaoua et al. [9] discussed the possibility to obtain 1,5-ben-
zodiazepines B from the reaction of o-PDA with DHAA using het-
eropolyacids as catalysts.
The A and B forms of 1,5-benzodiazepines constitute a pair of
functional tautomers due to the possible 1,5-prototropic shift be-
tween amino-en-one and imino-en-ol, respectively. To the best of
our knowledge, no report to date has revealed their coexistence.
It should also be mentioned that form A can display (E/Z)-amino-
en-one
isomerism
with
an
intramolecular
hydrogen
(C@Oꢁ ꢁ ꢁHANA) bond. In form B the intramolecular hydrogen bond
(AOHꢁ ꢁ ꢁN@CA) can impose the s-cis conformation depicted in
Scheme 1. Despite the presence of an asymmetric carbon (C-4)
on the diazepine ring, no enantioselective preparation of these
benzodiazepines is yet reported.
In this work our focus is oriented to the A and B prototropic
forms of 1,5-benzodiazepine reviewing their 1H NMR data analysis
reported in the literature. To this end, we have retaken their prep-
arations under the reported experimental conditions [2,3,6] in or-
der to establish a comparative study to allow the setup of adequate
operating conditions. The target benzodiazepines 4a–c were syn-
thesized by the reaction of o-PDA or p-Me-o-PDA with DHAA deriv-
atives 1a,b (Scheme 1) to study and confirm their structures in
solution based on 2D NMR analysis. While in the solid state, sin-
gle-crystal X-ray diffraction studies allowed the confirmation of
the structure A of the designed 1,5-benzodiazepine. We further up-
date the mechanistic explanations for the formation of 1,5-benzo-
diazepine on the basis of theoretical chemistry calculations of their
total energy and electronic density distribution of the starting
materials. This includes studies applied on some proposed reaction
intermediates involved in the A/B tautomerization and rearrange-
ment steps.
2. Experimental
Melting points were determined on a Stuart scientific SPM3
apparatus fitted with a microscope and are uncorrected. NMR
spectra were recorded on a Bruker Avance 300 spectrometer (oper-
ating at 300.13 for 1H and 75.47 MHz for 13C), in CDCl3 as solvent.
Chemical shifts (d) are reported in ppm and coupling constants (J)
in Hz. The signals are described as s (singlet), d (doublet), dd (dou-
blet of doublet), ddd (double doublet of doublet) and m (multiple).
The internal standard was TMS. Unequivocal 13C assignments were
made with the aid of 2D HSQC and HMBC (delays for one bond and
long-range JC/H couplings were optimized for 145 and 7 Hz, respec-
tively) experiments. NOESY spectra aided in the 1H resonances
assignments and in the spatial arrangements of the structures
(mixing time 800 ms). Electron impact mass spectra were obtained
X
a. R = styryl, X = H
b.
c.
R = ortho-hydroxyphenyl, X = H
R = styryl, X = methyl
( )-A
Z
X
OH
O
O
O
H
O
N
NH
R
B
H
R
o-PDAs
O
O
O
N
O
4a-c
NH
R
O
H
R"
1a,b
O
O
N
O
X
(E)-A
Pathway (a)
RCHO
H
O
N
NH
6
OH
O
R"NH2
R
O
O
O
O
MeOH / TFA
NH
o-PDA
R'OH
DHAA
X
X
o-PDAs
Pathway (b)
N
O
OH
O
N
O
OH
O
N
N
R'O
5
NH2
RCHO
R
O
3
2
Scheme 1. Different organic pathways for the synthesis of 1,5-benzodiazepines from DHAA and o-phenylenediamines.

A. Rabahi et al. / Journal of Molecular Structure 1061 (2014) 97–103
99
at 70 eV electron impact ionization using Nermag R 10-10C qua-
druple mass spectrometer. Exact mass measurements were re-
corded on high resolution mass spectrometer micrOTOF-Q and
elemental analysis on Truspec 630-200-200 equipment. Analytical
thin-layer chromatography (TLC) was performed on pre-coated sil-
ica gel plates 60-F-250 (Merck; 0.25 mm) developing with chloro-
form/methanol, 10/1 v/v). The PM6 semi-empirical Hamiltonian
method for the theoretical calculations of total energy and elec-
tronic density distribution is performed using Gaussian 09 [12].
Dehydroacetic acid DHAA and the aldehydes were obtained from
Sigma–Aldrich. Compounds 4a–c were prepared according to liter-
ature procedures [2,4–6].
