Synthesis, characterization and effect of the fluorine substitution on the redox reactivity and in vitro anticancer behaviors of N-polyfluorophenyl-3,5-di- tert-butylsalicylaldimines and their Cu(II) complexes

Article abstract of DOI:10.1016/j.jfluchem.2014.03.011

A series of new polyfluorinated bis(N-C₆F_nH _5-n3,5-^tBu₂salicylaldiminato)Cu(II) [n = 2; 2,4-F₂C₆H₃-3,5-DTBS (1), 2,5-F ₂C₆H₃-3,5-DTBS (2), 2,6-F₂C ₆H₃-3,5-DTBS (3); n = 3; 2,3,4-F₃C ₆H₂-3,5-DTBS (4), n = 4; 2,3,5,6-F₄C ₆H-3,5-DTBS (5), n = 5; 2,3,4,5,6-F₅C₆-3,5-DTBS (6); where 3,5-DTBS is 3,5-^tBu₂salicylaldiminato] with N-polyfluorophenyl-3,5-di-tert-butylsalicylaldi-mines (HL¹-HL ⁶) have been synthesized. Their structure, chemical and electrochemical redox-reactivity as well as cytotoxic activity of these compounds were characterized by analytical, spectroscopic (UV/vis, FT-IR, ¹⁹F NMR and EPR), magnetic, CV techniques and MTT assay. The in situ UV/Vis study have shown that the redox behavior of 1-6 and their Cu-phenoxyl radicals (1^?+-6^?+ and 1^??+- 6^??+) depend on the number and positions of F atoms on the aniline ring as well as of the nature of solvent. The in vitro cytotoxic activity studies of HL¹-HL⁶ and 1-6 against K562 cell lines indicate that the HL¹, HL⁴-HL⁶ and their corresponding complexes (1, 4-6) exhibited higher cytotoxic activity than HL², HL³ and their complexes. Compounds 1, 4 and 5 are more cytotoxic than their respective ligands.

Full text of DOI:10.1016/j.jfluchem.2014.03.011

Journal of Fluorine Chemistry 162 (2014) 78–89
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Synthesis, characterization and effect of the fluorine substitution on
the redox reactivity and in vitro anticancer behaviors of N-
polyfluorophenyl-3,5-di-tert-butylsalicylaldimines and their Cu(II)
complexes
b
c
d
a
V.T. Kasumov a, , F. Su¨zergo¨z , E. Sahin , O. C¸elik , M. Aslanog˘lu
¨
*
a Chemistry Department, Faculty of Science & Art, Harran University, 630190 Sanliurfa, Turkey
b Biology Department, Faculty of Science & Art, Harran University, 630190 Sanliurfa, Turkey
c Department of Chemistry, Ataturk University, Faculty of Science, TR-25240 Erzurum, Turkey
d Physics Department, Faculty of Education, Dicle University, 212800 Diyarbakr, Turkey
A R T I C L E I N F O
A B S T R A C T
A series of new polyfluorinated bis(N-C6FnH5ꢀn3,5-tBu2salicylaldiminato)Cu(II) [n = 2; 2,4-F2C6H3-3,5-
DTBS (1), 2,5-F2C6H3-3,5-DTBS (2), 2,6-F2C6H3-3,5-DTBS (3); n = 3; 2,3,4-F3C6H2-3,5-DTBS (4), n = 4;
2,3,5,6-F4C6H-3,5-DTBS (5), n = 5; 2,3,4,5,6-F5C6-3,5-DTBS (6); where 3,5-DTBS is 3,5-tBu2salicylaldi-
minato] with N-polyfluorophenyl-3,5-di-tert-butylsalicylaldi-mines (HL1–HL6) have been synthesized.
Their structure, chemical and electrochemical redox-reactivity as well as cytotoxic activity of these
compounds were characterized by analytical, spectroscopic (UV/vis, FT-IR, 19F NMR and EPR), magnetic,
CV techniques and MTT assay. The in situ UV/Vis study have shown that the redox behavior of 1–6 and
their Cu–phenoxyl radicals (1ꢁ+–6ꢁ+ and 1ꢁꢁ+–6ꢁꢁ+) depend on the number and positions of F atoms on the
aniline ring as well as of the nature of solvent. The in vitro cytotoxic activity studies of HL1–HL6 and 1–6
against K562 cell lines indicate that the HL1, HL4–HL6 and their corresponding complexes (1, 4–6)
exhibited higher cytotoxic activity than HL2, HL3 and their complexes. Compounds 1, 4 and 5 are more
cytotoxic than their respective ligands.
Article history:
Received 10 November 2013
Received in revised form 25 March 2014
Accepted 26 March 2014
Available online 3 April 2014
Keywords:
Polyfluorinated
Cu(II)–phenoxyl radicals
Spectroscopy
Anticancer
ß 2014 Elsevier B.V. All rights reserved.
1. Introduction
While one of the unique properties of first row transition metal
complexes with non-innocent redox-active phenoxyimine ligands
The design, synthesis and structural characterization of
salicylaldimine complexes bearing bulky tert-butyl groups are a
subject of current interest due to their interesting structural,
spectral, magnetic, catalytic, and redox properties, use as models
for enzymes and various theoretical interests [1–4]. Phenoxyl
radicals have been identified as an essential component of several
metalloenzyme active sites [3]. The discovery of Cu(II)–phenoxyl
radical centers in the active site of Cu(II)-containing enzymes
(galactose and glyoxal oxidases and other metal enzymes) [3,4] has
stimulated the development of various transition metal chelates
with redox-active noninnocent ligands which exhibits the ability
to form the directly coordinated M(II)/(III)–phenoxyl radical
complexes [1–7].
bearing bulky tert-butylated phenol fragments is their ability to
form stable metal(II)/(III)–phenoxyl type ligand-radical complexes
[3–6], the early 4- and 5-group transition metal complexes with
redox-active polyfluorinated phenoxyimine ligands have been
widely used as controlled living olefin polymerization catalysts [8–
18]. Depending on the number and position, a fluorine substitution
not only affects the electronic and structural properties of the
molecule itself but also gives rise to uncommon intermolecular
behavior [19,20]. It is well known that because of fluorine’s atom
strong electron-withdrawing effect and its small size, the presence
of fluorine in a molecule exerts a large influence on physicochemi-
cal properties such as lipophilicity, acidity or basicity, changes
quite drastically the charge distribution in compounds, thereby
altering their biological effects [19–21]. Literature survey revealed
that while the catalytic activities of the polyfluorinated sterically
hindered phenoxyimine ligands of early 4- and 5-group transition
metal complexes are widely studied, the redox reactivity and
biological features of the transition metal complexes are scarcely
*
Corresponding author. Tel.: +90 414 318 3588/+90 414 3440020 1265;
fax: +90 414 3180046/+90 4143440051.
E-mail address: vkasumov@harran.edu.tr (V.T. Kasumov).
http://dx.doi.org/10.1016/j.jfluchem.2014.03.011
0022-1139/ß 2014 Elsevier B.V. All rights reserved.

