Design and synthesis of novel thiourea metal complexes with controllable antibacterial properties

Article abstract of DOI:10.1002/aoc.4342

A new bidentate O,S donor thiourea ligand (L¹), namely N-(2-hydroxyethyl)-N′-2-chlorobenzoylthiourea, and its oxazolidine derivative (L²) were synthesized. Derivative L² was used for the preparation of Ni(L²)₂ and Cu(L²)₂ complexes. The compounds were investigated using X-ray crystallography and Fourier transform infrared, ¹H NMR and UV–visible spectroscopies. Single-crystal X-ray analysis showed strong hydrogen bonding interactions between carbonyl oxygen and N(10) ─ H in the L¹ ligand. In addition, the antibacterial activities of these compounds were evaluated against Gram-positive and Gram-negative bacteria, measured using the colony count method. The Cu(L²)₂ complex exhibited a significant antibacterial activity while the activity of the other compounds was much lower. Finally, the relationship between the structure and antibacterial properties of these compounds was investigated using highest occupied and lowest unoccupied molecular orbital energies calculated by density functional theory method based on the 6-31G*/LANL2DZ basis set.

Full text of DOI:10.1002/aoc.4342

Received: 26 December 2017
DOI: 10.1002/aoc.4342
Revised: 9 February 2018
Accepted: 12 February 2018
F U L L P A P E R
Design and synthesis of novel thiourea metal complexes
with controllable antibacterial properties
Sajjad Rakhshani1
| Ali Reza Rezvani1
| Michal Dušek2 | Václav Eigner2
1 Department of Chemistry, University of
Sistan and Baluchestan, PO Box 98135‐
674, Zahedan, Iran
2 Institute of Physics, Academy of Sciences
of the Czech Republic, Na Slovance 2, 182
21 Praha 8, Czech Republic
A new bidentate O,S donor thiourea ligand (L1), namely N‐(2‐hydroxyethyl)‐N′‐
2‐chlorobenzoylthiourea, and its oxazolidine derivative (L2) were synthesized.
Derivative L2 was used for the preparation of Ni(L2)2 and Cu(L2)2 complexes.
The compounds were investigated using X‐ray crystallography and Fourier
1
transform infrared, H NMR and UV–visible spectroscopies. Single‐crystal X‐
Correspondence
ray analysis showed strong hydrogen bonding interactions between carbonyl
oxygen and N(10)─ H in the L1 ligand. In addition, the antibacterial activities
of these compounds were evaluated against Gram‐positive and Gram‐negative
bacteria, measured using the colony count method. The Cu(L2)2 complex exhib-
ited a significant antibacterial activity while the activity of the other compounds
was much lower. Finally, the relationship between the structure and antibacte-
rial properties of these compounds was investigated using highest occupied and
lowest unoccupied molecular orbital energies calculated by density functional
theory method based on the 6‐31G*/LANL2DZ basis set.
Ali Reza Rezvani, Department of
Chemistry, University of Sistan and
Baluchestan, PO Box 98135‐674, Zahedan,
Iran.
Email: ali.reza_rezvani@yahoo.com
Funding information
Czech Science Foundation, Grant/Award
Number: 15‐12719S; University of Sistan
and Baluchestan
KEYWORDS
bacteria inactivation, colony count method, DFT, hydrogen bond, oxazolidine derivative
1 | INTRODUCTION
through changing the geometry of the complexes, type
of metal ions, oxidation state of metals and total charges
N‐Benzoylthiourea and its derivatives are versatile ligands
capable of coordinating a range of metal centres. The oxy-
gen, nitrogen and sulfur donors of these ligands provide
monobasic bidentate (O,S), neutral monodentate (S) and
neutral bidentate (O,N) binding modes.[1–4] Thiourea
derivatives and their metal complexes exhibit a vast range
of biological activities including antitumour,[5,6] antitu-
bercular,[7,8] insecticidal, antiprotozoal and herbicide
activities and they also act as inhibitors for prevention of
corrosion of a wide range of metals in various corrosion
environments.[9–11] Moreover, these compounds have
been extensively used as antibacterial agents, which is
important in light of a rapid increase of bacterial resis-
tance against known drugs.[12]
on complex ions.[13] In this respect, the kinetically labile,
square‐planar divalent (Cu, Ni and Pd) and octahedral tri-
valent (Fe, Co and Cr) complexes displayed more activity
than other inert complexes, even at low concentra-
tions.[14] For instance, the chelation of benzoylthiourea
derivatives to metal ions resulted in a significant increase
in their biological activities.[15] In this respect, weak inter-
actions are also important, especially intramolecular and
intermolecular hydrogen bonds. In the case of
benzoylthiourea derivatives, there are prominent intra-
molecular hydrogen bonds between N ─ H and carbonyl
moieties,[16] which can influence the coordination modes
of N‐benzoyl‐N′‐substituted thiourea compounds and
result in coordination through the sulfur atom of the
ligand.[17] Therefore, it is possible to influence the biolog-
ical properties of these compounds by elimination of the
One of the approaches for increasing the efficacy of
antibacterial agents consists of improving their activity
Appl Organometal Chem. 2018;e4342.
wileyonlinelibrary.com/journal/aoc
Copyright © 2018 John Wiley & Sons, Ltd.
1 of 13
https://doi.org/10.1002/aoc.4342