C@Oꢁ ꢁ ꢁHNA) can be, however, observed in the A or B forms. A cor-
rect assignment of the deshielded NMR proton signals to either an
NH or OH groups remains difficult due to a possible 1,5-prototropic
shift giving rise to interconverting functional tautomers. In the 1H
NMR spectra of phenylimino-DHAA intermediates (resulting from
DHAA + arylamino or o-PDA) such as derivatives 2 or 3 (Scheme 1),
it was found that the singlet signal attributed to the OH groups is
averaged between 15.5 and 15.9 ppm in CDCl3 solution. Dias
et al. reported, however, by both NMR (NH at 14.0–15.4 ppm, in
CDCl3) and X-ray diffraction the isolation of (E)-enaminopyran-
2,4-diones 6 from the reaction of DHAA with primary amines
[10] (Scheme 1). Likewise, a recent study discusses the reactivity
of structural analogues of DHAA with aromatic amines and proves,
with the aid of 2D NMR experiments, the existence of (E)-enamin-
opyran-2,4-dione isomer 6 as a unique product. The NH group cor-
responding to the amino-en-one A form in these compounds is
mainly assigned by a singlet appearing at ca. 16 ppm in the 1H
NMR spectra [11].
Based on these results, we assume the predominance of an (E/
Z)-enaminopyran-2,4-dione form (E/Z)-A in solution for the struc-
ture of our 1,5-benzodiazepines 4a–c derivatives. Theoretical cal-
culations using the PM6 semi-empirical Hamiltonian method
[12] have proved to yield useful evidence for the stereochemical
patterns by estimating the total energy (ET) of each of the proposed
1,5-benzodiazepine forms A and B (molecules are taken in gas
phase). For these studies we have selected 1,5-benzodiazepines
molecules with 4-styryl 4a or 4-o-hydroxyphenyl 4b substituents
while taking into consideration the effect of the (E/Z) and (4-R/S)
stereoisomery (Scheme 1, Table 1).
According to the results obtained for the energy calculation (ET)
(Table 1), the structures (E)-A-4a(4S), (E)-A-4b(4S) of the (E)-A
form, and B-4b(4R) of the B form seem to be the most likely fa-
vored concerning the stability in the gas phase. We have prepared
the 1,5-benzodiazepines compounds 4a and 4b following the
methods designed for the synthesis of forms A and B with the over-
all objective to isolate single-crystals suitable for a X-ray diffrac-
tion studies. Physical and spectral data analyses (mp, TLC, HPLC
and NMR) of the products 4a and 4b obtained by the distinct routes
(a) and (b) (Scheme 1) agree with those reported in the literature
[2–6,9]. Compounds 4a and 4b were further isolated as single crys-
tals which permitted their study using X-ray diffraction (Support-
ing Information) and the use of the molecular units as starting
molecular models for the energy calculations reported in Table 1.
ORTEP images of compounds 4a and 4b are given in Fig. 1. Both
compounds are composed of three main cycles: pyrone and diaze-
pine rings, plus a fused benzenic ring. The medium planes of these
different heterocycles show for both compounds 4a and 4b an al-
most total planarity for the pyrone and the benzenic rings. The
maximum deviations are 0.028(24) Å (atom C2) and 0.0383(19) Å
(atom C23) for 4a, and 0.0301(19) Å (atom C1) and 0.095(20) Å
(atom C21) for 4b. The dihedral angles between the two medium
planes are relatively small and calculated as 9.18(17)° and
5.20(19)° for 4a and 4b, respectively.