V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
79
Scheme 1. Chemical structure of the ligands (HL1–HL6) and their bis-ligands copper(II) complexes (1–6).
investigated. The fact that introduction of fluorine atom on
molecule exerts significant changes in its physicochemical and
biological properties, in the course of our interest on the structure
and redox reactivity of the transition metal complexes with di-tert-
butylphenol functionalized salicylaldimines [22,23], we have
decided to study the effects of fluorine atoms on the redox
reactivity and anticancer behavior of the polyfluorinated redox-
active 3,5-di-tertbutylsalicylaldimine ligands and their transition
metal(II) complexes. Since the presented ligands contain sterically
hindered phenol and polyfluorinated aniline fragments, one can
suggest that these ligands would possess both antioxidant and
antitumor efficiency [24]. It is well known that sterically hindered
phenolic compounds along with antioxidant activity also exhibit
anti-inflammatory, antimicrobial, hypolipidemic, antimutagenic
and anticarcinogenic effects [25a–c]. The incorporation of F atoms
on organic compounds also enhances their biological activity,
chemical reactivity, physicochemical properties as well as their
various material properties. In addition, F or CF3 substituent also
improves lipophilicity, and suppresses metabolic detoxification
processes to increase the in vivo lifetime of drugs [25d].
In this context, we wish to report the synthesis, spectroscopic
characterization, crystal structure, chemical redox-reactivity and
antitumor behavior of some N-polyfluorophenyl-3,5-di-tert-butyl-
salicylidene ligands (HL1–HL6) Cu(II) complexes (1–6) (Scheme 1).
The crystal structure and preliminary redox behavior of 4 and HL1–
HL6 ligands and their palladium(II) complexes have recently been
reported by us [23]. The synthesis, IR and 13C NMR characteristics
of HL1–HL6 and other numerous fluorinated salicylaldimines early
have been reported [8–17].
2. Results and discussion
2.1. Structure study
Despite of our efforts to crystallize the fluorinated 1–3 and 6
complexes, no crystals suitable for X-ray measurements were
obtained. The molecular structure of 5 including atom-numbering
is shown in Fig. 1. Crystal data and additional data collection
parameters and refinement details are presented in Table 1.
Selected bond lengths and angles of 5 are listed in Table 2.
Fig. 1. Molecular structures of the two independent complexes ((A) and (B)) in the crystal structure of 5 showing the atom numbering scheme. Thermal ellipsoids of non-
hydrogen atoms are drawn at the 50% probability level. H atoms are presented by arbitrarily small spheres.

80
V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
Table 1
Table 3
˚
Crystal data and structure refinement for compound 5.
Geometrical parameters (A) for hydrogen bonds and short intramolecular contacts.
Formula
C84H88 N4O4F16Cu2
˚
˚
˚
D–Hꢂ ꢂ ꢂA
D–H (A)
D–A (A)
Hꢂ ꢂ ꢂA (A)
D–Hꢂ ꢂ ꢂA (8)
Formula weight
Crystal system
Space group (no.)
1649.09
C1–H15ꢂ ꢂ ꢂF1i
0.936(5)
0.960(8)
0.897(5)
0.930(6)
0.960(4)
0.960(6)
0.960(6)
0.960(7)
0.960(7)
0.960(7)
0.960(8)
0.960(7)
0.930(6)
0.936(5)
0.960(8)
0.960(6)
0.960(6)
0.960(6)
0.960(8)
0.960(6)
0.960(7)
0.960(7)
2.866(7)
3.426(8)
2.862(7)
3.328(7)
2.862(7)
2.830(7)
3.029(7)
3.010(7)
3.021(8)
2.976(7)
3.005(8)
2.991(8)
3.303(7)
3.320(7)
3.421(8)
3.260(7)
3.272(7)
3.673(7)
3.584(7)
3.704(7)
3.828(7)
3.950(7)
2.397(4)
2.912(4)
2.343(5)
2.815(3)
2.296(4)
2.220(5)
2.395(3)
2.378(3)
2.372(4)
2.326(4)
2.367(3)
2.345(4)
2.805(4)
2.668(4)
2.822(4)
2.910(4)
2.779(4)
2.9379
110.80(4)
114.73(5)
116.83(4)
115.91(4)
117.00(3.0)
117.67(3.0)
123.22(4)
123.01(4)
124.49(4)
124.41(4)
123.46(4)
124.11(4)
114.72(4)
127.27(3.0)
121.29(4)
102.79(4)
112.72(4)
134.26
Monoclinic
P121/c1
C77–H77Bꢂ ꢂ ꢂF2
˚
C36–H36ꢂ ꢂ ꢂF8i
a = 23.7519(7) A
˚
C57–H57ꢂ ꢂ ꢂF9i
b = 34.2504(9) A
b=93.392(1)8
˚
3
C19–H19ꢂ ꢂ ꢂF12i
C78–H78ꢂ ꢂ ꢂF16i
C13–H13Bꢂ ꢂ ꢂO1i
C14–H14Bꢂ ꢂ ꢂO1i
C29–H29Cꢂ ꢂ ꢂO2i
C55–H55Bꢂ ꢂ ꢂO3i
C75–H75Cꢂ ꢂ ꢂO4i
C77–H77Cꢂ ꢂ ꢂO4i
C5–H5ꢂ ꢂ ꢂF7ii
c = 10.7424(3) A
˚
Cell volume (A )
Calculated density (Mg/m3) 1.255
8723.8(4)
Cell formula units Z
19
m
(Mo K
a
) (mmꢀ1
)
3.931
T (K)
293(2)
Crystal description
Color
Prism
Red
Crystal sizes (mm)
0.30 ꢃ 0.20 ꢃ 0.15
˚
C15–H15ꢂ ꢂ ꢂF7ii
C30–H30Aꢂ ꢂ ꢂF10iv
C13–H13Bꢂ ꢂ ꢂF7v
C54–H54Bꢂ ꢂ ꢂF15v
C13–H13Bꢂ ꢂ ꢂCg(4)i
C30–H30Cꢂ ꢂ ꢂCg(2)i
C54–H54Cꢂ ꢂ ꢂCg(8)i
C77–H77Cꢂ ꢂ ꢂCg(6)i
C33–H33Cꢂ ꢂ ꢂCg(1)vi
Radiation Mo K
umax (8)
a
(
l
= A)
0.71073
umin
–
2.0–21.6
F(0 0 0)
3416
Index rangers
ꢀ24ꢄhꢄ24, ꢀ35ꢄkꢄ35, ꢀ11ꢄlꢄ10
Reflection collected
74,602
2.8120
138.02
Num. of refined parameters 1008
Num. of reflection used
2.9458
136.83
9664
6972
0.072
0.100
3.0318
141.15
Observed data (I > 2s(I))
