Antimicrobial ferrocene containing quinolinones: Synthesis, spectral, electrochemical and structural characterization of 2-ferrocenyl-2,3- dihydroquinolin-4(1H)-one and its 6-chloro and 6-bromo derivatives

Article abstract of DOI:10.1016/j.poly.2011.11.006

Syntheses of three new ferrocene containing heterocycles - 2-ferrocenyl-2,3-dihydroquinolin-4(1H)-one and its 6-chloro and 6-bromo derivatives - starting from 2-aminoacetophenones and ferrocenecarboxaldehyde was achieved in two steps. The aldol condensati

Full text of DOI:10.1016/j.poly.2011.11.006

Polyhedron 31 (2012) 789–795
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Antimicrobial ferrocene containing quinolinones: Synthesis, spectral,
electrochemical and structural characterization of 2-ferrocenyl-2,
3-dihydroquinolin-4(1H)-one and its 6-chloro and 6-bromo derivatives
a
a
a
b
c
´
´
´
´
´
´
Anka Pejovic , Ivan Damljanovic , Dragana Stevanovic , Mirjana Vukicevic , Sladjana B. Novakovic ,
c
d
a,
⇑
´
´
´
´
Goran A. Bogdanovic , Niko Radulovic , Rastko D. Vukicevic
a
´
Department of Chemistry, Faculty of Science, University of Kragujevac, R. Domanovica 12, 34000 Kragujevac, Serbia
b Department of Pharmacy, Faculty of Medicine, University of Kragujevac, S. Markovi´ca 69, 34000 Kragujevac, Serbia
c
ˇ
Vinca Institute of Nuclear Sciences, Laboratory of Theoretical Physics and Condensed Matter Physics, PO Box 522, 11001 Belgrade, Serbia
d Department of Chemistry, Faculty of Science and Mathematics, University of Niš, Višegradska 33, 18000 Niš, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history:
Syntheses of three new ferrocene containing heterocycles – 2-ferrocenyl-2,3-dihydroquinolin-4(1H)-one
and its 6-chloro and 6-bromo derivatives – starting from 2-aminoacetophenones and ferrocenecarboxal-
dehyde was achieved in two steps. The aldol condensation of these substrates gave the corresponding 20-
aminochalcones in the first stage, whereas a further intramolecular cyclization gives the final products.
This cyclization was performed by either a solvent-free microwave irradiation (500 W/5 min) of a mix-
ture of chalcones and mortmorillonite K-10 or by using an acidic catalyst (the mixture of acetic and
orthophosphoric acid). The latter method, which can be performed by simple stirring at room tempera-
ture or by irradiation in an ultrasonic bath, gave much better results. The obtained compounds were spec-
trally and electrochemically (cyclic voltammetry) fully characterized, as well as by single-crystal X-ray
analysis. A microdilution assay revealed that the three dihydroquinolinones can be regarded as potential
lead compounds in the discovery of new antimicrobial drugs due to their very strong and unselective
Received 27 September 2011
Accepted 7 November 2011
Available online 15 November 2011
Keywords:
Ferrocene
2-Aminoacetophenones
2,3-Dihydroquinolin-4(1H)-ones
Cyclic voltammetry
Crystal structure
Antimicrobial activity
activity towards pathogenic bacteria and one yeast with MIC values (0.01–10.0
of tetracycline.
lg/mL) lower than that
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
SiO2 impregnated with NaHSO4 [16], through the use of microwave
irradiation, SiO2 supported TaBr5 [17] and Yb(OTf)3 [18], SiO2 and
Al2O3 supported CeCl3 [19], ZnCl2 combined with a polymer sup-
ported selenium reagent [20], molecular iodine [21], polyethylene
glycol [22], ionic liquids [23], SbCl3 [24], silica chloride [25], ZnCl2
[26], etc. Although ferrocene containing quinoline derivatives have
already been synthesized (see, for example [27,28]), to the best of
our knowledge there are no reports on the synthesis of neither
2-ferrocenyl-2,3-dihydroquinolin-4(1H)-ones nor 2-ferrocenyl-
quinolin-4(1H)-ones. It is well known that the interchange of an
aromatic ring with the ferrocene nucleus in some organic com-
pounds possessing a certain property (such as biological activity,
for example) might lead to a product with this property markedly
more prominent than that of the parent compound.
With this in mind and given our permanent interest in ferro-
cene chemistry [29–31], herein we report on the synthesis and
spectral, electrochemical and structural characterization of three
new ferrocene containing dihydroquinolinones – 2-ferrocenyl-
2,3-dihydroquinolin-4(1H)-one, 6-chloro- and 6-bromo-2-ferroce-
nyl-2,3-dihydroquinolin-4(1H)-ones. Our intention was also to
investigate the biological activity of this class of compounds and,
Quinoline and its derivatives have many applications, whereas
the use of these compounds as pharmaceutics is surely among
the most important ones. For example, quinolone antimicrobial
agents play the central role in the management of a broad range
of infections, like respiratory and urinary tracts infections, sexually
transmitted diseases, gastrointestinal, abdominal infections, etc.
[1]. 2-Aryl-2,3-dihydroquinolin-4(1H)-ones, in their turn, besides
possessing analgesic activity [2], nowadays attract considerable
attention as antimitotic antitumor agents [3–5]. In addition, they
are interesting as intermediates in the synthesis of other pharma-
ceuticals and active compounds [6,7]. The early procedures for the
synthesis of these aza analogues of flavanones included either base
[8–11] or acid [12,13] catalyzed isomerisation of the corresponding
20-aminochalcones. However, many new catalytic systems for this
intramolecular aza-Michael-type cyclization have been developed
in the recent years, such as montmorillonite K-10 [14,15] and
⇑
Corresponding author. Tel.: +381 34 30 02 68; fax: +381 34 33 50 40.
E-mail address: vuk@kg.ac.rs (R.D. Vukic´evic´).
0277-5387/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2011.11.006

´
A. Pejovic et al. / Polyhedron 31 (2012) 789–795
790
thus, we decided to screen their antimicrobial activity against sev-
eral bacteria and one fungal strain.