The calculation of the medium planes for the benzodiazepine
rings reveal significant deviations from planarity and highlights
the strong distortion of the cycle. A careful examination of
inter-atomic distances and angles resulted in consistent values
conventionally admitted for such chemical bonds underlining
the hybridization type (sp2, sp3). Accordingly, the C2AC7 bond
connecting the pyrone ring to that of the diazepine has a double
bond (Csp2ACsp2) character and the two carbonyl bonds, C1@O3
and C3@O2, have a quite similar inter-atomic distance values for
both compounds 4a and 4b. The inter-atomic distance correspond-
ing to the C1@O3 bond is close to the conventional value attributed
to a carbonyl bond, while that of the C3@O2 bond corresponds to a
clear elongation for a double bond (ca.+0.05 Å). This structural
(E,E)-6-Methyl-3-{4-styryl-4,5-dihydro-1H-benzo[b][1,4]diaze-
pin-2(3H)-ylidene}-3H-pyran-2,4-dione 4a: Mp 202–203 °C,
(198 °C [6]), Yield 80%. 1H NMR (300 MHz, CDCl3): d 2.15 (s, 3H,
60-CH3), 3.15 (dd, 1H, H-3, 2J 12.4 Hz, 3J 9.5 Hz), 3.82 (s, 1H, H-5),
3.95 (dd, 1H, H-3, 2J 12.4 Hz, 3J 3.7 Hz), 4.90 (ddd, 1H, H-4, 3J 3.7,
7.5 and 9.5 Hz), 5.75 (s, 1H, H-50), 6.30 (dd, 1H, H100, 3J 7.5 and
15.5), 6.65 (d, 1H, H-200, 3J 15.5 Hz), 6.62–7.75 (9H, Ar), 15.60 (s,
13
1H, NH) ppm. C NMR (75 MHz, CDCl3): d 19.9 (60-CH3), 34.9 (C-
3), 67.2 (C-4), 96.5 (C-30), 107.4 (C-50), 121.7 (C-6), 121.9 (C-8),
124.6 (C-11), 126.6 (C-400, 800), 127.6 (C-10), 127.9 (C-600), 129.4
(C-500, 700), 129.5 (C-7), 129.9 (C-100), 130.7 (C-200), 136.1 (C-300),
139.6 (C-11), 163.1 (C-2), 163.2 (C-60), 173.0 (C-20), 184.7 (C-40)
ppm. MS: m/z 373 (M + H, 100)+, 372 (M+°; 100), 287 (15), 255
(100), 242 (70). Anal. Calcd. for C23H20N2O3: C, 74.18; H, 5.41; N,
7.52. Found: C, 74.12; H, 5.38; N, 7.48.
(E)-3-{4-(2-Hydroxyphenyl)-4,5-dihydro-1H-benzo[b][1,4]dia-
zepin-2(3H)-ylidene}-6-methyl-3H-pyran-2,4-dione 4b: Mp 236–
238 °C (239–240 °C [2]), Yield 85%. 1H NMR (300 MHz, CDCl3): d
2.14 (s, 3H, 60-CH3), 3.10 (dd, 1H, H-3, 2J 12.5 Hz, 3J 9.8 Hz), 3.40
(dd, 1H, H-3, 2J 12.5 Hz, 3J 3.5 Hz), 3.85 (s, 1H, H-5), 5.40 (dd, 1H,
3
H-4, J 3.5 and 9.8 Hz), 5.95 (s, 1H, H-50), 6.85–7.40 (m, 4H aro-
matic), 7.45–7.60 (m, 4H, Ar), 16.10 (s br, 1H, NH) ppm. 13C NMR
(75 MHz, CDCl3): d 19.9 (60-CH3), 35.0 (C-3), 66.8 (C-4), 96.6
(C-30), 107.3 (C-50), 116.3 (C-300), 120.6 (C-500), 122.0 (C-6),
122.8 (C-8), 124.5 (C-9), 127.5 (C-10), 127.8 (C-400), 128.0
(C-600), 128.5 (C-7), 130.0 (C-100), 139.6 (C-11), 155.0 (C-200), 163.1
(C-2), 163.7 (C-60), 172.6 (C-20), 184.5 (C-40) ppm. HRMS (ESI+):
m/z calcd. For [C21H18N2O4 + Na]+ 385.3683, Found: 385.3685.
Anal. Calcd. for C23H20N2O4: C, 71.41; H, 5.19; N, 7.21. Found:
C, 71.36; H, 5.20; N, 7.18.