R [F2 > 2
3.0696
153.16
s(F2)]
Rw (F2)
Symmetry codes: (i) x, y, z, (ii) x, +y, +z + 1, (iii) x, ꢀy + 1/2, +z ꢀ 1/2, (iv) x, +y, +z ꢀ 1,
(v) x, ꢀy + 1/2, +z + 1/2, (vi) ꢀx, ꢀy, ꢀz.
w = 1/[s
2(Fo2) + (0.0000P)2 + 20.8677P],
where P = (Fo2 + 2Fc2)s/3
Goodness-ofꢀ-fi3t on F2
1.106
0.307
˚
(
(
rD)max
rD)min
(eAꢀ3
)
˚
(eA
)
ꢀ0.318
between C16–C21/C37–C43 phenyl planes in A and between C58–
C63/C79–C84 phenyl planes in B are 30.30(1)8 and 57.56(2)8,
respectively. On the other hand, the dihedral angles between the
two salicylidene planes in A (C1–C6/C22–C27) and in B (C43–C48/
C64–C69) are 75.4 (2)8 and 76.8(2)8, respectively. The dihedral
angles between phenyl/salicylidene planes, C37–C42/C22–C27
[61.10(1)8] and C1–C6/C16–C21 [26.88(1)8] in A, are quite
different. However, the same dihedral angles in B [(C64–C69/
C79–C84 = 29.1/(3)8 and C58–C63/C43–C48 = 31.4(3)8] are close to
each other. The two A and B complexes are linked by two strong
intermolecular C–Hꢂ ꢂ ꢂF hydrogen bonds [(C19–Hꢂ ꢂ ꢂF12–C12
Hydrogen bonds and short intramolecular contacts are given in
Table 3. X-ray diffraction analysis of 5 revealed that there are two
crystallographically independent molecules (A and B) in the
asymmetric unit (Fig. 1). These molecules are mirror images of
each other with virtually identical bond lengths and angles (Table
2). Each Cu(II) center is coordinated by two imine nitrogens and
two phenolate oxygens from two HL5 molecules, adopting a
slightly distorted square planar trans-[CuN2O2] geometry. The
˚
˚
˚
average Cu–O and Cu–N lengths of 1.884(5) and 1.980(4) A in A
(2.296(4) A) and C77–H77ꢂ ꢂ ꢂF2C18 (2.912(4) A] to form a dimer
˚
and B, respectively, correspond well with those reported for
in which the Cuꢂ ꢂ ꢂCu distance is 9.350 A (Fig. 2). Cu–O and Cu–N
nonfluorinated
bis-salicylaldimine
copper(II)
complexes
bond lengths are virtually the same as those in A and B (Table 2).
O–Cu–N bond angles of A differ slightly from 908. The dihedral
angles between the two planes O1–Cu1–N1 and O2–Cu1–N2 in A
and between O3–Cu2–N3 and O4–Cu2–N4 in B are 22.988 and
25.988, respectively. These data compares with 08 for a perfect
square-planar arrangement and 908 for a perfect tetrahedral
arrangement.
[22e,23b,26–29]. The two diagonal O2–Cu1–O1 and N2–Cu1–N1
angles of 160.94(2)8 and 166,76 (2)8, respectively, are less than
1808 as a result of tetrahedral distortion around Cu(II) center.
Similar values were also found for complex B (Table 2). Selected
data for hydrogen-bonding interactions are given in Table 3.
Although the C–O, CH55N, Cu–O and Cu–N bond lengths of A and B
complex molecules are very close to each other, the dihedral angles
In the crystal structure, there are many intramolecular and
intermolecular C–Hꢂ ꢂ ꢂF and C–Hꢂ ꢂ ꢂO type H-bonding interactions
(Table 3). In the crystal packing of 5, there are also a weak
Table 2
intermolecular and intramolecular C–Hꢂ ꢂ ꢂCg (
p-ring) type stack-
ing interactions. These intermolecular interactions are responsible
˚
Selected bond lengths (A), bond angles (8) and torsion angles (8) for compound 5.
Bond lengths
Bond angles
Torsion angle (8)
for constructing of the crystal structure of the complex 5.
Cu1–O1 1.892(4) O2–Cu1–O1 160.94(17) O2–Cu1–N2 –C37
161.2(4)
Cu1–O2 1.877(4) O2–Cu1–N1
Cu1–N1 1.980(4) O1–Cu1–N1
Cu1–N2 1.980(4) O2–Cu1–N2
Cu2–O3 1.888(3) O1–Cu1–N2
90.05(17) O1–Cu1–N1–C16 ꢀ158.1(4)
2.2. Spectroscopic properties of 1–6
91.02(17) N2–Cu1–O1–C1
91.77(17) O2–Cu1–O1–C1
91.52(17) N1–Cu1–O1–C1
150.2(4)
50.3(7)
The analytical, IR, UV/vis and 1H NMR spectroscopic character-
istics and chemical oxidation behaviors of the HLx were previously
reported [8–12, 23c]. The 19F NMR spectroscopic data of HLx and
selected IR data of the complexes 1–6 along with their
interpretation are presented in Section 4. It is worthwhile noting
that the electronic spectra of the polyfluorinated HLx ligands
unlike their non-fluorinated dimethyl [30a] and dimethoxy [30b]
substituted analogs, as well as monosubstituted ortho- and para-(F,
Cl, Br, OCH3, CH3 and (CH3)3C) analogs [30c] do not exhibit any
absorption band even in the concentrated methanolic and
ꢀ42.8(4)
63.1(7)
Cu2–O4 1.884(4) N1–Cu1–N2 166.76(18) O1–Cu1–O2–C22
Cu2–N3 1.982(4) O4–Cu2–O3 158.36(17) N1–Cu1–O2–C22
156.4(4)
ꢀ36.7(4)
161.9(4)
ꢀ56.8(6)
38.9(4)
Cu2–N4 1.981(4) O4–Cu2–N4
C1–O1 1.317(4) O3–Cu2–N4
92.03(16) N2–Cu1–O2–C22
90.85(16) O4–Cu2–N4–C79
91.08(16) O4–Cu2–O3–C43
91.62(16) N4–Cu2–O4–C64
C22–O2 1.305(4) O4–Cu2–N3
C15–N1 1.299(4) O3–Cu2–N3
C36–N2 1.299(4) N4–Cu2–N3 165.09(18) O3–Cu2–O4–C64
ꢀ58.5(6)
C43–O3 1.313(3)
C64–O4 1.315(3)
C57–N3 1.299(4)
C78–N4 1.299(4)
N3–Cu2–O4–C64 ꢀ155.6(4)
O3–Cu2–N3–C58
N4–Cu2–O3–C43 ꢀ154.5(4)
N3–Cu2–O3–C43 40.3(4)
158.8(4)
ethanolic solvents in the
l > 365 nm region [22].