2 of 13
RAKHSHANI ET AL.
intramolecular hydrogen bonds causing chelation of
ligand to metal ion through sulfur and oxygen atoms.
In this article, we report on a new thiourea ligand, N‐
(2‐hydroxyethyl)‐N′‐2‐chlorobenzoylthiourea (L1). We
applied the above‐mentioned hydrogen bond elimination
by preparing the oxazolidine derivative (L2) of L1 with a
different coordination mode (Scheme 1), and finally we
prepared Ni and Cu complexes of L2. The compounds
were characterized using X‐ray crystallography and Fou-
2.2 | Syntheses
2.2.1 | Synthesis of ligand L1
Ligand L1 was synthesized according to Scheme 2. An
amount of 10 mmol of ammonium thiocyanate was dis-
solvedindichloromethane, and 10 mmol of 2‐chlorobenzoyl
chloride was added to this solution in a dropwise fashion.
The mixture was refluxed for 30 min. After that 10 mmol
of 2‐aminoethanol was added dropwise to the crude 2‐
chlorobenzoyl isothiocyanate. Once again the solution was
heated to reflux for 3 h. Upon completion of heating, the
product was recrystallized from CH2Cl2–C2H5OH, resulting
in a white precipitate which was filtered off. Diffusion of
ether into a N,N‐dimethylformamide (DMF) solution of
the ligand gave colourless crystals. Yield 90%; m.p. 121–
122 °C. Selected FT‐IR data (ν, cm−1): 3525 (O ─ H), 3252
(N ─ H), 3051 (Ph─ H), 2937 (C ─ H), 1675 (C ─ O), 1291
1
rier transform infrared (FT‐IR), H NMR and UV–visible
spectroscopies, and their antibacterial activity was evalu-
ated against standard bacteria using the colony count
method. We also considered density functional theory
(DFT) results for the highest occupied molecular orbital
(HOMO) and lowest unoccupied molecular orbital
(LUMO) energies using the 6‐31G*/LANL2DZ basis set
for investigation of the relationship between the struc-
tures of these compounds and their antimicrobial
activities.
1
(C ─ S). H NMR (CDCl3; δ, ppm): 10.83 (s, 1H, N(8) ─
H), 9.17 (s, 1H, N(10)─ H), 7.29–7.76 (m, 4H, Ph), 3.95–
4.03 (m, 4H, ─ CH2), 1.91–1.94 (t, 1H, OH).
2 | EXPERIMENTAL
2.2.2 | Synthesis of N‐(2,2‐
dimethyoxazolidine)‐N′‐2‐
2.1 | Methods and Materials
chlorobenzoylthiourea ligand (L2)
FT‐IR spectra were recorded with KBr pellets using a
PerkinElmer 781 spectrophotometer. A Bruker 300 MHz
AVANCE III spectrometer was utilized to obtain NMR
spectra in CDCl3 solvent. Melting points were measured
with an Electrothermal 9100 apparatus. UV–visible spec-
tra were obtained using a JASCO‐570 spectrophotometer.
The spectra were measured in dichloromethane solution
at room temperature. Single‐crystal X‐ray data were
collected with a SuperNova diffractometer (Rigaku
Oxford Diffraction) using mirror‐collimated Cu Kα
radiation from a micro‐focus sealed X‐ray tube and CCD
detector (Atlas S2) at 95 K and with a Gemini diffractom-
eter (Rigaku Oxford Diffraction) using graphite‐
monochromated Mo Kα radiation from a sealed X‐ray
tube and CCD detector (Atlas S2) at 120 K. To visualize
compound structures, Diamond 3.2 K was used.
The oxazolidine derivative (L2) of the thiourea ligand was
synthesized following a procedure as given in the litera-
ture.[18–20] A mixture of 10 mmol of L1 and 20 ml of acetone
was refluxed for 3 h (Schemes 1 and 2). The product was
recrystallized from dichloromethane and identified using
1
FT‐IR and H NMR spectra. Yield 97%; m.p. 143–144 °C.
Selected FT‐IR data (ν, cm−1): 3241 (N ─ H), 3051 (Ph ─ H),
2985 (C ─ H), 1706 (C ─ O), 1417 (C ─ S), 826 (C ─ Cl). 1H
NMR (CDCl3; δ, ppm): 8.304 (s, 1H, N(8) ─ H), 7.29–7.74
(m, 4H, Ph), 4.04–4.13 (m, 4H, ─ CH2), 1.95 (s, 6H, ─ CH3).
2.2.3 | Synthesis of Ni(L2)2 complex
The desired complex was prepared by addition of an
ethanolic solution of nickel acetate to a solution of ligand
All solvents and chemical materials were of reagent
grade and were purchased from Merck, including 2‐
chlorobenzoyl chloride, ammonium thiocyanate, 2‐
aminoethanol, Ni(CH3COO)2⋅4H2O, Cu(CH3COO)2⋅H2O
and acetone.
SCHEME 1 Chemical structures of thiourea ligand (L1) and its
oxazolidine derivative (L2)
SCHEME 2 Synthetic route to ligand L2 and its Ni(II) and Cu(II)
complexes