(E,E)-6-Methyl-3-{7-methyl-4-styryl-4,5-dihydro-1H-
benzo[b][1,4]diazepin-2(3H)-ylidene}-3H-pyran-2,4-dione 4c: Mp
1
180–182, Yield 65%. H NMR (300 MHz, CDCl3): d 2.14 (s, 3H, 60-
CH3), 2.31 (s, 3H, 7-CH3), 3.10 (dd, 1H, H-3, 2J 12.4 Hz, 3J 9.8 Hz),
3.82 (s, 1H, H-5), 3.96 (dd, 1H, H-3, 2J 12.4 Hz, 3J 3.8 Hz), 4.87
3
(ddd, 1H, H-4, J 3.8, 9.8 and 12.4 Hz), 5.75 (s, 1H, H-50), 6.33 (dd,
3
1H, H-100, J 7.4 and 15.7 Hz) 6.65 (d, 1H, H-200, 3J 15.7 Hz), 6.65–
7.39 (5H, aromatic), 6.82 (d, 1H, H-9), 7.03 (d, 1H, H-8), 7.72 (s,
1H, H-6), 15.20 (s, 1H, NH) ppm. 13C NMR (75 MHz, CDCl3): d
19.9 (60-CH3), 21.0 (7-CH3), 35.0 (C-3), 66.9 (C-4), 96.5 (C-30),
107.3 (C-50), 122.0 (C-6), 122.7 (C-8), 124.3 (C-9), 126.6 (C-10),
127.8 (C-600), 126.6 (C-400, 800), 128.5 (C-500, 700), 130.0 (C-100), 130.6
(C-200), 136.1 (C-300), 138.6 (C-7), 139.5 (C-11), 163.0 (C-2), 163.7
(C-60), 172.6 (C-20), 184.5 (C-40) ppm. MS: m/z 387 (M + H, 100)+,
386 (M+°; 100), 301 (70), 281 (50), 269 (70), 256 (80). Anal. Calcd.
for C24H22N2O3: C, 74.59; H, 5.74; N, 7.25. Found: C, 74.56; H, 5.70;
N, 7.23.
3. Results and discussion
The NMR data reported for the B form of 1,5-benzodiazepines
confirms that the attached DHAA pyrone ring remains unchanged
[2,6]. Different intramolecular hydrogen bonds (AOHꢁ ꢁ ꢁN@CA or

100
A. Rabahi et al. / Journal of Molecular Structure 1061 (2014) 97–103
Table 1
Results of PM6 total energy (ET) calculations of 1,5-benzodiazepine derivatives 4a,b.
Form B
Form A (E/Z)
Compd.
ET (kJ molꢂ1
)
Compd.
ET (kJ molꢂ1
)
Compd.
ET (kJ molꢂ1
)
B-4a(4S)
B-4a(4R)
B-4b(4S)
B-4b(4R)
ꢂ192.701
ꢂ195.411
ꢂ429.110
ꢂ436.286
(E)-A-4a(4S)
(E)-A-4a(4R)
(Z)-A-4a(4S)
(Z)-A-4a(4R)
ꢂ197.047
ꢂ191.260
ꢂ188.935
ꢂ182.802
(E)-A-4b(4S)
(E)-A-4b(4R)
(Z)-A-4b(4S)
(Z)-A-4b(4R)
ꢂ432.904
ꢂ405.994
ꢂ424.371
ꢂ421.372
Bold values correspond to the most stable chemical entity in the group.
Fig. 1. Schematic representation of the molecular units present in the crystal structures of compounds 4a and 4b. Non-hydrogen atoms are represented as thermal ellipsoids
drawn at the 50% probability level; hydrogen atoms are depicted as small spheres with arbitrary radii. For geometrical details on the depicted intramolecular hydrogen bonds
see Tables S2 and S3 in the Supporting Information.
feature is associated with the strong intramolecular hydrogen
bond established between the hydrogen atom H1 (carried by the
N1 nitrogen) and the respective oxygen atom of the carbonyl in
the pyrone ring. In this way, the formation of the (E)-isomer is pre-
ferred for the two compounds 4a and 4b. Thus, the form (E)-A has
approximately reproduced the results obtained by the PM6 semi-
empirical calculations of ET for the 4a possible stereoisomers
(Fig. 1, Table 1). Nevertheless, both the (E)-A and B forms present
a quasi similar energy in the case of compound 4b. The crystallo-
graphic studies clarify the formation of chelating hexacyclic sys-
tem constructed by N1AH1ꢁ ꢁ ꢁO2@C3AC2AC7 atoms which is
almost planar: the highest deviation from the least squares plane
passing through the six atoms is 0.055(17) Å for 4a and
ꢂ0.06(20) Å for 4b. The medium planes of the pyrone ring and this
hexacyclic system are almost parallel subtending dihedral angles
of 5.3(6)° and 6.5(8)° for compounds 4a and 4b, respectively (Sup-
porting Information).
mainly based on 2D NMR (NOESY, HSQC and HMBC) experiments.
The position of the methyl group on 4c depends on which of the
amino groups in p-Me-o-PDA participates in the nucleophilic at-
tack on 1a (Scheme 1). Analysis of the NOESY spectrum of 4c shows
the sequential NOE cross peaks of H-4/5-NH, 5-NH/H-6 and H-6/7–
CH3 (Fig. 2). In this way, alongside with the HMBC correlations, the
position of the methyl group was unambiguously confirmed
(Fig. 2).