2.2.1. 2-Ferrocenyl-2,3-dihydroquinolin-4(1H)-one (4a)
M.p. 150 °C; 1H NMR (200 MHz, CDCl3, ppm) d = 7.86 (dd, J = 7.7,
1.4 Hz, 1H, H5), 7.33 (ddd, J = 7.7, 6.3, 1.4 Hz, 1H, H7), 6.77 (brt,
J ꢀ 7.4 Hz, 1H, H6), 6.72 (brd, J = 7.7 Hz, 1H, H8), 4.65 (brs, 1H,
NH), 4.45 (dd, J = 12.4, 4.6 Hz, 1H, H2), 4.27–4.31 (m, 1H, Fc),
4.19–4.25 (m, 8H, Fc, 5H from the unsubstituted Cp and 3H from
the substituted Cp), 2.87 (ddd, J = 16.2, 4.6, 1.2 Hz, 1H, H3eq),
2.74 (dd, J = 16.2, 12.4 Hz, 1H, H3ax); 13C NMR (200 MHz, CDCl3,
ppm): d = 193.5 (C4), 151.2 (8a), 135.3 (C7), 127.6 (C5), 118.9
(C4a), 118.1 (C6), 115.7 (C8), 89.3 (C10), 68.5 (C100–C500), 68.3, 68.2
(C20, C50), 66.7, 66.1 (C30, C40), 52.9 (C2), 45.9 (C3); IR (KBr):
2. Experimental
2.1. Materials and instruments
All chemicals were commercially available and used as re-
ceived, except that the solvents were purified by distillation.
Microwave Labstation for Synthesis, MicroSynth, Milestone appa-
ratus equipped with pressure and temperature control units was
used for the microwave assisted syntheses. Ultrasonic cleaner
Elmasonic S 10, 30 W was used for the ultrasonically supported
synthesis.
m
= 3323, 3078, 2991, 1651, 1608, 1507, 1480, 1321, 769 cmꢁ1
.
Anal. Calc. for C19H17FeNO: C, 68.90; H, 5.17; N, 4.23. Found: C,
68.87; H, 5.14; N, 4.25%.
Chromatographic separations were carried out using silica gel
60 (Merck, 230–400 mesh ASTM), whereas silica gel 60 on Al
plates, layer thickness 0.2 mm (Merck) was used for TLC. Melting
points (uncorrected) were determined on a Mel-Temp capillary
melting points apparatus, model 1001. Microanalysis of carbon,
hydrogen and nitrogen were carried out with a Carlo Erba 1106
microanalyser; their results agreed favorably with the calculated
values. The 1H and 13C NMR spectra of the samples in CDCl3 were
recorded on a Varian Gemini (200 MHz) spectrometer. Chemical
shifts are expressed in d (ppm), relative to the residual solvent pro-
tons or 13CDCl3 as internal standards (CHCl3: 7.26 ppm for 1H and
77 ppm for 13C). IR measurements were carried out with a Perkin–
Elmer FTIR 31725-X spectrophotometer.
2.2.2. 6-Chloro-2-ferrocenyl-2,3-dihydroquinolin-4(1H)-one (4b)
M.p. 144 °C; 1H NMR (200 MHz, CDCl3, ppm): d = 7.82 (d,
J = 2.5 Hz, 1H, H5), 7.26 (dd, J = 8.5, 2.5 Hz, 1H, H7), 6.66 (d,
J = 8.5 Hz, 1H, H8), 4.66 (brs, 1H, NH), 4.45 (dd, J = 13.5, 4.0 Hz,
1H, H2), 4.25–4.28 (m, 1H, Fc), 4.20–4.24 (m, 8H, Fc, 5H from the
unsubstituted Cp and 3H from the substituted Cp), 2.87 (ddd,
J = 16.5, 4.0, 1.5 Hz, 1H, H3eq), 2.73 (dd, J = 16.5, 13.5 Hz, 1H,
H3ax); 13C NMR (200 MHz, CDCl3, ppm): d = 192.4 (C5), 149.5
(C8a), 135.2 (C7), 126.9 (C5), 123.5 (C6), 119.6 (C4a), 117.3 (C8),
88.9 (C10), 68.6, 68.4 (C20, C50), 68.3 (C100–C500), 66.6, 66.1 (C30,
C40), 52.9 (C2), 45.4 (C3); IR (KBr):
m = 3340, 2924, 1657, 1615,
1501, 1480, 1408, 1294, 816 cmꢁ1. Anal. Calc. for C19H16ClFeNO:
C, 62.41; H, 4.41; N, 3.83. Found: C, 62.37; H, 4.44; N, 3.79%.
2.2. Preparation of quinolinones 4a–c
2.2.3. 6-Bromo-2-ferrocenyl-2,3-dihydroquinolin-4(1H)-one (4c)
M.p. 179 °C; 1H NMR (200 MHz, CDCl3, ppm): d = 7.96 (d,
J = 2.5 Hz, 1H, H5), 7.38 (dd, J = 8.5, 2.5 Hz, 1H, H7), 6.61 (d,
J = 8.5 Hz, 1H, H8), 4.66 (brs, 1H, NH), 4.44 (dd, J = 13.0, 3.5 Hz,
1H, H2), 4.25–4.27 (m, 1H, Fc), 4.20–4.24 (m, 8H, Fc, 5H from the
unsubstituted Cp and 3H from the substituted Cp), 2.87 (ddd,
J = 16.5, 4.0, 1.5 Hz, 1H, H3eq), 2.73 (dd, J = 16.5, 13.0 Hz, 1H,
H3ax); 13C NMR (200 MHz, CDCl3, ppm): d = 192.2 (C4), 149.9
(C8a), 137.8 (C7), 130.0 (C5), 120.1 (C4a), 117.6 (C8), 110.4 (C6),
88.8 (C10), 68.5, 68.4 (C20, C50), 68.3 (C100–C500), 66.6, 66.1 (C30,
The solution of 214 mg (1 mmol) of ferrocenecarboxaldehyde
(1), 1 mmol of the corresponding o-aminoacetophenone (1-(2-ami-
nophenyl)ethanone (2a), 1-(2-amino-5-chlorophenyl)ethanone
(2b) and 1-(2-amino-5-bromophenyl)ethanone (2c)) and 100 mg
of NaOH in 10 mL of ethanol was stirred overnight at room temper-
ature. The solvent was evaporated and to the residue 10 mL H2O
was added. The solution was neutralized with 2 M HCl (litmus pa-
per), extracted with CH2Cl2 (three 20 mL portions) and the com-
bined organic layers dried overnight (anhydrous Na2SO4). The
solvent was evaporated and the residue chromatographed over a
short pad of SiO2 (hexane/ethyl acetate 9:1, v/v). After the evapo-
ration of the solvent, the obtained solids (chalcones 3a–c) were
submitted to an intramolecular cyclization reaction by applying
one of the following three methods:
C40), 52.7 (C2), 45.3 (C3); IR (KBr):
m = 3327, 2924, 1657, 1600,
1494, 1394, 1284, 820 cmꢁ1. Anal. Calc. for C19H16BrFeNO: C,
55.65; H, 3.93; N, 3.42. Found: C, 55.60; H, 3.97; N, 3.41%.