RAKHSHANI ET AL.
3 of 13
L2 in a molar ratio of 1:2 at room temperature. The
resulting mixture was refluxed for 3 h. The purple precip-
itate of the complex was filtered and recrystallized from
DMF. Yield 90%. Selected FT‐IR data (ν, cm−1): 3048
of each bacterial suspension was dispersed on nutrient
agar medium, and finally, the number of colony forming
units was enumerated with consideration to the dilution
factor after 24 h of incubation at 37 °C.
1
(Ph ─ H), 2932 (C ─ H), 1519 (C ─ O), 1255 (C ─ S). H
NMR (CDCl3; δ, ppm): 7.25–7.55 (m, 8H, Ph), 3.84–4.06
(m, 8H, 4CH2), 1.65 (s, 12H, ─ CH3).
2.5 | Calculation Details
The total computations were accomplished using the DFT
(B3LYP) method at the 6‐31G*/LANL2DZ basis set level
using the Gaussian 03 program. After optimization, some
DFT results for the HOMO and LUMO energies were con-
sidered by using the 6‐31G* basis set for C, H, N, O, S and
Cl and LANL2DZ for Ni and Cu atoms for investigation of
antibacterial properties of the title compounds.
2.2.4 | Synthesis of Cu(L2)2 complex
A similar procedure was carried out using L2 (2 mmol)
and Cu(CH3COO)2⋅H2O (1 mmol) and refluxing for 3 h.
The green precipitate was filtered and recrystallized from
ethanol. Yield 80%. Selected FT‐IR data (ν, cm−1): 3048
(Ph ─ H), 2932 (C ─ H), 1514 (C ─ O), 1249 (C ─ S).
3 | RESULTS AND DISCUSSION
3.1 | Synthesis and Structural Analysis
2.3 | X‐ray Crystallographic Analysis
The measurements of L1 and Cu(L2)2 were performed
with a Gemini diffractometer and those of Ni(L2)2 were
performed with a SuperNova diffractometer, as described
in Section 1.1. The data reduction and absorption correc-
tion were done with CrysAlis PRO software.[20] The struc-
tures were solved using charge flipping methods and
refined using Jana2006.[21,22] The residual electron den-
sity maps were visualized using MCE.[23] All hydrogen
atoms were discernible in difference Fourier maps and
could be refined to reasonable geometry. According to
common practice, H atoms bonded to C atoms were kept
in ideal positions with C ─ H = 0.96 Å while positions of
H atoms bonded to N and O atoms were refined with
restrained bond lengths. In both cases Uiso(H) was set to
1.2Ueq(C,N,O). All non‐hydrogen atoms were refined
using harmonic refinement. The disordered part of het-
erocycle in Cu(L2)2 was refined with restrained bond
lengths and angles. The overall occupancy was
constrained to full with final occupancy ratio of 74:26.
According to Scheme 2, the reaction between 2‐
chlorobenzoyl chloride and ammonium thiocyanate in
CH2Cl2 produced 2‐chlorobenzoyl isothiocyanate, and
condensation of this compound with 2‐aminoethanol
resulted in ligand L1. This ligand was purified and recrys-
tallized by ether diffusion into a DMF solution of the
ligand.
Interestingly, ligand L1 cannot coordinate metal ions
(Ni(II) and Cu(II)) in a bidentate manner because the
intramolecular hydrogen bond between N(10) ─ H and
carbonyl moiety forces this ligand into a monodentate
coordination by keeping the carbonyl oxygen in the trans
position with respect to sulfur. In order to eliminate this
hydrogen bond, we converted L1 to its oxazolidine deriva-
tive L2 according to Scheme 1.[28,29] Ligand L2 was recrys-
tallized from CH2Cl2. The complexes Ni(L2)2 and Cu(L2)2
were prepared by adding a hot ethanolic solution of the
respective metal acetate to a solution of L2 in a molar ratio
of 1:2 (Scheme 2). Recrystallization of the complexes from
DMF and ethanol yielded suitable crystals for X‐ray crys-
tallography. The synthesized Ni(II) and Cu(II) complexes
are soluble in DMF, dimethylsulfoxide, CH2Cl2 and tolu-
ene, and are relatively soluble in water at room tempera-
ture. These compounds were characterized using 1H
NMR, UV and FT‐IR spectroscopies and single‐crystal X‐
ray diffraction analysis.
2.4 | In Vitro Antimicrobial Studies
According to the literature, the antibacterial properties of
the synthesized compounds were measured using bacteria
as per the colony count method.[24–27] In order to measure
the rate of bacterial growth, the bacterial strains were
grown in nutrient broth medium supplemented with sus-
pensions of ligand L2, or Ni2+ or Cu2+ metal complexes at
a concentration of 20 ppm. These samples were added
into 10 ml of nutrient broth medium to which was added
200 μl of bacteria at a concentration of 1.5 × 105 CFU.
Blank samples were prepared by growth of the culture
in ligand‐ and metal complex‐free media under the same
conditions. Then these samples were agitated with a
shaker platform at a speed of 200 rpm. After that, 100 μl
3.2 | Spectroscopic Analysis
3.2.1 | FT‐IR spectral studies
According to the FT‐IR spectrum of ligand L1, the band at
3525 cm−1 should be attributed to the stretching of the
O─ H group. The medium broad peak at 3252 cm−1

4 of 13
RAKHSHANI ET AL.
corresponds to the stretching vibration of N ─ H groups.
Stretching vibrations of C ─ O and C ─ S bonds are
observed at 1676 and 1291 cm−1, respectively.[30] Symmet-
ric and asymmetric stretching vibrations of ( ─ CH2)
appeared at 2892–2937 cm−1. The absence of the
ν(S─ H) band at about 2570 cm−1 and the presence of
the ν(N ─ H) band at about 3252 cm−1 can be explained
by formation of the thione tautomer (L1) in the solid state.
FIGURE 1 1H NMR spectra of (a) L1,
(b) L2 and (c) Ni(L2)2 in CDCl3 at room
temperature