V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
81
of
intensity new bands at 475–560 cmꢀ1 assignable to the
and (Cu–N) [31a], further confirmed the coordination of ligands
n(C–O) +
n(C55C–O) stretching mode [31b], and moderate
n
(Cu–O)
n
through phenolic oxygen and imine nitrogen atoms to the Cu(II)
metal ion.
The electronicspectra of1–6complexesinCHCl3, MeCNand DMF
solutions exhibit very similar two intense doublets in the region
206–365 nm, assignable to intraligand
intense absorption band at 412–425 nm (
p
!
p
* and n !
p* [32,33],
e
= 3.06–4.02 (M cm)ꢀ1
)
(Table 4) and unresolved broad shoulder at 650–670 nm. The band
at 412–425 nm according to its highest molar extinction coefficients
has been assigned to the phenolate-to-Cu(II) ½Oðp Þ !
ꢀ
ꢁ
p
dx
2
CuðIIÞꢅ [3–5] charge transfer (LMCT). The unresolved broad
2
ꢀy
shoulders at 650–670 nm are attributed to convolution of three
h
ꢀ
ꢁ
2
allowed d–d transitions
2B1g ! A1g
d
2 ! d2
;
2B1g
!
2
ꢀ
ꢁ
ꢀ
ꢁꢀy
z
x
2B2g dx 2 ! d
and B1g ! Eg dx 2 ! d2 ꢅ, in a tetrahedrally
2
2
2
2
xy
z
ꢀy
ꢀy
distorted square-planar geometries [33].
Fig. 2. The structure of 5, with displacement ellipsoids drawn at the 50% probability
level for the H-bonded dimers. Dotted lines indicate hydrogen bonds. Molecules A
and B are mirror images.
The room temperature (r.t.) effective magnetic moment (meff) of
1 (1.65
m
B) is lower than that expected for magnetically non
B, suggesting the existence of a
interacting spin only value of 1.73
m
The IR spectra of all compounds 1–6 exhibit sharp bands in the
region of 2860–2950 cmꢀ1 due to the asymmetric and symmetric
weak intermolecular antiferromagnetic interaction between
neighboring molecules of 1 [34]. The meff values found for 2
n
(C–H) stretching vibrations of the C(CH3)3 groups. The strong
bands observed in the range 1602–1616 cmꢀ1 were assigned to
C55N stretching vibrations of the 1–6 complexes [31a]. These
(1.72mB) and 4–6 (1.80–1.86mB) (Section 4.4) are typical for
magnetically non interacting Cu(II) centers in square planar or
slightly distorted square-planar mononuclear geometries [34,35a].
n
bands were shifted to lower frequencies relative to those of free
HL1–HL6 ligands (1622–1630 cmꢀ1), indicating that the ligands are
coordinated through the imines nitrogen atoms to copper(II). A
The r.t. magnetic moment of 3 (2.08mB) is typical for significantly
tetrahedrally distorted Cu(II) complexes [34].
Solid state and solution spectral spin Hamiltonian parameters
of 1–6 complexes are listed in Table 5. The solid-state EPR spectra
of 1–6 at 300 and 173 K are characterized by an axial g tensors with
weak broad feature centered at 2700–2800 cmꢀ1, due to
n(OH) of
the intramolecularly H-bonded OHꢂ ꢂ ꢂN in HL1–HL6, disappears in
the spectra of all the complexes. These observations suggest the
coordination of the ligands via imine nitrogen atom and
deprotonated phenolic oxygen atom to Cu(II). The appearance of
new strong band at 1526–1533 cmꢀ1, attributable to the coupling
gII > g > ge, indicating that the copper site has a d
ground state characteristic for tetrahedral, square-planar or a
or dxy
2
2
?
x
ꢀy
slightly distorted square planar geometries of the Cu(II) ion [35].
No half-field
D
Ms = ꢆ2 signals typical for binuclear triplet-state
species were observed, ruling out the possibility of the dimeric forms.
The EPR spectra of 1, 3 and 4 exhibit one broadened anisotropic signal
(Fig. 3) with gav = 2.110, gav = 2.116 and gav = 2.079, respectively,
which is attributable to dipolar interaction between crystallographi-
cally nonequivalent Cu(II) centers in the solid state [35a]. On the other
hand, the powder spectra of 2, 5 and 6 (Fig. 3c) complexes showed
three well resolved copper(II) hyperfine components in the gII region
(Table 5) at r.t., indicating that the Cu(II) centers are magnetically
diluted in these complexes [23,35].
The r.t. EPR spectra of 1–6 in CHCl3 solution exhibit a typical
four-line pattern without 14N-shfs resolutions on the high-field
components. The cryogenic solution spectra recorded in CHCl3
glasses at 173 of all 1–4 and that of 5 and 6 complexes at 120 K,
Table 4
UV/vis spectral data for 1–6 complexes and their oxidized intermediates.
Complex Solvent
Electronic spectra, l (nm) (log e
(M l)ꢀ1) in CH3CN
1
2
3
4
5
6
CHCl3
MeCN
291(4.43), 412(4.00), 655(2.41)
235, 291, 412, 650*
MeCN + Ox 275, 305*, 325*, 352, 705
DMF
292(4.37), 412(3.94), 680*(2.22)
276, 310*, 325*, 763, 870*
DMF + Ox
CHCl3
MeCN
248(sh), 295(4.70), 322**, 416(4.25), 670*(2.35)
232, 295, 415, 668
MeCN + Ox 279, 305*, 320*, 355, 706
DMF
293(4.25), 415 (4.12), 675*(2.27)
279, 313, 330*, 359, 747
DMF + Ox
CHCl3
MeCN
exhibit usual gII > g > 2.04 and AII ꢇ A pattern consistent with a
250**, 293(4.54), 324**, 412(4.11), 660*(2.20)
236, 293, 412, 650*
?
dx
2
ground state in a tetrahedrally distorted or square-planar
copper(II) centers (Table 5) [35]. The representative frozen glass
2
ꢀy
MeCN + Ox 240, 306*, 325*, 352, 701
DMF
291(3.47), 413(3.06), 680*(2.25)
274, 306*, 325*, 355, 764, 870*
250**, 292(4.82), 337**, 415(4.29), 665*(2.31)
238, 292, 415, 680*
DMF + Ox
CHCl3
MeCN
Table 5
EPR parameters of 1–6 complexes.
MeCN + Ox 282, 307*, 325*, 353, 707
Compex Solid state spec- Solution spectral parameters
DMF
293(4.68), 415(4.19), 680*(2.29)
tral
DMF + Ox
CHCl3
MeCN
276, 306*, 321*, 357, 764, 860*
parameters
269(4.98), 308(4.83), 430(4.18)
232(4.48), 246(4.46), 293(4.53), 417(4.03), 680*(2.20)
gII
g
giso
gII
g
*Aiso *AII
*A
?
?
?
MeCN + Ox 272, 353, 677
1
2
2.189
2.069 2.124 2.237 2.067 66
2.057 2.126 2.238 2.071 68
166 16.0
164 20.0
DMF
292(4.52), 417(4.44), 680*(2.15), 830**(1.69)
2.221,
*AII = 128
2.198
DMF + Ox
CHCl3
MeCN
276, 300**,330**, 356, 737
251(4.59), 293(4.74), 415(4.26), 670*(2.12)
250(4.59), 293(4.74), 415(426), 680*(1.15), 830**(1.66)
3
4
5
2.075 2.124 2.232 2.069 69
2.071 2.116 2.232 2.058 69
2.054 2.117 2.224 2.036 64
168 19.0
158 24.4
165 13.5
2.223
MeCN + Ox 277, 305**, 320**, 352, 707
2.228,
*AII = 165
2.232,
DMF
291(4.52), 415(4.00), 670*(2.19), 830**(1.76)
277, 308*, 325*, 746
DMF + Ox
6
2.047 2.116 2.223 2.041 63
166 11.5
*AII = 154
*
Shoulder.
**
*
Very week shoulder.
A parameters are given in 104 cmꢀ1
.