2.3. X-ray crystallography
Method A: Chalcones 3a–c (1 mmol) were mixed with 100 mg of
montmorillonite K-10 in a mortar with a pestle, placed into a Tef-
lon cuvette and irradiated for 5 min in a microwave oven at 500 W,
without the presence of a solvent. After 10 min of cooling down to
room temperature, the crude mixture was extracted with ethyl
acetate and the obtained solution dried overnight (anhydrous
Na2SO4). The solvent was evaporated and the residue purified by
column chromatography (SiO2, hexane/ethyl acetate 9:1, v/v).
Method B: Chalcones 3a–c (1 mmol) were dissolved in the mix-
ture of 3 mL glacial acetic acid and 3 mL of 90% orthophosphoric
acid and stirred at room temperature for 50 min. The reaction mix-
ture was poured into ice–water mixture, extracted with ethyl ace-
tate (three 25 mL portions), the obtained solution washed with
NaHCO3, and dried overnight over anhydrous Na2SO4. After the sol-
vent was evaporated, the residue was purified by column chroma-
tography (SiO2, hexane/ethyl acetate 9:1, v/v).
Single-crystal X-ray analysis of three ferrocene derivatives:
2-ferrocenyl-2,3-dihydroquinolin-4(1H)-one (4a), 6-chloro-2-ferr-
ocenyl-2,3-dihydroquinolin-4(1H)-one (4b) and 6-bromo-2-ferr-
ocenyl-2,3-dihydroquinolin-4(1H)-one (4c) was performed. X-ray
diffraction data for all three compounds were collected at room
temperature and using two single-crystal diffractometers, Enraf–
Nonius CAD4 (for 4a) and Oxford Diffraction Xcalibur Sapphire3
Gemini (for 4b and 4c). Both diffractometers were equipped with
Mo Ka radiation (k = 0.71073 Å). X-ray data 4b and 4c were
processed with CrysAlis software [33] with multi-scan absorption
corrections applied using SCALE3 ABSPACK [33].
All three crystal structures were solved with SHELXS [34] and re-
fined using SHELXL [34]. The H1n atom attached to N1 was located
by difference Fourier synthesis and refined isotropically. All other
H atoms were placed at geometrically calculated positions with
the C–H distances fixed to 0.93 from C(sp2); 0.97 and 0.98 Å from
methylene and methine C(sp3), respectively. The corresponding
isotropic displacement parameters of the hydrogen atoms were
equal to 1.2 Ueq and 1.5 Ueq of the parent C(sp2) and C(sp3),
respectively.
Method C: A round bottom flask containing the solution of chal-
cones 2a–c, 1 mmol in the above mentioned mixture of acetic and
phosphoric acids (around 6 mL), was placed in an ultrasonic clea-
ner and irradiated for 50 min. The reaction mixture was worked
up as given in Method B.

´
791
A. Pejovic et al. / Polyhedron 31 (2012) 789–795
Table 1
Crystallographic data for 4a, 4b and 4c.
Compound
4a
4b
4c
Empirical formula
Formula weight
Color, crystal shape
Crystal size (mm3)
T (K)
C19 H17 Fe N O
331.19
orange, prism
0.27 ꢂ 0.24 ꢂ 0.23
293(2)
C19 H16 Cl Fe N O
365.63
orange, prism
0.25 ꢂ 0.22 ꢂ 0.19
293(2)
C19 H16 Br Fe N O
410.09
orange, prism
0.21 ꢂ 0.18 ꢂ 0.18
293(2)
k (Å)
0.71073
0.71073
0.71073
Crystal system
Space group
Unit cell dimensions
a (Å)
b (Å)
c (Å)
monoclinic
P21/c
monoclinic
P21/c
monoclinic
P21/c
12.991(4)
8.482(2)
13.972(3)
90
13.7845(6)
8.0608(3)
13.9697(4)
90
14.1005(7)
7.9990(3)
14.0146(5)
90
a
(°)
b (°)
98.80(2)
90
1521.4(7)
4
1.446
0.991
94.997(3)
90
1546.33(10)
4
1.571
1.151
96.104(4)
90
1571.74(11)
4
1.733
3.505
c
(°)
V (Å3)
Z
Dcalc (mg/m3)
l
(mmꢁ1
)
F(000)
688
752
824
h range for data collection (°)
Reflections collected
1.59–25.98
3109
2.97–29.11
8320
3.21–29.03
7835
Independent reflections (Rint
Data/restraints/parameters
Goodness-of-fit (GOF) on F2
)
2977 (0.0210)
2977/0/203
1.094
3573 (0.0290)
3573/0/212
1.098
3622 (0.0334)
3622/0/212
1.132
Final R1/wR2 indices [I > 2
r
(I)]
0.0733/0.2334
0.0475/0.0906
0.0562/0.1054
A summary of the crystallographic data is given in Table 1. Fig-
ures were produced using ORTEP-3 [35] and MERCURY, Version 2.4 [36].
The software used for the preparation of the materials for presen-
tation: WINGX [37], PLATON [38], PARST [39].
racycline and nystatin served as positive controls, while the sol-
vent (10% DMSO(aq)) was used as a negative control.