RAKHSHANI ET AL.
5 of 13
For L2, the main bands at 3241, 1706 and 1417 cm−1 are
ascribed to the stretching frequencies of the N(8) ─ H,
C ─ O and C ─ S groups. On the other hand, the loss of
N(10) ─ H and OH vibrational frequencies after conver-
sion of L1 to its oxazolidine derivative L2 confirmed the
formation of the oxazolidine heterocycle.[18,19]
absorption spectra were recorded in the range 200–
800 nm. For L2, the absorption bands observed at 230,
285 and 350 nm were assigned to intra‐ligand charge
transfer transitions π–π* (phenyl ring) and n–π* (corre-
sponding to C ─ O and C ─ S).[30]
After coordination of metal ions to ligand L2, the
absorption bands corresponding to the n–π* transitions
of C ─ O and C ─ S display a blue and red shift by 4–
10 nm relative to the free ligand.[32] In the spectra of
metal complexes Ni(L2)2 and Cu(L2)2, the bands corre-
sponding to the n–π* transitions of C ─ O appear at 269
After coordination of ligand L2 to the metal ions, the
N(8) ─ H stretching vibrations disappeared, indicating
the loss of H bonded to the N atom of amide group. Also
the vibrational frequencies corresponding to C ─ O and
C ─ S groups in the free ligand were shifted towards lower
wavenumber upon complexation, confirming that L2 is
coordinated to Ni2+ and Cu2+ through the sulfur and oxy-
gen atoms.[28,29,31]
and 275 nm, respectively, thus indicating
a
hypsochromic shift. On the other hand, the bands at
358 and 360 nm corresponding to the n–π* transitions
of C ─ S indicate a bathochromic shift due to the com-
plexation of a thione sulfur atom to the metal centre.
The band at about 295 nm for the Ni(L2)2 complex is
related to ligand‐to‐metal charge transfer.[33] Absorption
bands in the visible region, 503 and 586 nm for Ni(L2)2
and Cu(L2)2, respectively, are due to d–d transitions
(Figure 2).[34–39]
3.2.2 | NMR spectral studies
1
The H NMR spectra were recorded in CDCl3 at room
temperature at 300 MHz for ligands and at 250 MHz for
the Ni(II) complex. The presence of NH resonances and
absence of SH resonances revealed that the thione tauto-
mer instead of thiol tautomer was formed for both L1 L2.
1
The H NMR spectrum of L1 shows the N(10) ─ H and
N(8) ─ H resonances at 9.17 and 10.83 ppm. The signal
at 1.92 ppm and multiplets at 7.29–7.76 ppm are assigned,
respectively, to the OH and phenyl protons in the ligand
(Figure 1a).
The same results were obtained for L2 except the dif-
ferences caused by the missing N(10) ─ H and O ─ H
bonds due to the formation of the heterocyclic ring.
Therefore, the spectrum of metal‐free L2 exhibits four
well‐resolved signals at room temperature, N(8) ─ H,
Ph ─ H, ─ CH2 and ─ CH3 resonances appearing at
8.30, 7.29–7.74, 4.04–4.13 and 1.95 ppm, respectively
(Figure 1b).
According to Figure 1(c), the deprotonation of L2
after complexation to Ni(II) is corroborated by the loss
of the N(8) ─ H signal at 8.30 ppm. In the spectrum of
Ni(L2)2, each peak is shifted downfield from its normal
peak position corresponding to L2. The diastereotopic
hydrogens of ─ CH2 moieties at 4.04–4.13 ppm for L2
are shifted downfield to 3.84–4.05 ppm for Ni(L2)2 due
to the negative induction effect. Similarly, the
characteristic signal of the methyl ( ─ CH3) protons on
oxazolidine ring at 1.95 ppm for L2 is shifted downfield
to 1.65 ppm for Ni(L2)2 (Figure 1c). Both FT‐IR and
NMR data are in agreement with the proposed
structures.
FIGURE 2 Electronic spectra of ligand L2 and its Ni(II) and
Cu(II) complexes in dichloromethane
3.2.3 | UV–visible spectral studies
CH2Cl2 solutions of thiourea ligand L2 and its Ni(II) and
Cu(II) complexes were prepared and the electronic
FIGURE 3 Molecular structure and intramolecular hydrogen
bonding of ligand L1

6 of 13
RAKHSHANI ET AL.
3.3 | Description of Crystal Structures
3.3.1 | Crystal structure of L1
The structures of L1 and the Ni(II) and Cu(II) complexes
of its oxazolidine derivative were studied using single‐
crystal X‐ray diffraction. Figures 3–5 show the features
of their atomic structures. Experimental details of data
collection and structure refinement are given in
Table 1. Selected bond lengths and bond angles are sum-
marized in Table 2.
According to Table 2, the bond length of carbonyl
(C7 ─ O1 = 1.221(6) Å) group of L1 corresponds to a dou-
ble bond, while the thiocarbonyl (C8 ─ S1 = 1.681(3) Å)
bond length lies between those of a single and a double
bond.[40] The C ─ N bonds, C7 ─ N1 = 1.367(7) Å,
C8─ N1 = 1.393(2) Å, C8 ─ N2 = 1.315(8) Å, are shorter
than the typical single C ─ N bond length of 1.479 Å due
FIGURE 4 (a) Molecular structure of
Ni(L2)2 and (b) view of unit cell of
Ni(L2)2 along the b‐axis

RAKHSHANI ET AL.
7 of 13
FIGURE 5 (a) Molecular structure of
Cu(L2)2 and (b) view of unit cell of Cu(L2)2
along the b‐axis
to the delocalization in the ligand.[10,19] This delocaliza-
tion is also confirmed by values of the angles C7–N1–
C8 = 127.22(1)° and C8–N2–C9 = 123.00(9)° indicating
sp2 hybridization on the N1 and N2 atoms.[20] The
conformation of L1 with regard to the thiocarbonyl and
carbonyl moieties is twisted, as reflected by the
torsion angles O1–C7–N1–C8 = 4.520(7)°, C7–N1–C8–
N2 = −1.106(2)° and S1–C8–N1–C7 = 177.022(0)°. The
C ─ O and C ─ S moieties are located at opposite sides
of the molecule.
As evident from Figure 3 and Table 2, the N2 atom
participates in a bifurcated intramolecular hydrogen bond
to carbonyl oxygen (O1) and hydroxyl oxygen (O2). The
strong 1.966(2) Å hydrogen bond towards O1 closes the