82
V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
Table 6
Voltammetric data for the compounds 1–6.
Compound
Epa (V)
1.059
Epa (V)
0.131
Epc (V)
Epc (V)
1
0.098
ꢀ0.398
ꢀ1.005
ꢀ0.762
–
2
3
4
5
6
1.048
1.172
1.071
1.134
1.190
0.366
0.672
–
ꢀ0.338
–
–
0.206
–
0.169
–
ꢀ0.322
ꢀ1.062
Supporting electrolyte = 0.05 M Et4NBF4, scan rate = 100 mV/s.
complexes exhibit irreversible oxidation peak within 1.050–
1.119 V, which can be assigned to the ligand centered Cu(II)–
phenoxide/Cu(II)–phenoxyl couples. Cyclic voltammogram of 1
exhibits an irreversible oxidation peak at 1.059 V (Fig. 5a). The
irreversibility suggests the instability of the generated radical
species under electrochemical time scale. Interestingly, some more
oxidation and reduction waves were generated in the second scan
as shown in (Fig. 5b). Complex 1 exhibits an oxidation at 0.131 V
and a corresponding reduction peak at 0.098 V. The separation in
the peak potentials,
DEp, is 33 mV indicating a quasi-reversible
two electron transfer process which is assigned to the oxidation/
reduction of imine moiety of the ligand [37a–c]. Similar
electrochemical behavior also was detected for complex
5
Fig. 3. The powder EPR spectra of 2 (a) at 300 K and (b) 173 K and 6 (c) at 300 K.
(Table 6 and Fig. 6). Interestingly, the anodic peak for 1–6
complexes was shifted to more positive values as the number of
electron-withdrawing F atoms increased. On the other hand, there
is no correlation between the number F atoms and the cathodic
peaks values (ꢀ0.322/ꢀ1.062 V) which are assigned to the
reduction of copper(II) to copper(I) (Table 6). The presented
electrochemical study indicates that all anodic oxidation processes
are ligand centered. These results suggest that the observed
chemical oxidation in 1–6 complexes are ligands centered. The Ip/
v1/2 value is almost constant for all scan rates. This establishes the
electrode process as diffusion controlled [37d].
spectra for 1, 3 and 6 complexes are given in Fig. 4. The gII and AII
values of these complexes are very similar to each other, indicating
that their molecular geometry is governed by the steric effect of
salicylaldimine tBu2 groups. Except 4 (G = 4.12), the exchange
interaction parameter, G = (gII–ge)/(g –ge), values (3.43–3.66) for
?
1–6 complexes being <4 suggest the existence a small exchange
coupling between CuII centers [35]. Another convenient empirical
index of tetrahedral distortion, f(a) = gII/AII, is regarded as an extent
of deviation from idealized planar geometry [36]. The obtained
ratio gII/AII for all frozen glass 1–6 samples (133–141 cm) falls
within the range 130–150 cm and indicate the existence of a
slightly tetrahedral distorted geometry around Cu(II) centers in the
above complexes [36].
2.4. Chemical oxidations of 1–6
The chemical oxidation properties of 1–6 were investigated by
in situ UV/vis spectral measurements in MeCN and DMF. The
oxidation of the complexes was carried out with one and two
equivalent molar amounts (NH4)2Ce(NO3)6 (CAN). Addition of an
equiv. amount of CAN to 1–6 in above solvents at r.t. caused a color
change from dark brown to green. The spectral changes for 1–
6 + CAN system were monitored by UV/vis spectroscopy during the
progress of their oxidation in CH3CN and DMF solutions. The
2.3. Electrochemistry
Electrochemical properties of copper complexes were studied
on a Pt disc electrode in acetonitrile containing 0.05 M Et4NBF4 as
the supporting electrolyte. The main electrochemical data of 1–6
complexes are given in Table 6. As can be seen from Table 6, all
Fig. 4. Frozen glass EPR spectra of 1 (a), 3 (b) (173 K) and 5 (c) (120 K) in CHCl3.