3. Results and discussion
3.1. Synthesis
2.4. Electrochemistry
Synthesis of the title compounds followed the concept of an
intramolecular aza Michael addition, mentioned in the introduc-
tory section. In the first step of this protocol ferrocenecarboxalde-
hyde (1) was subjected to a simple mixed aldol condensation with
three o-aminoacetophenones 2a–c, followed by an isomerisation of
the obtained chalcone analogues 3a–c to quinolines 4a–c during
the second stage (Scheme 1). The preparation of the derivatives
3a–c (60–90%) was accomplished by stirring the ethanol solution
of reactants and NaOH overnight. Prompted by a recent report on
the easy, environmentally benign and efficient synthesis of 2-
aryl-2,3-dihydroquinolin-4(1H)-one from 20-aminochalcones un-
der mild reaction conditions (70–80%) [14], we adsorbed com-
pounds 3a–c on the montmorillonite K-10 clay surface and
Cyclic voltammetry experiments were performed at room tem-
perature under an argon atmosphere in a three-electrode cell using
an Autolab potentiostat (PGSTAT 302N). The working electrode
was a platinum disk (2 mm diameter). The counter electrode was
a platinum wire, and an Ag/AgCl electrode was used as the refer-
ence one.
2.5. Biology
2.5.1. Test microorganisms
The synthesized dihydroquinoliones 4a–c were tested against a
panel of microorganisms (American Type Culture Collection
strains), including Gram-positive (Bacillus cereus ATCC 10876,
Staphylococcus aureus ATCC 6538 and Clostridium perfringens ATCC
19404) and Gram-negative bacteria (Escherichia coli ATCC 25922,
Klebsiella pneumoniae ATCC 10031 Proteus vulgaris ATCC 8427,
Pseudomonas aeruginosa ATCC 27853 and Salmonella enterica ATCC
13076), as well as against a yeast Candida albicans ATCC 10231.
Bacterial strains were maintained on Nutrient agar (37 °C), while
the pathogenic yeast was cultured (30 °C) on potato dextrose agar
(PDA).
H2N
X
H2N
+
X
NaOH
O
Fe
Fe
O
O
1
2a-c
3a-c
a)X=H; b)X=Cl, c)X=Br
8
7
5
H
N
K-10,MW
8a
4
6
2.5.2. Screening of antimicrobial activity
X
4'
5'
or
Antimicrobial activity was evaluated using a broth microdilu-
tion method according to NCCLS [40]. Minimum inhibitory concen-
trations (MIC) and minimal bactericidal/fungicidal concentrations
(MBC/MFC) were determined as described in full detail in Radul-
3'
AcOH/H3PO4r.t.
or
1'
2
3
4a
2'
3a-c
Fe
4''
1''
2''
AcOH/H3PO4,r.t.,))))))
O
3''
4a-c
´
ovic et al. [41]. Stock solutions of the compounds 4a–c were pre-
pared in 10% (v/v) aqueous dimethyl sulfoxide (DMSO) in the
concentration range 0.01–10000 lg/mL (the diluting factor 2). Tet-
Scheme 1. Synthesis and numbering scheme of 2-ferrocenyl-2,3-dihydroquinolin-
4(1H)-ones 4a–c.

´
A. Pejovic et al. / Polyhedron 31 (2012) 789–795
792
Table 2
nyl-2,3-dihydroquinolin-4(1H)-one given in the paper by Lee et al.
Yields of the synthesized quinolinones 4a–c.
[43] show a deviation from all the other published ones (e.g.
[42,44]) and from the ones obtained in this study for the ferrocenyl
analogue 4a.
o-
Quinolinone Yielda (%)
Aminoacetophenone
Method
Ab
Method
Bb
Method
Cb
The conformation of the dihydroquinolinone ring of compounds
4a–c could be inferred from their 1NMR. The large value of the cou-
pling of H2 with one of the C3 protons (12.4, 13.5 and 13.0 Hz, for
4a–c, respectively) is only compatible with a diaxial antiperiplanar
arrangement of H2 and H3ax and this clearly establishes the con-
formation of the dihydroquinolinone ring as a half-chair with the
2-ferrocenyl substituent in pseudoequatorial position. A conspicu-
ous feature of the 1H NMR spectra of 4a–c was an additional small
splitting of the signal of H3eq (1.2, 1.5 and 1.5 Hz, for 4a–c, respec-
tively). This splitting can arise from a four-bond NH–H3eq interac-
tion. Due to the known stereospecificity of the four-bond couplings
this provides additional evidence for the quasi perfect tetrahedral
geometry of N, C2 and C3 atoms, and, hence, for the half-chair con-
formation of the ring. Observation of the NH–H3eq long-range
splitting implies a slow NH-exchange on the NMR time scale
(>1 sꢁ1, estimated from the magnitude of the residual NH–CH cou-
pling constant). The expected three-bond coupling between NH
and H2, on the other hand, causes only a broadening of the H2 mul-
tiplet lines when measured in CDCl3-solution.
2a
2b
2c
4a
4b
4c
14
13
11
60
41
36
70
73
74
a
Isolated yield based on the starting ferrocenecarboxaldehyde.
See Section 2.
b
submitted it to microwave irradiation. However, in our hands, this
procedure gave only very poor yields (11–14%; Table 2, Method A)
of the target 2-ferrocenyl-2,3-dihydroquinolin-4(1H)-ones 4a–c.
Our attempts to improve the yields by varying reaction conditions
(increase of the amount of catalyst and the time of exposure to
MW) did not produce better results.
We decided, then, to perform this synthesis through the use of
orthophosphoric acid as the catalyst [12,13]. Again when the origi-
nal reaction conditions were adopted (heating at reflux) [12], very
poor yields of the quinolinones 4a–c were attained. But when the
reaction was conducted at room temperature, compounds 4a–c
were obtained in much higher yields – up to 60% (Method B, Table
2). The best yields, however, were arrived at if the reaction was
carried out with orthophosphoric acid but also promoted by ultra-
sonic irradiation (70–74%; Table 2, Method C).