The structural similarity of compounds 4a–c is supported by 13
C
NMR (Table 2). A careful analysis of the 13C NMR chemical shifts (d)
of the enaminopyran-2,4-diones A scaffold in 4a–c would be infor-
mative in order to define whether or not the imino-4-hydroxypy-
ran-2-one B exists in solution. We have noticed a great similarity
with the enaminopyran-2,4-dione compound 7 synthesized by
Aït-Baziz et al. [11] from which the convincing connectivities
found in the HMBC revealed the amino-en-one sequence with an
intramolecular hydrogen (C@Oꢁ ꢁ ꢁHANA) (Fig. 3). Besides, it is
important to underline that the chemical shifts of carbons C-30,
C-40 (pyrone ring) and @CANH in 7 evidence a ‘‘Push–Pull’’ effect
in the 13C NMR spectra of this fragment. In a previous work [5],
the same authors have also described the synthesis and 13C NMR data
of 5-(2-hydrazino-6-phenyl-3E,5E-hexadien)-3-methyl-1H-pyrazole 8
Compounds 4a and 4b crystallize in the centrosymmetric
monoclinic and triclinic crystallographic systems (space groups
P21/n and P-1), respectively, in which the presence of both (E)-A-
4a-b 4S and 4R enantiomers is imposed and both could be distin-
guished in the unit cell (Supporting Information). According to ET
theoretical calculations, both (E)-A-4a 4S and 4R enantiomers pres-
H2
ent also
a
quite similar value of energy (ꢂ197.047 and
ꢂ191.260 kJ molꢂ1, respectively) as molecular models, thereby,
both forms are energetically favored. However, in the case of (E)-
A-4b the 4S enantiomer template shows a greater stability than
C
H2C H
HMBC
H
8
7
NOESY
9
H
7
6
H
6
11
H
H
the 4R one (ꢂ432.904 and ꢂ405.994 kJ molꢂ1
,
respectively),
O
1
N
O
N1
5 N
H
NH
4
3'
although both are experimentally achievable (Table 1).
4'
4'
3'
3
Compound 4c was prepared from the reaction of 1a and p-Me-
o-PDA using the conventional method (refluxing toluene). One
product was separated just as for 4a, but all attempts to obtain sin-
gle crystals of 4c have failed, with the structural analysis being
H
H
H3C
O
O
2'
6'
C
H
O
O
H2
Fig. 2. Main HMBC correlations and NOE cross effects in compound 4c.

A. Rabahi et al. / Journal of Molecular Structure 1061 (2014) 97–103
101
C-10 (extra-cyclic carbonyl) is the most suitable to undergo a nucle-
ophilic attack by amino groups of o-PDAs (Scheme 2). The starting
material 1a is similar to an a,b-unsaturated ketone system capable
Table 2
13C NMR data for 1,5-benzodiazepines 4a–c.
Atom
13C NMR Chemical shift (d, ppm)
to undergo two different nucleophilic attacks (Routes I or II,
Scheme 2) of o-PDAs in the initial step of the reaction: I) conden-
sation of amine on the most electrophilic site C-10 carbonyl to give
the imine intermediate (1,2-addition) Ia ? Ib; or II) 1,4-conjugate
addition (aza-Michael addition) leading to b-amino ketone inter-
mediate IIa ? IIb. Fig. 4 summarizes the calculated total energy
profile for pathway (a) (Scheme 1) using the semi-empirical meth-
od PM6. The total energy (ET) calculated for the intermediates Ia/Ib,
IIa/IIb shows that IIa/IIb are much more stable (ꢂ443.185 and
ꢂ432.317 kJ molꢂ1), than the final product 4c (ꢂ191.047 kJ molꢂ1).