8 of 13
RAKHSHANI ET AL.
TABLE 1 Summary of X‐ray diffraction data
L1
Ni(L2)2
Cu(L2)2
Empirical formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
a (Å)
C10H11Cl1N2O2S1
258.7
C26H28Cl2N4Ni1O4S2
654.3
C26H28Cl2Cu1N4O4S2
659.1
120
95
120
0.71073
Triclinic
P ‐1
1.54184
Monoclinic
P 1 21/c 1
11.3603(2)
11.7544(2)
21.6307(4)
90
0.71073
Monoclinic
P 1 21/c 1
11.2812(3)
11.7893(2)
21.6896(6)
90
4.5487(2)
10.3397(6)
12.4969(7)
78.052(5)
88.006(4)
89.013(5)
574.64(5)
2
b (Å)
c (Å)
α (°)
β (°)
100.7293(15)
90
98.525(2)
90
γ (°)
V (Å3)
2837.93(9)
4
2852.79(12)
4
Z
Dcalc (g cm−3
μ (Mo Kα)
F(000)
)
1.4953
1.5313
1.5346
0.499
4.411
1.138
268
1352
1388
θ range (°)
3.52 to 29.48
6442
3.96 to 74.53
20 408
2.43 to 29.43
26 710
Reflections collected
Independent reflections
Rint
2763
5693
7120
0.013
0.0278
0.0199
Data/parameters
GOF on F2
2763/155
1.580
5693/353
1.670
7120/360
1.36
R1 [I > 3σ(I)]
0.0265
0.0282
0.0259
wR2 [I > 3σ(I)]
R1 (all data)
0.0853
0.0863
0.0778
0.0299
0.0294
0.0335
wR2 (all data)
0.0876
0.0875
0.0815
Largest diff. Peak/hole (e Å−3
)
0.28/−0.22
0.69/−0.31
0.30/−0.27
ring C8–N1–C7–O1–H1n2–N2 into a planar structure,
which can compel the monodentate coordination of L1
to metal ions.[17]
The carbonyl and thiocarbonyl bond distances
changed from average values of 1.2216 and 1.6813 Å for
the free ligand to 1.267 and 1.74 Å for the complex
(C7a ─ O2a
= 1.26(6) Å, C1a ─ S1a = 1.74(1) Å;
C7b ─ O2b = 1.26(5) Å, C1b ─ S1b = 1.73(6) Å). The
bond lengths of C ─ N moieties (C7 ─ N2a = 1.32(2)
3.3.2 | Crystal structure of Ni(L2)2
complex
Å, C1a ─ N2a
= 1.34(6) Å, C1a ─ N1a = 1.33(5)
Å, C7b ─ N2b = 1.31(9) Å, C1b ─ N2b = 1.34(6) Å,
C1b ─ N1b = 1.33(3) Å) are shorter than for a typical sin-
gle C ─ N bond. This variation in C ─ O, C ─ S and C ─ N
bond lengths of the complex indicates a marked electronic
delocalization in the chelate rings.
Figure 4(a) shows that the Ni centre is chelated by two
deprotonated ligands coordinated in the bidentate O,S
mode in cis arrangement, giving rise to a nearly square
planar geometry around Ni (O2b–Ni1–Sa1 = 179.29(3)°,
O2a–Ni1–S1b = 174.88°, O2a–Ni1–S1a = 94.72°, O2b–
Ni1–S1b = 94.79°, Ni1–O2b–C7b–N2b = 3.651°, Ni1–
S1b–C1b–N2b = 19.534°).
The crystal packing of Ni(L2)2 demonstrates a weak inter-
molecular hydrogen bonding, C6b─ H1c6b⋅⋅⋅N2bi (symme-
try code: (i) 1 − x, −0.5 + y, 0.5 – z) (Table 2; Figure 4b).