V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
83
Fig. 5. The cyclic voltammograms of complex 1 in acetonitrile containing 0.05 M Et4NBF4 as supporting electrolyte. Scan rate 100 mV/s. First scan (a) and second scan (b).
examination revealed that the redox behavior of complexes 1–6 is
distinctively different from their non-fluorinated dimethyl [22d]
and dimethoxy [23c] substituted analogues and mono substituted
(CH3, CH3O, F, Cl, Br) bis(N-X-Ph-3,5-di-tert-butylsalicylaldimi-
nato)-Cu(II) complexes [30c]. On the other hand, the redox feature
of the radical intermediates generated from these complexes by
using two equiv. of CAN in CH3CN and DMF are different from each
others. Upon one-electronoxidation of 1–6 complexes by one or two
equiv molar amounts of CAN in MeCN at r.t and aerobic conditions,
along with instant disappearance of d–d bands at about 660 nm and
LMCT absorptions around 412–415 nm region of the parent
complexes, the appearance of a new asymmetric or symmetric
broad maximum bands centered within 350–360, 701–707 and
730–770 nm regions, assignable to Cu(II)/Cu(III)–phenoxyl radical
species, were observed (Table 4) [2–5,38]. The UV/vis spectral
changes accompanied by the oxidations of 1–6 are shown in Figs. 7–
10. For all 1–6 complexes, under above experimental conditions, the
decrease in intensities and blue or red shifts in lmax of the first
detected bands during the 3–5 successive scanning with a period of
50 s were observed. Upon further scanning, the decrease in the
intensity of 701–707 nm (MeCN) and 730–765 nm (DMF) bands for
all oxidized species 1+–6+ꢁ practically stopped when the oxidation is
carried out with one or two equiv of CAN in MeCN and in the case of
the oxidation by one equiv of CAN in DMF (Fig. 7). The energy and
shoulder about 870 nm were detected (Figs. 8 and 9 and Table 4).
Generally, after the fourth/fifth successive scans, the intensity of
new band starts to increase gradually upon further scans of the
oxidized reaction mixture and then after about 60–80 min its
absorbance again begins to decrease. For example, addition of 2
equiv of CAN to solution 1 (1.14 ꢃ 10ꢀ2 M) in DMF at r.t., a new
asymmetric broad maximum at 765 nm with a shoulder at ca
870 nm appears in the spectra of the reaction mixture (Fig. 8A).
Upon further successive scans witha period 50 s, the intensity of this
band rapidly decreases from 765(2.88), 718(1.35), 700(1.21) to
695(1.21) and then, at forth scan a new band begins to grow at
698(1.39) nm and upon further scanning its intensity is increased to
709(1.99)nm(Fig. 8A). Asimilarspectralbehaviorwasalsoobserved
in the oxidation of 3 (Fig. 8B) and 4 (Fig. 9) complexes by two equiv
CAN under the same experimental conditions. In the chemical
oxidation of complexes 5 and 6 with 2 equiv CAN in DMF, together
with disappearance of their d–d bands, the appearance of new
absorptions at 737(2.69) and 746(3.40), respectively, were detected.
Upon three successive scans, the intensity of these bands rapidly
decreased from 737(2.69) to 750(0.61) nm and from 746(3.40) to
758(0.37) nm, respectively, for 5+ (Fig. 10a) and 6+ (Fig. 10b).
However, after fourth scan with a period of 50 s, the appearance of
new weak shoulder band at about 630 nm for both complexes was
observed. These bands can be assigned to non radical ligand Cu(II)
complexes.
intensity of these features are consistent with the
p
!
p*
transitions associated with a bisphenoxyl radical species [38].
These bands can be detected even after 24 h under aerobic
conditions. However, when the oxidation of complexes 1–4 is
carried out with two equiv of CAN in DMF, along with above spectral
changes, after third or fourth scan the appearance of a new broad
maximumbandwitha slowly growing intensityatca. 730 nmwitha
To the best of our knowledge, the observed spectral changes
in the chemical oxidation of polyfluorinated Cu(II)–phenoxyl
systems by two equivalent of Ce(IV) in DMF is unusual
phenomenon. Unfortunately, while there are a wide variety
chemical or electrochemical oxidation studies of various
transitions metal complexes with di-tert-butylphenol function-
Fig. 6. The cyclic voltammograms of complex 5 in acetonitrile containing 0.05 M Et4NBF4 as supporting electrolyte. (a) Scan rate 100 mV/s (first scan) and (b) Scan rates
increasing from100 mV/s to 250 mV/s.

84
V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
Fig. 7. UV–vis spectral changes in during one-electron oxidation of 1 (7.5 ꢃ 10ꢀ3 M, (A (1)) and 3 (2.5 ꢃ 10ꢀ3 M, (B (1) by one equiv of CAN at r.t. in DMF. Spectral changes for
1
ꢁꢁ+ (A) and 3ꢁꢁ+ (B): In both figure number 1 denote spectrum of parent complexes. The numbers 2, 3, 4 are spectra of 1(3) +1 equiv CAN systems scanned with a period 50 s.
Spectrum 5 in A and B recorded after 20 h.
alized bi-, tri- and tetradendate non-fluorinated salicylaldimine
ligand systems [1–7], we could not find any report in literature
similar to spectral changes observed in our study. The fact that
secondary spectral changes do not appear in non-fluorinated
various di-CH3 and di-CH3O substituted N-XPh-3,5-DTBS ligands
Cu(II) complexes [22d,23c], we suggest that the observed
unusual redox behavior may be originates from fluorine atoms.
Thus, the redox-reactivity of the polyfluorinated sterically
hindered di-tert-butylphenol functionalized 1–6 complexes and
reactivity of the generated Cu–phenoxyl radicals depends on the
number and positions of the fluorine atoms in the aniline ring
and solvent nature. The observed spectral changes originate
from the conversions of the initially generated Cu-phenoxy
radical species to another secondary phenoxy radical species.
Fig. 8. UV/vis spectral changes during one-electron oxidation of 1 (a) and 3 (b) by two equiv of CAN in DMF solution at r.t. In both, Fig. 1 denotes spectrum of neutral
complexes. The intensity of the spectra of 1/3 +2 equiv CAN systems rapidly decreased from 2 to 4 and then upon further scanning with a period 50 s, the appearance of a new
spectra which the intensities increase from 5 to 15 (a) and from 5 to 12 (b) were detected. The arrow indicates the direction of the increasing absorbance.