From these data, the 1,4-conjugate addition leading to the b-amino
ketone intermediate (IIa ? IIb) is less probable because the energy
barrier of the IIa/IIb ? 4c transition is high and the reaction is
4a
4b
4c
C-20
C-30
C-40
C-50
C-60
60-CH3
C-2
173.0
96.7
184.7
107.4
163.1
19.9
163.2
34.9
67.2
172.8
96.6
184.6
107.3
163.0
19.8
163.3
35.1
64.7
172.6
96.5
184.5
107.4
163.0
19.9
163.7
35.0
66.8
C-3
C-4
and 1-phenyl-2-pyrazoline 9 through which the carbons’ chemical
shifts due to the imino function C@N are similar (149.1 ppm in 8)
and (150.6 ppm in 9). However, the chemical shifts of the carbonyl
4-C@O in the enaminopyronic scaffolds 4a–c and 7 (184.7, 184.6,
184.5 and 185.7 ppm, respectively) and 4-CAOH (170.5 ppm) in
the imino-4-hydroxypyran-2-one 9 differ in about 15 ppm offering
a possibility to tentatively distinguish between the A and B forms
with the aid of solution 13C NMR. Aside, Claramunt et al. [3] have
reported the synthesis and isolation of 1,5-benzodiazepine 10 sug-
gesting its occurrence in the form A identically to our 1,5-benzodi-
azepines 4a–c herein reported. It is worth noting that the chemical
shifts for the carbonyl 4-C@O in the Push–Pull system for 4a–c and
10 (184.7, 184.6, 184.5 and 184.5 ppm, respectively) are also sim-
ilar like those arising in compound 7 (185.7 ppm), Furthermore,
from the 13C NMR data of benzodiazepine 11 synthesized by Fodili
et al. [6], it can be deduced that the chemical shifts for C-30, C-40
and C-2 of the proposed imino-4-hydroxypyran-2-one scaffold B
seem quasi equal to those of enaminopyran-2,4-diones A in our
models 4a–c and in the reported benzodiazepine 10 depicted in
Fig. 3. With this information in hand, we point out that the
prototropy between A and B forms is still not observed in solution
or even though to be deduced. Thus, if such a phenomenon exists
in equilibrium, it should greatly favor the form A according to
the results obtained from theoretical chemistry calculation of
ET and the solid state X-ray analysis.
thermodynamically unexpected
(DET > 0). The mechanistic
hypothesis following the formation of the imine intermediate
(1,2-addition) Ia ? Ib (ꢂ177.329 and ꢂ186.459 kJ molꢂ1) is, never-
theless, the most plausible and the relevant transition states are
determined to be thermodynamically possible (
b ? 4c). In this respect, similar mechanisms have been proposed
concerning the action of hydrazines (RNHANH2) on ,b-unsatu-
DET < 0 for Ia/
a
rated ketone systems like chalcones which afford, in most cases,
an isolated hydrazone intermediate (imine formation via 1,2-addi-
tion) followed by 2-pyrazoline cyclisation through condensation of
the second amino group on the b-carbon (1,2-addition) [13a–e].
Alternatively, this reaction could also proceed via an aza-1,4-con-
jugate addition of hydrazine on the a,b-unsaturated ketone system
(alkylation of hydrazine). The remaining amino group can then be
condensed on the carbonyl group to form an imine function, hence
2-pyrazoline [13f]. These chalcone ? 2-pyrazoline transformations
are usually non-regioselective and are strongly dependent on the
hydrazine R substituent which controls the first attack of the ami-
no group to yield 1,3,5-triarylpyrazolines [13g].
The mechanism for the formation of compound 4b (R =
o-hydroxyphenyl) through pathway (b) (Scheme 1) was proposed
by Fodili et al. [6] according to route I depicted in Scheme 3. This
mechanism is not the sole possibility and remains questionable
especially regarding the site of protonation (using TFA) that
remains illusive. As a matter of fact, the protonation sites in 3
are numerous. Theoretical studies of the electronic density distri-
bution may indicate the privileged site susceptible to the proton-
ation event. For instance, these calculations were performed for
prototropic forms 3A and 3B with an o-hydroxyphenyl substituent
(R = o-hydroxyphenyl) (Fig. 5). The calculated charge values clearly
The study of the reaction mechanism for the formation of 4a
and 4c (R = styryl) via pathway (a) (Scheme 1) starts by a prior
determination of the electrophilic sites in the DHAA conjugate
reagent 1a that are most susceptible for the first attack of the
amino group. The theoretical calculation of the electronic density
distribution shows that in the starting material 1a, the carbon
H2N
H
H
150.6
N
NH2
185.71 O
O
N
O
N
170.5 OH
N
3'
N
N
179.82
Cl
Cl
4
3
94.8
96.62
149.1
O
O
8
9
7
H
184.5
O
4'
N
OH
185.1
N
2
163.5
96.6
163.3
N
H
N
H
2
4'
3'
3'
95.6
O
O
O
O
11
10
Fig. 3. The ‘‘Push–Pull’’ effect in the 13C chemical shifts showing differences between the enaminopyran-2,4-diones A and the imino-4-hydroxypyran-2-one B of some
reported scaffolds 7–11.