RAKHSHANI ET AL.
9 of 13
TABLE 2 Selected bond lengths, angles, torsion angles and interactions
L1
Ni(L2)2
Cu(L2)2
Bond lengths (Å)
Bond lengths (Å)
Bond lengths (Å)
C7–O1
1.221(6)
1.681(3)
1.367(7)
1.393(1)
1.315(8)
0.680(9)
0.859(7)
Ni1–O2b
Ni1–S1b
1.861(9)
2.145(3)
1.857(1)
2.146(7)
1.265(0)
1.736(6)
1.319(3)
1.346(7)
1.333(5)
1.266(8)
1.741(7)
1.322(1)
1.346(8)
1.335(6)
Cu1–O1b
Cu1–S1b
1.918(3)
2.244(3)
1.925(4)
2.227(3)
1.258(5)
1.735(9)
1.262(4)
1.743(0)
1.318(4)
1.350(8)
1.331(1)
1.326(3)
1.342(2)
1.331(8)
C8–S1
C7–N1
Ni1–O2a
Cu1–O1a
Cu1–S1a
C8–N1
Ni1–S1a
C8–N2
C7b–O2b
C1b–S1b
C7a–O1a
C8a–S1a
N2–H1n2
N1–H1n1
C7b–N2b
C1b–N2b
C1b–N1b
C7a–O2a
C1a–S1a
C7b–O1b
C8b–S1b
C7a–N1a
C8a–N1a
C8a–N2a
C7b–N1b
C8b–N1b
C8b–N2b
Bond angles (°)
C7a–N2a
C1a–N2a
C1a–N1a
Bond angles (°)
Bond angles (°)
O1–C7–N1
122.97(5)
119.11(3)
117.58(2)
122.22(1)
123.00(9)
O2b–Ni1–S1a
O2a–Ni1–S1b
O2a–Ni1–S1a
O2b–Ni1–S1b
176.29(3)
174.88(3)
94.72(3)
94.79(3)
O1a–Cu1–S1b
O1b–Cu1–S1a
O1a–Cu1–S1a
O1b–Cu1–S1b
174.99(0)
173.19(2)
92.89(1)
91.79(3)
S1–C8–N1
N1–C8–N2
C7–N1–C8
C8–N2–C9
Torsion angles (°)
Torsion angles (°)
Torsion angles (°)
O1–C7–N1–C8
C7–N1–C8–N2
S1–C8–N1–C7
4.52(1)
−1.10(6)
177.02(2)
Ni1–O2b–C7b–N2b
Ni1–S1b–C1b–N2b
Ni1–O2b–C7a–N2a
Ni1–S1a–C1a–N2a
3.65(1)
19.54(3)
27.75(5)
18.07(1)
Cu1–S1a–C8a–N1a
Cu1–O1a–C7a–N1a
Cu1–S1b–C8b–N1b
Cu1–O1b–C7b–N1b
−16.13(6)
4.68(6)
−21.41(7)
−33.99(9)
Hydrogen bond geometries
Compound
D–H⋅⋅⋅A
D–H (Å)
H⋅⋅⋅A (Å)
D⋅⋅⋅A (Å)
D–H⋅⋅⋅A (°)
L1
N2–H1n2⋅⋅⋅O1
0.860(8)
0.860(8)
0.824(6)
0.960(0)
0.959(9)
1.966(2)
2.465(9)
2.481(0)
2.904(3)
2.888(7)
2.645(3)
2.748(3)
2.996(0)
3.473(6)
3.684(2)
134.92(0)
99.92(4)
N2–H1n2⋅⋅⋅O2
Ni(L2)2
Cu(L2)2
C6b–H1c6b⋅⋅⋅N2bi
C13b–H1c13b⋅⋅⋅O1bii
C12a–H3c12a⋅⋅⋅N1biii
113.49(1)
119.08(1)
140.94(4)
(i) 1 − x, −0.5 + y, 0.5 − z; (ii) 1 − x, 1 − y, −z; (iii) 1 + x, −1 + y, z
3.3.3 | Crystal structure of Cu(L2)2
complex
S1a = 92.83°, O1b–Cu1–S1b = 92.80°; Cu1–S1a–C8a–
N1a = −16.136°, Cu1–O1a–C7a–N1a = 4.686°), and it also
features a decrease of the bond order of carbonyl and
thiocarbonyl moieties as well as an increase of C ─ N bond
orders. This again indicates the presence of marked delo-
calization of electron density in the chelate ring of the
complex.[33,34] Unlike the Ni complex, one of the
The Ni and Cu complexes are isotypic. The Cu complex
(Figure 5a) has a similar slightly distorted square
planar geometry around the metal atom (O1a–Cu1–
S1b = 174.99°, O1b–Cu1–S1a = 173.19°, O1a–Cu1–

10 of 13
RAKHSHANI ET AL.
oxazolidine ring moieties is disordered with occupancy of
the minor disordered part refined to 0.259(4). The atom
labels of the minor disordered part are indicated by a
prime: O2b′, C10b′, C11b′ and corresponding hydrogen
complexes compared with the free ligand and increases
their ability to permeate the peptidoglycan layer of cell
walls or the lipid layer of cell membranes. While ligand
L1 is unable to coordinate to the metal ions in a bidentate
manner due to the strong intramolecular hydrogen bond
between N2─ H1n2 and the carbonyl oxygen atom (O1)
(Figure 3), its oxazolidine derivative L2 can provide such
complexes. As a result, the elimination of the intramolec-
ular hydrogen bond in L2 leads to the increase of antibac-
terial properties of complexes.
atoms.
The
main
intermolecular
interactions
are C13b ─ H1c13b⋅⋅⋅O1bi and C12a ─ H3c12a⋅⋅⋅N1bii
(symmetry code: (i) 1 − x, 1 − y, −z; (ii) 1 + x, −1 + y, z)
(Table 2; Figure 5b).
3.4 | Antibacterial Assay
We found that L2 and its Ni(II) and Cu(II) complexes
were more powerful bactericidal agents against the
Gram‐negative bacterium than against the Gram‐positive
bacterium. The difference in the efficacy of these agents
can be ascribed to the differences in the cell wall struc-
ture of the otherwise similar Gram‐negative and Gram‐
positive bacteria; a Gram‐negative bacterium has a thin
peptidoglycan layer and an outer membrane that con-
tains proteins, phospholipids and lipopolysaccharide,
while a Gram‐positive bacterium has a thick peptidogly-
can layer that contains teichoic and lipoteichoic acid.
Therefore, the cell wall of a Gram‐negative bacterium is
more polar, and the permeation of Ni(L2)2 and Cu(L2)2
complexes into the microorganism is facilitated by this
polarity.[43] Therefore, the efficacy of the investigated
The free ligand (L2) and its Ni2+ and Cu2+ metal com-
plexes were investigated for their antibacterial activity
against Gram‐negative (Escherichia coli; E. coli ATCC
25922) and Gram‐positive (Staphylococcus aureus; S.
aureus ATCC 25923) bacteria. The studied compounds
have varying bactericidal effects on the tested bacterial
strains. For both tested bacterial strains, the antibacterial
activity of the complexes was better than that of the free
ligand, as explained by Tweedy's chelation theory.[41]
According to this theory, the chelation reaction dimin-
ishes the polarity of the metal ions through partial sharing
of their positive charge with donor groups and delocaliza-
tion of electrons over the chelating ring.[42] This augments
the lipophilic feature of the Ni(L2)2 and Cu(L2)2
FIGURE 6 Plot of survival of E. coli
with L2, Ni(L2)2 and Cu(L2)2
FIGURE 7 Plot of survival of S. aureus
with L2, Ni(L2)2 and Cu(L2)2