V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
85
Fig. 9. Visible spectrum of 4 (1.2 ꢃ 10ꢀ2 M) (a) and spectral changes of 4 + 2 equiv CAN mixture; (b) the 1, 2, 3 and 4 spectra are subsequently scanned in the progress of the
chemical oxidation via of 50 s; the spectra 5–12 are scanned with a period of 2–10 min; (c) spectrum is scanned after 15 h; (d) spectrum of the dilute solution sample of (c).
2.5. Cytotoxic activity
monofluorinated HL7 (p-F) and HL8 (o-F) revealed that their IC50
data (Table 7) significantly affected by the positions of F atoms
on the aniline ring. Based on the IC50 values, pentafluorinated
HL6 ligand was found to be the most potent cytotoxic one. In
addition, according to IC50 value of 5-FU and HL3 the cytotoxic
influence on the K562 cell lines is negligible. Comparison of IC50
values of above mentioned ligands strikingly revealed that, in
generally, upon increase in the numbers of F atoms in the aniline
rings the cytotoxic efficiency of the fluorinated ligands also
increases. But in some cases there is a deviation from this trend.
It is noteworthy that for 2,4-F2Ph-3-tert-butylsalicylaldimine
and 2,4-F2Ph-5-tert-butylsalicylaldimine compounds, which
contains only one tert-butyl substituent on the salicylaldimine
The cytotoxic effects of ligands HL1–HL6, and their correspond-
ing Cu(II) complexes 1–6 were assessed by MTT [(3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)] assay.
K562 cells were treated for 72 h with increasing doses of the
ligands and 5-Fluorouracil (5-FU). Upon assessment of viability,
IC50 (micromolar concentration inhibiting 50% cell growth)
values were calculated for compounds HL1 (2,4-F2), HL2 (2,5-F2),
HL3 (2,6-F2), HL4 (2,3,4-F3), HL5 (2,3,5,6-F) and HL6 (2,3,4,5,6-
F5). The IC50 values for these ligands were found as 8.78, 91,
>100, 2.81, 3.73 and 0.04
conditions for reference 5-FU drug, the IC50 value was found to
be as >100 M. The compactions of cytotocxic activity data
between difluorinated HL1, HL2 and HL3, as well as between
mM, respectively. Under the same
m
moiety, the evaluated IC50 values were found to be about 16
mM.
On the other hand, the analogous non-fluorinated 3,5-di-tert-
Fig. 10. UV–vis spectral changes in the progress of the one-electron chemical oxidation of 5 (1.26 ꢃ 10ꢀ2 M, (A (1)) and 6 (9.88 ꢃ 10ꢀ3 M, (B (1) by two equiv CAN in DMF at r.t.
Spectral changes for 5ꢁꢁ+ (A) and 6ꢁꢁ+ (B) 2, 3 and 4 are 1st, 2nd and 3rd scans of oxidized 5 and 6, respectively, with a period of 50 s; 5-diluted spectrum of 6ꢁꢁ+. 6-diluted
solution spectrum of 6. Inset: Spectrum of the dilute solution samples of 5ꢁ+
.

86
V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
Table 7
Chemical structure and IC50 data for ligands and their Cu(II) complexes.
Chemical structure
HLx
HL1
IC50
(
m
M)
Chemical structure
Cu(Lx)2
IC50
(
m
M)
8.78
1
1.27
HL2
91
2
3
>100
HL3
>100
90
HL4
2.81
4
5
2.8
HL5
3.73
1.12
HL6
HL7
HL8
0.04
6
7
8
3.21
348
105
30
75
HL9
745
HL10
703

V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
87
butylsalicylaldimine ligands having various di-CH3 (2,3-, 3,4-,
2,5-difluouroaniline, 2,6-difluouroaniline, 2,3,4-trifluouroaniline,
2,5- and 3,5-di CH3), 2-CH3, 4-CH3 and 4-CH3O substituents on
2,3,5,6-tetrafluouroaniline
and
2,3,4,5,6-pentafluouroaniline)
the aniline rings exhibited negligible (IC50 = 568–2078
cytotoxic effects under the same conditions (Table
m
M)
7).
reagents were purchased from Sigma-Aldrich. 3,5-tBu2-salicylalde-
hyde was prepared according to a published procedure [40].
These results demonstrate that the observed cytotoxic effects
of HLx Schiff bases originate from the number and positions of
fluorine atoms on the aniline rings. In order to investigate the
influence of the complexation of the ligands with Cu(II) on
their cytotoxic effects under above conditions, for 1–6 com-
plexes, the IC50 values were evaluated. The obtained values of
IC50 for complexes 1–6 were found to be as 1.27, >100, 90, 2.8,
4.2. Instrumentation
The C, H, N elemental analyses were performed on a LECO
CHNS-932 model analyzer. UV/vis spectra were measured on a
Perkin-Elmer Lambda 25 spectrometer operating between 200 and
1100 nm. IR spectra were recorded as KBr pellets on a Perkin-Elmer
FTIR spectrometer in the 450–4000 cmꢀ1 region. The 19F NMR
measurements were performed on an Agilent–NMR–VNMRS
400 MHz spectrometer operating at 376 MHz. in CDCl3. 19F NMR
chemical shifts were determined relative to CDCl3 as the external
standard and low field is positive. EPR spectra were recorded on a
Varian model E 109C spectrometer in X-band with 100 kHz
modulation frequencies. The g-values were determined by
1.12 and 3.21
mM, respectively. These data indicate that the
complexation of the HL1, HL4, and HL5 ligands with Cu(II)
increase their cytotoxic activities on the K562 cells. Thus, some
polyfluorinated redox-active salicyilaldimine ligands and their
Cu(II) complexes showed strong cytotoxic effects in comparison
with that of 5-FU which showed value IC50 > 100 against K562
cell lines. While in many cases, it has been found that the
biological activity of ligands is increases in their complexation
with metal ions [39], in our case a significant increases in the
cytotoxic effects for some coordinated ligands have been
observed.
comparison with
a 2,2-diphenyl-1-picrylhidrazyl (DPPH) of
g = 2.0036. The room temperature magnetic susceptibility was
measured by using a Sherwood Scientific magnetic balance and the
diamagnetic corrections were evaluated from Pascal’s constants
[41]. The phenoxyl radical species of the complexes were
generated in situ by adding an equimolar amount (1 or 2 equiv
3. Conclusions
amount) of (NH4)2[Ce(NO3)6] (CAN) (2,5–5.0 ꢃ 10ꢀ3 M) to
a
A new series of polyfluorinated Cu(II) complexes with N-
polyfluorophenyl-3,5-di-tert-butylsalicylaldimine ligands have
been prepared and their structure characterized by various
spectroscopic, magnetic and X-ray diffraction techniques. UV/vis
and EPR spectral studies along with magnetic moment data
suggest slightly distorted square-planar geometry for all com-
plexes. X-ray study reveals that the asymmetric unit of 5 contains
two crystallographically independent molecules. UV/vis study
reveals that the redox properties of the polyfluorinatrd 3,5-di-tert-
butylated Schiff base Cu(II) complexes are greatly affected by the
number and the location of the fluorine atom(s) in the phenyl rings.
The chemical oxidation of all complexes with one and two equiv of
CAN in MeCN or with one equiv of CAN in DMF, leads to the
immediate disappearance of d–d bands and the appearance a
broad strong band at 730–760 with a shoulder around 850 nm
which are assigned to Cu(II)–phenoxyl radical complexes. The
intensity of these bands within 2–3 min decreased about %70–80
and upon further scanning their intensity practically remained
constant. Unexpectedly, upon chemical oxidation of 1–4 with two
equiv of CAN in DMF at r.t. along with above mentioned spectral
changes, the appearance of a new bands at arꢀo1und 715–740 nm,
solution (DMF or MeCN) of corresponding 1–6 in a UV cell (1-
cm path length, with a silicon cap) at r.t. under aerobic conditions.
An EcoChemie Autolab-12 potentiostat with the electrochemical
software package GPES 4.9 (Utrecht, The Netherlands) was used for
voltammetric measurements. A platinum disk (2 mm o.d.) was
employed as a working electrode, a platinum coil as a counter
electrode, and an Ag/AgCl as reference electrode, respectively. All
measurements were performed in MeCN containing 0.05 M
Et4NBF4 as a supporting electrolyte at room temperature (r.t.)
and under nitrogen atmosphere.
4.3. Synthesis of ligands
N-(3,5-di-tert-butylsalicylidene)-2,4-difluoroaniline (HL1), N-
(3,5-di-tert-butylsalicyl-idene)-2,5-difluoroaniline (HL2), N-(3,5-
di-tert-butylsalicylidene)-2,6-difluoroaniline (HL3), N-(3,5-di-tert-
butylsalicylidene)-2,3,4-trifluoroaniline (HL4), N-(3,5-di-tert-butyl-
salicylide-ne)-2,3,5,6-tetrafluoroaniline (HL5) and N-(3,5-di-tert-
butylsalicylidene)-2,3,4,5,6-penta-fluoroaniline (HL6) ligands were
synthesized according to literature procedures [9,13,23b].
The IR, UV/vis and 1H NMR spectral data as well as redox-
reactivity behaviors of HLx ligands have been presented in recent
works [12–17,23]. As the 19F NMR spectra of HLx ligands have not
been reported, we present herein fluorine-19 NMR spectral data for
above ligands. Analysis of the 19F NMR spectral data of HLx ligands
revealed the existence of 19F-19F and 19F–1H interactions in these
assignable to Cu(III)(phenoxyl)(phenolat)NO3
radical species,
were observed. Upon oxidation of 5 and 6 with two equiv of CAN in
DMF, the spectra of the generated radicals after 4 scans gradually
converted to another spectra typical for nonradical Cu(II) com-
plexes. The electrochemical oxidation of 1–6 revealed that these
complexes possess ligand centered oxidation in the region 1.0–
1.2 V. In vitro study of cytotoxic activities of the HL1–HL6 and their
complexes against K562 cell lines revealed that some of these
compounds. The 19F NMR spectrum of HL1: 19F NMR (376 MHz,
3
CDCl3)
d
ꢀ112.4 to ꢀ112.5 ppm (m, JFF
,
3JFH = 2.8–7.2 Hz, 1F),
ꢀ120.9 to ꢀ121.0 ppm (m, 3JFF, 3JFH = 3.2–8.0 Hz, 1F). HL2: 19F NMR
3
compounds possess higher (IC50 = 0.04–8.78
m
M for HLx and 1.12–
(376 MHz, CDCl3)
d
ꢀ117.6 to ꢀ117.7 ppm (m, JFH
,
3JFF = 2.3–
3
3.21
m
M for 1, 4, 5, 6) cytotoxic activity. It has been found that as
8.0 Hz,1F), ꢀ131.3 to ꢀ131.4 ppm (m, JFH
,
3JFF = 2.8–6.8 Hz, 1F).
the number of F atoms increases in HLx their IC50 values decrease.
HL3: 19F NMR (376 MHz, CDCl3)
d
ꢀ123.3 ppm (t, JFH
,
3JFF = 7.2,
3
7.2 Hz, 2F). HL4: 19F NMR (376 MHz, CDCl3)
d
ꢀ136.6 to
3
4. Experimental
ꢀ136.7 ppm (m, JFH
,
3JFF = 2.8–6.0 Hz, 1F), ꢀ145.4 to 145.6 ppm
(dt, 1F) 3JFH
,
3JFF = 2.8–8.0 Hz 30.4 Hz), ꢀ158.5 to ꢀ158.6 ppm (td,
4.1. Materials
3JFH
,
3JFF = 4.4, 20.4 Hz, 1F). HL5: 19F NMR (376 MHz, CDCl3)
d
3
ꢀ139.3 to ꢀ139.6 ppm (q, JFH
,
3JFF = 6.8–12.8 Hz, 2F), ꢀ152.5 to
3
All reagents and solvents were obtained commercially and used
without further purification. All 2,4-di-tert-butylphenol, hexameth-
ylenetetramine, copper(II)acetate monohydrate, (NH4)2Ce(NO3)6
(CAN), acetic acid and all fluorinated anilines (2,4-difluouroaniline,
ꢀ152.6 ppm (m, JFH
,
3JFF = 2.8–8.8 Hz, 2F). HL6: 19F NMR,
3
(376 MHz, CDCl3)
d
ꢀ152.2 to ꢀ52.3 ppm (dd, JFF = 7.2, 17.0 Hz,
2F), ꢀ158.9 to ꢀ159.0 ppm (t, 3JFF = 23.2, 21.6 Hz, 1F), ꢀ162.55 to
3
ꢀ162.7 ppm (td, JFF = 7.2, 5.6, 5.6 Hz, 2F).