102
A. Rabahi et al. / Journal of Molecular Structure 1061 (2014) 97–103
H2N
HN
OH
O
+ 0.050 - 0.103
+ 0.663
(II)
OH
O
O
OH
O
N
O
1'
5'
3'
(I)
4
NH2
+ 0.670
6
1a
2
O
O
+
Ia
Ib
IIa
p
-Me-
o
-PDA
O
H2N
OH HN
O
O
O
O
HN
O
HN
O
O
O
NH
NH2
IIb
O
4c
Scheme 2. Proposed mechanism of 1,5-benzodiazepine 4c formation through pathway (a).
ET (kJ mol-1)
1a
+
Ia
Ib
- 177.329
p-Me-o-PDA
4c
- 186.459
- 191.047
IIb
IIa
- 432.317
- 443.185
Reaction coordinate
Fig. 4. Total energy profile for pathway (a) for the formation mechanism of 1,5-benzodiazepine 4c.
H
H
H
O
O
N
O
O
O
N
O
OH
O
N
O
N
CHR
N+ CHR
H
N+
H
H+
CHR
CH2
H
CH2
(I)
(I)
3
(I)
(II) H+
H
H
H
O+ HN
O
O
N
OH
N
O
H
R
N
CHR
(II)
N
C
N
CHR
CH2
H
CH2
(II)
H
O
O
O
O
4b
Scheme 3. Proposed mechanism for the formation of 1,5-benzodiazepine 4b through pathway (b) depicted in Scheme 1.
show that the oxygen of the carbonyl 4-C@O in the form 3A, or that
of the hydroxyl 4-OH in the form 3B are most expected to undergo
the protonation under acidic conditions. This data contradicts the
proposed mechanism (I) where the protonation of the amine
groups is suggested (Scheme 3, route I). According to our results,
and some of recent literature [10,11], the intermediary compounds
3 would take the 3A or 3B forms, in which the mechanism follow-
ing route II is reasonably acceptable (Scheme 3, route II). For the
formation of 1,5-benzodiazipines, different mechanisms could be
envisaged to obtain the (E/Z)-A and B forms. However, the results
of the present work have guided us to a unique (E)-A form which
-0.619
-0.593
-0.358
-0.364
OH
N
O
HN
N-0.356
H
N-0.338
H
CH3
CH3
-0.472
-0.492
OH
OH
O
O
O
O
-0.431
-0.493
-0.472
-0.455
3A
3B
Fig. 5. Electronic density distribution on heteroatoms of the intermediary com-
pounds 3A and 3B.

A. Rabahi et al. / Journal of Molecular Structure 1061 (2014) 97–103
103
has been convincely evidenced in the solid state using X-ray
References
diffractometry and in solution by NMR.
[1] (a) C. Page, C. Michael, M. Sutter, M. Walker, B.B. Hoffman, Integrated
Pharmacology, second ed., C.V. Mosby, 2002;
4. Conclusions
(b) D.G. Dikeos, C.G. Theleritis, C.R. Soldatos (Eds.), Benzodiazepines: Effects
on Sleep in Sleep Disorders: Diagnosis and Therapeutics, Informa Healthcare,
London, 2008;
In summary, distinct experiments have allowed us to assert that
the most advantageous methods for the synthesis of 1,5-benzodi-
azepines are the coupling of dehydroacetic acid or conjugate deriv-
atives with ortho-phenylenediamines. We have discussed the
difficulties of characterizing the proposed A and B prototropic
forms of 1,5-benzodiazepines which are the main subject of the
present manuscript. We have also noted that in the literature both
A and B forms are tentatively characterized based on 2D NMR data,
but the mechanisms of their formation are insufficiently explored.
With the aid of theoretical chemistry calculations, comprehensive
exploitation of NMR data and especially the valuable contribution
of single-crystal X-ray diffraction, we could confirm the formation
of 1,5-benzodiazepines A in their (E)-configuration. These studies
permitted the establishment of reasonable mechanisms of forma-
tion. The phenomenon of prototropy between A and B forms is still
not observed in solution. This motivated us to revise the NMR data
reported in the literature for 1,5-benzodiazepines and hopefully
will encourage future analysis in different solvent systems.