RAKHSHANI ET AL.
11 of 13
compounds against E. coli is greater than against S.
aureus.[25,44,45]
the crystallographic data (Table 2), the C ─ O and C ─ S
bond lengths are larger for Cu(L2)2, and indeed the anti-
bacterial activity of this complex is greater than that of
Ni(L2)2 due to better connection of the carbonyl and the
thiocarbonyl to the microorganism.[47–49] Thus, Cu(L2)2
is the most efficient inhibitor among the three investi-
gated compounds (Figures 6 and 7).
Furthermore, the strengths of the antibacterial activity
are associated with the elongation or shortening of the
C ─ O and C ─ S bond lengths in the complexes, because
the elongation of such bonds increases both the polarity
and reactivity of C ─ O and C ─ S bonds.[46] Pursuant to
FIGURE 8 Frontier molecular orbitals of the studied compounds

12 of 13
RAKHSHANI ET AL.
Ali Reza Rezvani http://orcid.org/0000-0003-2681-9906
Václav Eigner http://orcid.org/0000-0003-1014-3980
3.5 | Structure–Activity Relationship
The optimization of the molecular structures of L1, L2 and
Ni(II) and Cu(II) complexes of L2 was accomplished using
the DFT method (B3LYP) at 6‐31G* and LANL2DZ basis
set level. After that, DFT results for the frontier molecular
orbital energies were obtained. The frontier molecular
orbitals of these samples are shown in Figure 8.
REFERENCES
[1] D. G. Patil, M. R. Chedekel, J. Org. Chem. 1984, 49, 997.
[2] Z. F. Li, X. X. Cheng, G. Li, H. J. Lu, H. F. Zhang, J. Lumin.
2010, 130, 2192.
These results confirm the inverse connection between
the size of the band gap (ΔEL‐H) and the antibacterial
activities of these samples.[38,39,50–52] In other words, the
smaller is the difference between LUMO and HOMO
energies, the greater is the antibacterial activity. Accord-
ing to Figure 8, the Cu(L2)2 complex has the narrowest
band gap, which conforms to its greatest antibacterial
activity (Figures 6 and 7) among the investigated
compounds.
[3] N. Selvakumaran, N. S. Weng, E. R. T. Tiekink, R. Karvembu,
Inorg. Chim. Acta 2011, 376, 278.
[4] A. M. Mansour, Appl. Organometal. Chem. 2018, 32, e3928.
[5] W. Schiessl, R. Puchta, Ž. D. Bugarčić, F. W. Heinemann, R.
Eldik, Eur. J. Inorg. Chem. 2007, 1390.
[6] D. C. Schroeder, Chem. Rev. 1955, 55, 181.
[7] E. Rodrıǵ uez‐Fernández, E. Garcı́a, M. R. Hermosa, A.
Jiménez‐Sánchez, M. Mar Sánchez, E. Monte, J. J. Criado,
J. Inorg. Biochem. 1999, 75, 181.
[8] K. Chandrasekhara Pillai, R. Narayan, J. Electrochem. Soc. 1978,
125, 1393.
4 | CONCLUSIONS
[9] A. Mesbah, S. Jacques, E. Rocca, M. François, J. Steinmetz, Eur.
J. Inorg. Chem. 2011, 1315.
The intramolecular hydrogen bond in ligand L1 prevents
coordination of the carbonyl moiety to metal ions and
causes synthesis of a mixture of cis and trans complexes.
To eliminate this intramolecular hydrogen bond and favour
a bidentate chelating of ligand to metal ion, we prepared the
oxazolidine derivative of the L1 ligand, L2, and used it for
synthesis of the bis‐chelate Ni(II) and Cu(II) complexes.
The crystal structures of Ni(L2)2 and Cu(L2)2 revealed that
the Ni and Cu atoms display a nearly square planar geome-
try with two bidentate ligands having oxygen and sulfur as
donor atoms in cis position. In vitro antibacterial activity
experiments against E. coli and S. aureus showed that the
Cu(L2)2 complex has the strongest antibacterial activity
among the investigated compounds. This is due to the che-
lation of L2 to Cu2+ ion, which decreases its polarity and
increases the lipophilic property. Theoretical computations
illustrate that the antibacterial properties of the synthesized
compounds are related to the difference between frontier
orbital energies, where a decrease in the band gap energy
corresponds to an increase of the antibacterial activity.
[10] M. Ozcan, I. Dehri, M. Erbil, Appl. Surf. Sci. 2004, 236, 155.
[11] K. F. Khaled, Appl. Surf. Sci. 2010, 256, 6753.
[12] S. Saha, D. Dhanasekaran, S. Chandraleka, N. Thajuddin, A.
Panneerselvam, Adv. Biol. Res. 2010, 4, 224.
[13] H. M. Parekh, P. B. Pansuriya, M. N. Patel, Polish J. Chem. 2005,
79, 1843.
[14] V. Bobbarala, A Search for Antibacterial Agents, InTech, Rijeka,
Croatia 2012.
[15] B. H. Abdullah, Y. M. Salh, Orient. J. Chem. 2010, 26, 763.
[16] G. Kurt, F. Sevgi, B. Mercimek, Chem. Pap. 2009, 63, 548.
[17] K. R. Koch, Y. Wang, A. Coetzee, J. Chem. Soc. Dalton Trans.
1999, 63, 1013.
[18] E. D. Bergmann, Chem. Rev. 1953, 53, 309.
[19] M. Al‐Masum, B. W. Lott, N. Ghazialsharif, Int. J. Org. Chem.
2012, 362.
[20] E. D. Bergmann, E. Zimkin, S. Pinchas, Recl. Trav. Chim. Pays‐
Bas 1952, 71, 168.
[21] Crys AlisPro, Rigaku Oxford Diffraction, 2015.
[22] L. Palatinus, G. Chapuis, J. Appl. Crystallogr. 2007, 40, 786.
[23] V. Petricek, M. Dusek, L. Palatinus, Z. Kristallogr. 2014, 229, 345.
[24] J. Rohlicek, M. Husak, J. Appl. Crystallogr. 2007, 40, 600.
ACKNOWLEDGEMENTS
[25] N. Talebian, S. M. Amininezhad, M. Doudi, J. Photochem.
Photobiol. B 2013, 120, 66.
We thank USB for financial support. The crystallographic
investigation was supported by project 15‐12719S of the
Czech Science Foundation using instruments of the ASTRA
laboratory established within the Operation Program
Prague Competitiveness – project CZ.2.16/3.1.00/24510.
[26] S. M. Amininezhad, A. R. Rezvani, M. Amouheidari, S. M.
Amininejad, S. Rakhshani, J. Zahedan, Res. Med. Sci. 2015, 17,
e1053.
[27] H. Naeimi, Z. S. Nazifi, S. M. Amininezhad, J. Photochem.
Photobiol. B 2015, 149, 180.
[28] S. Bourne, K. R. Koch, J. Chem. Soc. Dalton Trans. 1993, 2071.
ORCID
[29] K. R. Koch, C. Sacht, S. Bourne, Inorg. Chim. Acta 1995,
232, 109.
Sajjad Rakhshani http://orcid.org/0000-0002-0490-3130