3.3. Structural comparisons of 4a, 4b and 4c
All three compounds have very similar molecular geometry
(Fig. 1 and Fig. S1). The ferrocene unit in 4b and 4c display a con-
formation close to eclipsed (Fig. S1). The C1–Cg1–Cg2–C10 torsion
angle is ꢁ7.6° in 4b and ꢁ7.9° in 4c (Cg1 and Cg2 are centroids of
the corresponding Cp rings). In 1, the cyclopentadienyl (Cp) rings
are somewhat more eclipsed and the C1–Cg1–Cg2–C10 torsion an-
gle is ꢁ3.3°. In all three compounds the Cp rings within ferrocenyl
unit are almost parallel with interplanar angles 2.6(6)°, 3.4(2)° and
3.7(4)° for 4a, 4b and 4c, respectively. The Cg1–Cg2 distance
(3.29 Å) and the Cg1–Fe–Cg2 angle (178.5°, 177.4° and 178.5°)
are also very similar for all crystal structures.
Bond lengths (Table 3) show that only C–C and C–N pure single
bonds exist in the N1–C11–C12–C13 fragment for all three com-
pounds. Corresponding bond lengths for all molecules are very
similar except to some extent for the C1–C5 bond.
Conformation and geometrical parameters of the N1–C11 six-
membered ring is very similar for all three compounds. This ring
is twisted around the C11–C12 bond. Bond angles within the ring
vary from 108.0(5)° to 121.2(6)° but all corresponding angles for
all three compounds are comparable (Table 4). Comparison of
some torsion angles within the ring are given in Table 4 for the
three compounds.
A closer analysis of the unit cell dimensions reveals some inter-
esting points. Namely, all three compounds crystallize in the
monoclinic system and the same space group, P21/c. The unit cell
dimensions are similar and, as expected, due to the presence of
bromine, the unit cell volume for 3 is the largest (1521.4(7),
1546.33(10) and 1571.74(11) Å3 for 4a, 4b and 4c, respectively).
However, surprisingly, the length of the unit cell edge c is almost
equal for all three crystal structures (Table 1). A ready explanation
can be found in the existence of the N1–Hꢃ ꢃ ꢃO1 hydrogen bond.
This H-bond is the only classical hydrogen bond in all three crystal
structures. It forms a chain of molecules exactly along the c axis
keeping the molecules in almost the same orientation and distance
for all three ferrocene derivatives (Fig. S2). The N1ꢃ ꢃ ꢃO1 distance is
3.144(8), 3.098(3) and 3.104(4) Å for 4a, 4b and 4c, respectively
(The N1–Hꢃ ꢃ ꢃO1 angle, ranging from 176(5)° to 179(7)°, is very
close to 180° for all crystal structures).
3.2. Spectral characterisation
The three synthesized compounds (4a–c) were characterized by
spectral (1H and 13C NMR, IR) and chemical means (elemental anal-
ysis), and the accumulated data agreed favourably with the pro-
posed structures. The IR spectra, as expected, displayed
a
carbonyl stretch (1651–1657 cmꢁ1) and N–H vibration (3323–
3340 cmꢁ1). The 1H NMR spectra of 4b and 4c showed a character-
istic doublet at 7.82 and 7.96 ppm, respectively, with the coupling
constant of 2.5 Hz, in both 4b and 4c, that can be ascribed to the C5
proton. A proton under the influence of a carbonyl anisotropy and
with a meta interaction in these spectra clearly confirms the pres-
ence of a chlorine or bromine at the sixth position. The 13C NMR
spectra of the dihydroquinolinones 4a–c exhibit signals in the aro-
matic, carbonyl and aliphatic regions. The assignment of 13C reso-
nances was achieved on the basis of chemical shift theory, signal
intensities, substituent effects and by comparison with literature
data [42] for analogues. The carbonyl carbon (C-4) in these com-
pounds resonates in a narrow range, dC 192.2–193.5 ppm, and is
relatively insensitive to the presence and nature of the halogens
in 4b and 4c. However, the difference in electronegativities of the
halogens (Cl > Br) and the nature of the heavy atom is clearly dem-
onstrated by the larger value of chemical shifts of the directly
bonded carbon C6 (123.5 and 110.4 for 4b and 4c, respectively)
and the reversal of the same for the adjacent carbons C5 (130.0
and 133.1 for 4b and 4c, respectively) and C7 (135.2 and 137.8
for 4b and 4c, respectively), when the two halogenated analogues
are mutually compared. If the parent compound 4a is compared to
an azaflavanone analogue (2-phenyl-2,3-dihydroquinolin-4(1H)-
one) [42], the 13C resonances of the A- and C-ring nuclei of 4a illus-
trate their insensitivity towards the interchange of the phenyl
group for the ferrocenyl one (C2-52.9/58.5, C3-45.9/46.6, C4-
193.6/193.2, C4a-118.9/119.1, C5-127.6/127.6, C6-118.1/118.4,
C7-135.3/135.4, C8-115.7/115.9, C8a-151.2/151.5, for 4a/phenyl
analogue [42], respectively). This once again puts out a pro argu-
ment for the general viewpoint of the effect that a benzene–ferro-
cene substitution produces. Perhaps worth mentioning, the
spectral data, and especially the 13C NMR shifts in CDCl3, for 2-phe-
Besides the N1–Hꢃ ꢃ ꢃO1 hydrogen bond there is no other (even
weak) H-bond in the three crystal structures. However, numerous

´
793
A. Pejovic et al. / Polyhedron 31 (2012) 789–795
Table 4
Selected bond and torsion angles (°).
4a
4b
4c
C19–N1–C11
N1–C11–C10
N1–C11–C12
C10–C11–C12
C13–C12–C11
O1–C13–C14
119.2(5)
108.3(5)
108.0(5)
114.7(6)
112.0(6)
122.8(6)
120.6(6)
116.5(5)
119.3(6)
121.2(6)
47.8(8)
119.4(2)
109.1(2)
108.7(2)
115.8(2)
112.4(2)
123.1(2)
121.4(2)
115.5(2)
119.6(2)
120.7(2)
48.0(3)
119.7(3)
109.7(3)
108.4(4)
115.9(4)
112.6(4)
123.1(4)
121.4(4)
115.4(3)
119.5(4)
121.0(4)
47.8(5)
O1–C13–C12
C14–C13–C12
C19–C14–C13
N1–C19–C14
C19–N1–C11–C12
N1–C11–C12–C13
C11–C12–C13–C14
ꢁ54.6(8)
ꢁ54.7(3)
ꢁ54.3(5)
35.2(9)
34.3(4)
34.5(6)
Table 5
Electrochemical data of compounds 4a–c.