(c) H. Ashton (Ed.), Benzodiazepine Dependence in Adverse Syndromes and
Psychiatric Drugs: A Clinical Guide, Oxford University Press, 2004;
(d) A. McIntosh, D. Semple, R. Smyth, J. Burns, R. Darjee, Depressants, first ed.,
Oxford Handbook of Psychiatry, Oxford University Press, 2005;
(e) K.T. Olkkola, J. Ahonen, Handb. Exp. Pharmacol. 182 (2008) 335–360;
(f) T. Saïas, T. Gallarda, Encephale 34 (2008) 330–336;
(g) M. Lader, Expert. Rev. Neurother. 8 (2008) 1189–1191;
(g) M. Lader, A. Tylee, J. Donoghue, CNS Drugs 23 (2009) 19–34;
(h) C.H. Ashton, Curr. Opin. Psychiatry 18 (2005) 249–255;
(i) A.D. Fraser, Ther. Drug Monit. 20 (1998) 481–489.
[2] O. Prakash, A. Kumar, A. Sadana, S.P. Singh, Synth. Commun. 32 (2002) 2663–
2667.
[3] R.M. Claramunt, D. Sanz, S. Aggarwal, A. Kumar, O. Prakash, S.P. Singh, J.
Elguero, Arkivoc xiv (2006) 35–45.
[4] N. Aît-Baziz, Y. Rachedi, M. Hamdi, A.M.S. Silva, F. Belagroune, R. Thiery, N.
Sellier, J. Heterocycl. Chem. 41 (2004) 587–591.
[5] N. Aît-Baziz, Y. Rachedi, F. Chemat, M. Hamdi, Asian J. Chem. 20 (2008) 2610–
2622.
[6] M. Fodili, M. Amari, B. Nedjar-Kolli, A. Robert, M. Baudy-Floch, P. Le Grel,
Synthesis 5 (1999) 811–814.
[7] D. Chergui, M. Hamdi, M. Babouléne, V. Spéziale, A. Lattes, J. Heterocycl. Chem.
23 (1986) 1721–1724.
[8] M. El-Abbassi, B. Djerrari, E.M. Essassi, J. Fifani, Tetrahedron Lett. 30 (1989)
7067–7070.
[9] R. Kaoua, N. Benamane, S. Bakhta, S. Benadji, C. Rabia, B. Nedjar-Kolli,
Molecules 16 (2011) 92–99.
Acknowledgements
[10] L.C. Dias, A.J. Demuner, V.M.M. Valente, L.C.A. Barbosa, F.T. Martins, A.C.
Doriguetto, J. Ellena, J. Agric. Food Chem. 57 (2009) 1399–1405.
[11] N. Aït-Baziz, Y. Rachedi, A.M.S. Silva, Arkivoc x (2010) 86–97.
[12] R.D. Dennington, T.A. Keith, J.M. Millam, Gaaussian 09W and GausView 5.0.8.
Semichem, 2000–2008.
[13] (a) E.M. Sharshira, N.M.M. Hamada, Molecules 17 (2012) 4962–4971;
(b) R. Aggarwal, R. Kumar, Synth. Commun. 39 (2009) 2169–2177;
(c) B. Maleki, D. Azarifar, M.K. Moghaddam, S.F. Hojati, M. Gholizadeh, H.
Salehabadi, J. Serb. Chem. Soc. 74 (2009) 1371–1376;
Thanks are due to the University of Aveiro, Fundação para a
Ciência e a Tecnologia (FCT, Portugal), the European Union, QREN,
FEDER, COMPETE, for funding the Organic Chemistry Research
Unit (QOPNA) (project PEst-C/QUI/UI0062/2013), the Associated
Laboratory CICECO (Pest C-CTM/LA0011/2013), and the Portuguese
National NMR Network (RNRMN).
(d) V.G. Desai, P.C. Satardekar, S. Polo, K. Dhumaskar, Synth. Commun. 42
(2012) 836–842;
(e) L. Ji-Tai, Z. Xiao-Hui, L. Zhi-Ping, Beilstein J. Org. Chem. 3 (2007) 13–17;
(f) O. Mahé, D. Frath, I. Dez, F. Marsais, V. Levachera, J.-F. Brière, Org. Biomol.
Chem. 7 (2009) 3648–3651;
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molstruc.2013.
12.082.
(g) K.S. Smuilovich, N.A. Orlova, E.V. Karpova, M.M. Shakirov, V.V.
Shelkovnikov, Russ. Chem. Bull. Int. Ed. 59 (2010) 1408–1413.