RAKHSHANI ET AL.
13 of 13
[30] M. Atiş, F. Karipcin, B. Sarıboğa, M. Taş, H. Çelik, Spectrochim.
Acta A 2012, 98, 290.
[46] J. Xie, Z. Cheng, W. Yang, H. Liu, W. Zhou, M. Li, Y. Xu, Appl.
Organometal. Chem. 2015, 29, 157.
[31] A. Irving, K. R. Koch, M. Matoetoe, Inorg. Chim. Acta 1993,
206, 193.
[47] B. A. Kellogg, L. Garrett, Y. Kovtun, K. C. Lai, B. Leece, M.
Miller, G. Payne, R. Steeves, K. R. Whiteman, W. Widdison,
Bioconjugate Chem. 2011, 22, 717.
[32] J. Barret, F. S. Deghaidy, Spectrochim. Acta A 1975, 31, 707.
[33] A. Shrivastav, N. K. Singh, S. M. Singh, BioMetals 2003, 16, 311.
[48] H. Dong, T. Hou, Y. Zhao, X. Fu, Y. Li, Comput. Theor. Chem.
2012, 1001, 51.
[34] A. B. P. Lever, Inorganic Electronic Spectroscopy, 2nd ed.,
Elsevier, Amsterdam 1984.
[49] E. Movahedi, A. R. Rezvani, J. Mol. Struct. 2017, 1139, 407.
[50] L. Padmaja, C. Ravikumar, D. Sajan, I. H. Joe, V. S. Jayakumar,
[35] M. Mikami, I. Nakagawa, T. Shimanouchi, Spectrochim. Acta A
1967, 23, 1037.
G. R. Pettit, O. F. Nielsen, J. Raman Spectrosc. 2009, 40, 419.
[51] N. Tidjani‐Rahmouni, N. E. H. Bensiradj, S. Djebbar, O. Benali‐
Baitich, J. Mol. Struct. 2014, 1075, 254.
[36] M. C. Prabahkara, H. S. Bhojya Naik, BioMetals 2008, 21, 675.
[37] M. Hanif, Z. H. Chohan, Appl. Organometal. Chem. 2013, 27, 36.
[52] U. M. Rafi, D. Mahendiran, A. K. Haleel, R. P. Nankar, M.
[38] W. Yang, H. Liu, M. Li, F. Wang, W. Zhou, J. Fan, J. Inorg.
Biochem. 2012, 116, 97.
Doble, A. K. Rahiman, New J. Chem. 2016, 40, 2451.
[39] M. H. Klingele, P. D. W. Boyd, B. Moubaraki, K. S. Murray, S.
Brooker, Eur. J. Inorg. Chem. 2006, 573.
SUPPORTING INFORMATION
[40] W. Hernández, E. Spodine, J. C. Muñoz, L. Beyer, U. Schröder,
Additional Supporting Information may be found online
in the supporting information tab for this article.
J. Ferreira, M. Pavani, Bioinorg. Chem. Appl. 2003, 1, 271.
[41] N. Dharmaraj, Transition Met. Chem. 2001, 26, 105.
[42] T. D. Thangadurai, K. Natarajan, Transition Met. Chem. 2001,
26, 500.
How to cite this article: Rakhshani S, Rezvani
AR, Dušek M, Eigner V. Design and synthesis of
novel thiourea metal complexes with controllable
antibacterial properties. Appl Organometal Chem.
2018;e4342. https://doi.org/10.1002/aoc.4342
[43] F. Khosravi, H. Mansouri‐Torshizi, J. Biomol. Struct. Dyn. 2017,
35, 1.
[44] A. G. Rincón, C. Pulgarin, Appl. Catal. B 2004, 49, 99.
[45] A. Pal, S. O. Pehkonen, L. E. Yu, M. B. Ray, J. Photochem.
Photobiol. A 2007, 186, 335.