Compound
Eox1 (V)
Eox2 (V)
Ered (V)
E1/2 (V)
DE (mV)
4a
4b
4c
0.458
0.476
0.473
0.1340
0.1373
0.1401
0.360
0.378
0.381
0.409
0.427
0.427
98
98
92
The greatest difference in the packing of molecules for the three
crystal structures is illustrated in Fig. S4. Obviously, two neighbor-
ing molecules are significantly more distant in the case of crystal
structure 4a. (Although the mutual orientation of the two mole-
cules is almost the same for all three crystal structures.) An expla-
nation may be found in the fact that only compounds 4b and 4c
possess halogen substituents (X = Cl and Br for 4b and 4c, respec-
tively) which participate in the C–Hꢃ ꢃ ꢃX intermolecular interac-
tions between two molecules. These interactions although very
weak, are stabilizing and keep the van der Waals surfaces of two
molecules close to each other (Fig. S4). The unoccupied space be-
tween molecules in the crystal structure of 4a, gives a ‘‘Potential
Solvent Accessible Area’’ of 37.2 Å3.
3.4. Electrochemistry
Cyclic voltammetry experiments in acetonitrile containing
0.1 mol/L of lithium perchlorate as the supporting electrolyte have
been performed to evaluate the electrochemical properties of com-
pounds 4a–c. On the basis of some preliminary investigations we
chose the potential window between 0 and 1.5 V. The voltammo-
grams obtained for compound 4a are presented here (Fig. 2) as a
representative example, whereas the data of the other compounds
are summarised in Table 5.
As it can be seen from Fig. 2, curve a, the ferrocene containing
quinolinone 2a exhibited at the first scan two well defined oxida-
tion waves on the forward potential sweep (O1, at 0.458 and O2, at
1.340 V, respectively) and one reduction wave on the back poten-
tial sweep (R1, at 0.592 V). The reduction peak R1 appeared also
when the potential was reversed after O1, and, therefore, O1 and
R1 should correspond to interdependent electrochemical events.
Since the difference between the values of these two potentials is
close to the theoretical one, O1 and R apparently belong to a
reversible redox couple, appearing due to the presence of the ferro-
cene nucleus. Their position lays more than 50 mV higher than that
of the unsubstituted ferrocene (Fig. 2, curve c). This might be
attributed to a slightly raised electro-positivity of the carbon atom
connected to the cyclopentadienyl ring (caused by the negative
inductive effect of the nitrogen atom). Both the anodic (O1) and
cathodic (R1) peak currents are proportional to the square root of
Fig. 1. Molecular structures of 4a (a), 4b (b) and 4c (c) with the atom numbering
scheme. Displacement ellipsoids are drawn at the 50% probability level. Additional
projections are shown in the Supplementary material file (Fig. S1).
Table 3
Selected bond lengths (Å).
4a
4b
4c
O1–C13
N1–C19
N1–C11
C1–C5
1.241(8)
1.375(9)
1.470(8)
1.443(11)
1.503(8)
1.534(8)
1.511(9)
1.460(9)
1.405(9)
1.427(9)
1.368(11)
1.411(11)
1.373(11)
1.410(9)
1.218(3)
1.366(3)
1.455(4)
1.413(4)
1.499(4)
1.499(4)
1.507(4)
1.474(4)
1.394(4)
1.408(3)
1.361(4)
1.391(4)
1.365(4)
1.410(4)
1.217(5)
1.358(5)
1.457(6)
1.413(7)
1.496(6)
1.499(5)
1.513(6)
1.472(6)
1.390(6)
1.410(5)
1.360(6)
1.400(6)
1.352(6)
1.411(6)
C10–C11
C11–C12
C12–C13
C13–C14
C14–C15
C14–C19
C15–C16
C16–C17
C17–C18
C18–C19
intermolecular CH/p interactions exist in the crystal packing of all
three compounds. Some selected CH/p interactions are illustrated
in Fig. S3.

´
A. Pejovic et al. / Polyhedron 31 (2012) 789–795
794
the scan rate (as depicted in Fig. 3), and their ratio is independent
of the scan rate, indicating a diffusion-controlled process.
An irreversible electrochemical event is responsible for the
occurrence of the second oxidation wave (O2) at 1.340 V, since
the corresponding reduction wave interdependent to O2 was not
observed. To the best of our knowledge, there are no literature re-
ports on the cyclic voltammetry of 2,3-dihydroquinolin-4(1H)-one
or its derivatives. One can assume a certain degree of similarity
with the cyclic voltammetry of N-alkyl anilines [45–47], and we
believe that an electron transfer from the nitrogen atom to anode
occurs at this potential. The product of this event should be a rad-
ical cation, which surely undergoes some irreversible transforma-
tions, because of its high reactivity. In the case of the above
mentioned anilines this species gives products that can be detected
by cyclic voltammetry at the second and subsequent scans [45–
47], but here we did not observe any similar behaviour, most prob-
able due to the overlapping with the electrochemical response of
the ferrocene unit. The electrochemical properties of compounds
4a–c are apparently interesting and surely deserve additional
investigations, but both the extent and the type of the future inves-
tigations exceed the scope of this work.
Fig. 2. Cyclic voltammograms of 1 mM solution of 2-ferrocenyl-2,3-dihydroquin-
olin-4(1H)-one at the platinum electrode (2 mm diameter) by a 0.1 Vsꢁ1 scan rate in
a 0.1 M acetonitrile solution of LiClO4: (a) the first scan, (b) the second scan, (c)
1 mM ferrocene and (d) electrolyte.
3.5. Biology
To date, the contribution of ferrocene to antibacterial, anti-
fungal and other biological properties of ferrocene-containing
compounds remains uncertain. Some authors believe that ferro-
cene fulfills a structural role, possibly that of a hydrophobic spacer
in place of a phenyl ring [48]. Others proposed a more ‘‘active’’ role
for ferrocene, namely as a source of ferricenium (Fe3+) or reactive
oxygen species that had a direct influence on activity [49]. In addi-
tion to this, dihydroquinoline moiety is found in a wide variety of
natural products and an extensive array of medicinally interesting
compounds [50,51]. Motivated by these ideas and our previous re-
sults [29–32], we decided to screen compounds 4a–c for their
in vitro antimicrobial activity on a group of representative Gram-
positive, Gram-negative bacteria and one human pathogen fungal
strain. The antimicrobial activities of the substances expressed as
MIC and MBC/MFC are reported in Table 6. In particular, 4a and
its halogen derivatives 4b and 4c showed a very strong and unse-
lective activity against all tested microorganisms with MIC values
ranging from 0.01 to 10.0 lg/mL, but with usually one order or
two of magnitude larger MBC/MFC values. Some of the compounds
proved to be more potent in inhibiting the growth of bacteria than
the used positive control (the commercially available tetracycline).
The bromo-compound 4c, that was the most active of all three, had
MIC in all cases lower that that of tetracycline, although its anti-
Fig. 3. Anodic and cathodic peak currents obtained at different scan rates of 1 mM
solution of 2-ferrocenyl-2,3-dihydroquinolin-4(1H)-one (4a) at the platinum
electrode (2 mm diameter) in a 0.1 M acetonitrile solution of LiClO4.
Table 6
Antimicrobial activity of 4a–c.
Compound
g/mL)
4a
4b
4c
Tetracycline
MIC = MBC
Nystatin
(
l
MIC
MBC
MIC
MBC
MIC
MBC
MIC = MFC
Gram-positive bacteria
B. cereus
S. aureus
0.40
0.39
2.25
20.0
5.00
20.0
0.31
0.10
1.25
10.0
5.00
10.0
0.31
0.07
0.15
10.0
5.00
10.0
1.56
0.09
1.56
nt
nt
nt
C. perfringens
Gram-negative bacteria
E. coli
K. pneumoniae
P. vulgaris
P. aeruginosa
S. enterica
2.25
0.05
0.40
10.0
1.25
20.0
5.00
10.0
40.0
20.0
0.31
0.01
0.40
5.00
0.62
10.00
1.25
5.00
20.0
5.00
0.07
0.01
0.10
1.25
0.15
10.0
2.50
5.00
10.0
20.0
1.56
0.05
0.11
3.12
3.12
nt
nt
nt
nt
nt
Yeast
C. albicans
1.25
10.0
0.45
20.0
0.31
20.0
nt
0.04
nt, not tested.

´
795
A. Pejovic et al. / Polyhedron 31 (2012) 789–795
candidal activity was weaker than that of nystatin (but still close to
References
it). In general, the activity noted against Gram-negative bacteria
compared to Gram-positive ones was lower as expected from the
structure of their cell walls, with the G(ꢁ) bacterium P. aeruginosa
that turn out to be the most resistant strain in this test (this also
stands for tetracycline). However, the most susceptible microor-
ganism was also a G(ꢁ) bacterium K. pneumoniae with a MIC value
[1] A.S. Wagman, M.P. Wentland, in: J.B. Taylor, D.J. Triggle (Eds.), Comprehensive
Medicinal Chemistry II, vol. 7, Elsevier, Amsterdam, 2006, pp. 567–596.
[2] M.S. Atwal, L. Bauer, S.N. Dixit, J.E. Gearien, R.W. Morris, J. Med. Chem. 8 (1965)
566.
[3] Y. Xia, Z.-Y. Yang, P. Xia, K.F. Bastow, Y. Tachibana, S.-C. Kuo, E. Hamel, T. Hackl,
K.-H. Lee, J. Med. Chem. 41 (1998) 1155.
[4] S.-X. Zhang, J. Feng, S.-C. Kuo, A. Brossi, E. Hamel, A. Tropsha, K.-H. Lee, J. Med.
Chem. 43 (2000) 167.
of 0.01 lg/mL for the halogen-containing compounds and 0.05 lg/
[5] L.N. Bheemanapalli, A. Kaur, R. Arora, Sangeeta, R.R. Akkinepally, N.M. Javali,
Med. Chem. Res., in press. doi:10.1007/s00044-011-9688-z.
[6] O.V. Singh, R.S. Kapil, Synth. Commun. 23 (1993) 277.
[7] O. Prakash, D. Kumar, R.K. Saini, S.P. Singh, Synth. Commun. 24 (1994) 2167.
[8] C. Mannich, M. Dannehl, Chem. Ber. 71 (1938) 1899.
[9] H. de Diesbach, H. Kramer, Helv. Chim. Acta 28 (1945) 1399.
[10] G. Janzso, E.M. Philbin, Tetrahedron Lett. (1971) 3075.
[11] A.L. Tökés, L. Szilágy, Synth. Comm. 17 (1987) 1235.
[12] J.A. Donnelly, D.F. Farrell, J. Org. Chem. 55 (1990) 1757.
[13] J.S. Mahanty, M. De, P. Das, N.G. Kundu, Tetrahedron 53 (1997) 13397.
[14] R.S. Varma, R.K. Saini, Synlett (1997) 857.
mL for 4a. It seems worthwhile to note that the activity of the
compounds follows a trend of decreasing hydrophilic character
(estimated [52] logPo/w values for the phenyl analogues of com-
pounds 4a–c are 2.901, 3.564 and 3.740, respectively), hence, indi-
cating that the solubility of the compounds in the bilipid cell
membranes may play a significant role in the observed activity.
However, further work is necessary in establishing the true Modus
operandi. Overall, these results are highly promising and suggest
that a more detailed study of the antimicrobial activity of this class
of compounds could identify further derivatives with improved
antibacterial and antifungal properties.
[15] R. Varma, J. Heterocyclic Chem. 36 (1999) 1565.
[16] K.H. Kumar, P.T. Perumal, Can. J. Chem. 84 (2006) 1079.
[17] N. Ahmed, J.E. van Lier, Tetrahedron Lett. 47 (2006) 2725.
[18] J. Li, L. Jin, C. Yu, W. Su, J. Chem. Res. (2009) 170.
[19] N. Ahmed, J.E. van Lier, Tetrahedron Lett. 48 (2007) 13.
[20] E. Tang, B. Chen, L. Zhang, W. Li, J. Lin, Synlett (2011) 707.
[21] L. Wu, B. Niu, W. Li, F. Yan, Bull. Korean Chem. Soc. 30 (2009) 2777.
[22] D. Kumar, G. Patel, B.G. Mishra, R.S. Varma, Tetrahedron Lett. 49 (2008) 6974.
[23] D. Kumar, G. Patel, A. Kumar, R.K. Roy, J. Heterocyclic Chem. 46 (2009) 791.
[24] R.N. Bhattacharya, P. Kundu, G. Mait, Synth. Commun. 40 (2010) 476.
[25] M. Muthukrishnan, M. Mujahid, V. Punitharasu, D.A. Dnyaneshwar, Synth.
Commun. 40 (2010) 1391.
[26] J.I. Lee, H.J. Jung, J. Korean Chem. Soc. 51 (2007) 106.
[27] W.J. Lee, J.M. Chea, Y. Jahng, Bull. Korean Chem. Soc. 30 (2009) 3061.
[28] M. Zora, Ö. Veliog˘lu, J. Organomet. Chem. 693 (2008) 2159.
[29] M.D. Vukic´evic´, K. Wurst, A.G. Müller, G. Laus, R.D. Vukic´evic´, Polyhedron 24
(2005) 533.
4. Conclusion
Three new ferrocene containing 2,3-dihydroquinolin-4(1H)-
ones were synthesized and fully characterised by spectral, chemi-
cal and crystallographic data. Electrochemical investigations, per-
formed by cyclic voltammetry, showed that these compounds
exhibit two oxidation waves and a reduction one corresponding
to one reversible redox pair and a certain irreversible event. In a
microdilution assay, the three compounds showed a very strong
and unselective activity against the tested pathogenic bacteria
[30] I. Damljanovic´, M. Vukic´evic´, N. Radulovic´, R. Palic´, E. Ellmerer, Z. Ratkovic´,
M.D. Joksovicc´, R.D. Vukic´evic´, Bioorg. Med. Chem. Lett. 19 (2009) 1093.
´
´
´
´
´
[31] Z. Ratkovic, S.B. Novakovic, G.A. Bogdanovic, D. Šegan, R.D. Vukicevic,
Polyhedron 29 (2010) 2311.
and a yeast with MIC values ranging from 0.01 to 10.0 lg/mL. In
all cases the bromo-derivative had lower MIC values that those
of tetracycline. The obtained results make these compounds excel-
lent leads for the development of new more potent ferrocene-con-
taining antimicrobial agents.
´
´
´
´
´
´
[32] D. Ilic, I. Damljanovic, D. Stevanovic, M. Vukicevic, N. Radulovic, V. Kahlenberg,
G. Laus, R.D. Vukic´evic´, Polyhedron 29 (2010) 1863.
[33] Oxford Diffraction (2008). CrysAlis CCD and CrysAlis RED. Versions
1.171.32.24. Oxford Diffraction Ltd., Abington, England.
[34] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[35] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
[36] C.F. Macrae, P.R. Edgington, P. McCabe, E. Pidcock, G.P. Shields, R. Taylor, M.
Towler, J. van de Streek, J. Appl. Crystallogr. 39 (2006) 453.
[37] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
Acknowledgements
[38] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
[39] M. Nardelli, J. Appl. Crystallogr. 28 (1995) 659.
´
We thank Prof. Vladimir Divjakovic for collecting the diffraction
data of 4b and 4c compounds. This work was supported by the
Ministry of Education and Science of the Republic of Serbia (Grant
172034).
[40] NCCLS – National Committee for Clinical Laboratory Standards, Document
M100-S11. Performance standards for antimicrobial susceptibility testing.
National committee for clinical laboratory Standard, Wayne, PA, USA, 2003.
[41] N. Radulovic´, M. Dekic´, Z. Stojanovic´-Radic´, S. Zoranic´, Chem. Biodivers. 7
(2010) 2783.
[42] M.J. Mphahlele, P.T. Kaye, Magn. Reson. Chem. 36 (1998) 69.
[43] M.-S. Park, J.-I. Lee, Bull. Korean Chem. Soc. 25 (2004) 1269.
[44] J.-I. Lee, J.-S. Youn, Bull. Korean Chem. Soc. 29 (2008) 1853.
[45] Z. Galus, R.N. Adams, J. Phys. Chem. 67 (1963) 862.
[46] J. Bacon, R.N. Adams, J. Am. Chem. Soc. 90 (1968) 6596.
[47] R. Ojani, J.-B. Raoof, B. Norouzi, J. Mater. Sci. 44 (2009) 4095.
[48] P. Beagley, M.A.L. Blackie, K. Chibale, C. Clarkson, R. Meijboom, J.R. Moss, P.J.
Smith, H. Su, Dalton Trans. 39 (2003) 3046.
[49] J.C. Swarts, D.M. Swarts, M.D. Mare, E.W. Neuse, C. La Madeleine, J.E. Van Lier,
Anticancer Res. 21 (2001) 2033.
[50] A.R. Katrizky, S. Rachwal, B. Rachwal, Tetrahedron 52 (1996) 15031.
[51] J.C. Craig, P.E. Person, J. Med. Chem. 14 (1971) 1221.
[52] W. Renxiao, F. Ying, L. Luhua, J. Chem. Inf. Comput. Sci. 37 (1997) 615.
Appendix A. Supplementary data
CCDC 839535, 839536 and 839537 contain the supplementary
crystallographic data for 4a, 4b and 4c. These data can be obtained
free of charge via http://www.ccdc.cam.ac.uk/conts/retriev-
ing.html, or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223 336 033; or
e-mail: deposit@ccdc.cam.ac.uk. Supplementary data associated
with this article can be found, in the online version, at doi:10.
1016/j.poly.2011.11.006.