88
V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
4.4. Synthesis of 1–6 copper(II) complexes
10 min, and resuspended in fresh medium at a cell density 3 ꢃ 104
cells/ml. The cell suspension was seeded per well in 96-well
microplate in triplicate order. Each well was cultured at 37 8C
under 5% CO2 in RPMI1640 medium supplemented with %10 fetal
calf serum (FCS) 72 h with 5-Fluorouracil (5-FU) or ligands (HL1–
HL6) and their respective Cu(II) complexes (1–6) that the
Complexes 1–6 were prepared using following general proce-
dure. To a stirring solution of the ligand (1 mmol) in MeOH (50–
60 ml) at ca. 60 8C a solution of copper acetate monohydrate
(0.5 mmol, 0.1 g) in MeOH (10 ml) was added. The resulting dark
green solution was refluxed with stirring for 1.0–1.5 h and then
was concentrated to ca. 30 ml. After cooling to r.t. the formed
microcrystalline solid was filtered off, washed with methanol and
dried in air. The copper(II) complexes were recrystallized from
chloroform/methanol mixture (yields 68–85%). Ligand HL4 and its
complex 4 are recently reported by us [23b]. However, because
their detailed redox reactivity and anticancer behavior have not
been reported, for comparative properties, these compounds have
been included in present study.
concentrations ranged from 31.25
dilutions. Then MTT solution (diluted phosphate buffered saline,
5 mg/ml, 10 l) was added to each well and incubated for 4 h. The
formazan crystals were dissolved by adding DMSO (100 l) to each
mM to 1000 mM as serial 1:2
m
m
well after centrifuged for 10 min under 1800 rpm, and the
absorbance valued was measured at 570 nm.
Acknowledgments
Harra University Research Fund (HUBAK, Project Number 712)
and the Scientific and Technological Research Council of Turkey
(TUBITAK, Project Number: ARDEB-113Z134) are kindly acknowl-
edged for financial support.
1. Yield 68%. M.p. > 280 8C. Anal. Calc. for C42H48N2O2F4Cu: C,
67.05; H, 6.43; N, 3.72%. Found: C, 66.81; H, 6.53; N, 3.48%. IR (
cmꢀ1, KBr): (C–H) of tBu 2867–2963;
(C55N) 1618; new strong
band at 1525 assigned to (C–O) + (C55C–O); (C–O) 1324;
(Cu–N) 536, 568; (Cu–O) 449, 492. eff = 1.69mB at r.t.
2. Yield 77%. M.p. 231 8C. Anal. Calc. for C42H48N2O2F4Cu: C, 67.05;
H, 6.43; N, 3.72%. Found: C, 66.95; H, 6.32; N, 3.68%. IR (
cmꢀ1
(C55N) 1618; new band 1533
(Cu–N) 446, 498; (Cu–O) 536,
n
n
n
n
n
n
n
n
m
n
,
Appendix A. Supplementary data
t
KBr):
assigned to
568. eff = 1.72mB at r.t.
3. Yield 77%. M.p. > 280 8C. Anal. Calc. for C42H48N2O2F4Cu: C,
67.05; H, 6.43; N, 3.72%. Found: C, 66.85; H, 6.59; N, 3.62%. IR (
cmꢀ1, KBr): (C–H) of tBu 2867–2964;
(C55N) 1613; new band
at 1525 assigned to (C–O) + (C55C–O); (Cu–N) 506, 497;
(Cu–O) 534, 519. eff = 2.08mB at r.t.
4. Yield 67%. M.p. 276 8C. Anal. Calc. for C42H44N2O2F8Cu: C, 61.19;
H, 5.38; N, 3.40%. Found: C, 60.97; H, 4.98; N, 3.32%. IR (
cmꢀ1
KBr): (C55N) 1615; new band at 1531
(C–H) of tBu 2870–2958;
assigned to (C–O) + (C55C–O); (Cu–N) 540, 487; (Cu–O) 595,
576 eff = 1.86mB at r.t.
5. Yield 67%. M.p. 252 8C. Anal. Calc. for C42H42N2O2 F10Cu: C, 58.64;
H, 4.92; N, 3.25. Found: C, 58.90 H, 4.75; N, 3.06%. IR (
cmꢀ1
KBr): (C55N) 1614; new band at
(C–H) of tBu 2870–2959;
1531assigned to (C–O) + (C = C–O); (Cu–N) 540, 487; (Cu–
O) 595, 576. eff = 1.80mB at r.t.
n
(C–H) of Bu 2867–2963;
n
n
(C–O) + (C55C–O);
n
n
n
Supplementary material related to this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.jfluchem.2014.03.011.
m
n
References
n
n
n
n
n
[1] C.G. Pierpont, C.W. Lange, in: K.D. Karlin (Ed.), Prog. Inorg. Chem. 41 (1994) 331–
442.
n
m
[2] P. Chaudhuri, K. Wieghardt, Prog. Inorg. Chem. 50 (2001) 151–216.
[3] B.A. Jazdzewski, W.B. Tolman, Coord. Chem. Rev. 200 (2000) 633–685.
[4] J.W. Whittaker, Chem. Rev. 103 (2003) 2347–2364 (and references therein).
[5] J. Stubbe, W.A. van der Donk, Chem. Rev. 98 (1998) 705–762.
[6] V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283–315.
[7] L. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85–93.
[8] H. Makio, N. Kashiwa, T. Fujita, Adv. Synth. Catal. 344 (2002) 477–493.
[9] R. Furuyama, J. Saito, S. Ishii, H. Makio, M. Mitani, H. Tanaka, T. Fujita, J.
Organomet. Chem. 690 (2005) 4398–4413.
[10] H. Makio, H. Terao, I. Washita, T. Fujita, Chem. Rev. 111 (2011) 2363–2449.
[11] A.F. Mason, G.W. Coates, J. Am. Chem. Soc. 126 (2004) 16326–16327.
[12] A. Sakuma, M.S. Weiser, T. Fujıta, Polym. J. 39 (2007) 193–207.
[13] M. Mitani, R. Furuyama, J. Mohri, J. Saito, S. Ishii, H. Terao, T. Nakano, H. Tanaka, T.
Fujita, J. Am. Chem. Soc. 125 (2003) 4293–4305.
n
,
n
n
n
n
n
n
m
n
,
n
n
n
n
n
n
m
[14] (a) T. Matsugi, T. Fujita, Chem. Soc. Rev. 37 (2008) 1264–1277;
(b) A.F. Mason, G.W. Coates, J. Am. Chem. Soc. 126 (2004) 16326–16327;
(c) D.P. Song, Y.G. Li, R. Lu, N.H. Hu, Y.S. Li, Appl. Organ. Chem. 22 (2008) 330–340.
[15] K. Kawai, T. Fujita, Top. Organomet. Chem. 26 (2009) 3–46.
[16] T. Matsugi, T. Fujita, Chem. Soc. Rev. 37 (2008) 1264–1277.
[17] M. Mitani, J. Mohri, Y. Yoshida, J. Saito, S. Ishii, K. Tsuru, S. Matsui, R. Furuyama, T.
Nakano, T. Fujita, J. Am. Chem. Soc. 124 (2002) 3327–3336.
[18] (a) N. Iwasa, S. Katao, J. Liu, M. Fujiki, Y. Furukawa, K. Nomura, Organometallics
28 (2009) 2179–2187;
4.5. X-ray structural determination of 5
Single crystalX-raydiffraction was carried out at 293(2)K usinga
Bruker APEX-II CCD diffractometer. The intensity data were
˚
collected using graphite monochromated Mo K
a (l = 0.71073 A)
radiation. The lattice parameters were determined by the least-
squares methods on the basis of all reflections with F2 > 2 (F2).
(b) S. Chen, C. Guo, L. Liu, H. Xu, J. Dong, Y. Hu, Polymer 46 (2005) 11093–11098;
(c) S. Reinartz, A.F. Mason, E.B. Lobkovsky, G.W. Coates, Organometallics 22
(2003) 2542–2544.
s
Integration of the intensities, correction for Lorentz and polarization
effects and cell refinement was performed using Bruker AXS APEX II
software suit [42]. The structures were solved by direct methods
using SHELXS-97 [43] and refined by a full-matrix least-squares
procedure using the program SHELXL-97 [43]. H atoms were
positioned geometrically and refined using a riding model. The final
difference Fourier maps showed no peaks of chemical significance.
The molecular structure plots were prepared using ORTEP III [44].
The crystal and structural data for complex 5 are given in Table 1.
[19] T. Hiyama, in: H. Yamamoto (Ed.), Organofluorine Compounds. Chemistry and
Application, Springer, Berlin, 2000, pp. 137–182.
[20] (a) P. Jeschke, Chem. Biol. Chem. 5 (2004) 570–589;
(b) R.K. Filler, Y. Kobayashi (Eds.), Biomedical Aspects of Fluorine Chemistry,
Kodama Ltd, Elsevier, Tokyo, Amsterdam, 1982;
(c) R. Li, R.L. Smith, H.I. Kentta¨maa, J. Am. Chem. Soc. 118 (1996) 5056–5061.
[21] (a) D.B. Harper, D. O’Hagan, Nat. Prod. Rep. 11 (1994) 123–133;
(b) T. Tsuzuki, N. Shirasawa, T. Suzuki, S. Tokita, Adv. Mater. 15 (2003)
1455–1458;
(c) G. Theodoridis, Advances in Fluorine Science, 2, Elsevier, Amsterdam, 2006,
pp. 121–175;
(d) D. Chopr, T.N.G. Row, Cryst. Eng. Commun. 13 (2011) 2175–2186.
[22] (a) A.A. Medzhdov, V.T. Kasumov, H.S. Mamedov, Koord. Khim. 7 (1981) 66–71;
(b) V.T. Kasumov, A.A. Medzhidov, Russ. J. Coord. Chem. 21 (1995) 783–787;
(c) V.T. Kasumov, F. Ko¨ksal, Y. Zeren, Spectrochim. Acta, A: Mol. Biomol. Spec-
trosc. 63 (2006) 330–336.
[23] (a) O. Celik, V.T. Kasumov, E. Sahin, Acta Crystallogr. E65 (2009) o2786;
(b) V.T. Kasumov, I. Ucar, A. Bulut, J. Fluorine Chem. 131 (2010) 59–65;
(c) V.T. Kasumov, E. Sahin, Spectrochim. Acta, A: Mol. Biomol. Spectrosc. 107
(2013) 31–38.
4.6. K562 cell line and proliferation assay
MTT [(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide)] assay were used for IC50 determination. K562 (human
chronic myelogenous leukemia) cells in exponential growth stage
were harvested from culture and centrifuging at 1800 rpm for

V.T. Kasumov et al. / Journal of Fluorine Chemistry 162 (2014) 78–89
89
[24] (a) M.D. Mulican, M.W. Wilson, D.T. Conner, C.R. Kostlan, D.J. Schrier, R.D. Dyer, J.
Med. Chem. 36 (1993) 1090–1099;
[34] (a) A. Earnshaw, Introduction to Magneto Chemistry, Academic Press, London
and New York, NY, 1968;
(b) I. Leventis, I. Andreadou, A. Papalois, I. Sfiniadakis, V. Gorgoulis, D. Korkolis, D.
Hadjipavlou-Litina, P. Kourounakis, C. Fotiadis, In Vivo 18 (2004) 161–169;
(c) V. Weber, C. Rubat, E. Duroux, C. Lartigue, M. Madesclaire, P. Coudert, Bioorg.
Med. Chem. 13 (2005) 4552–4564.
(b) W.E. Hatfield, J.A. Barnes, D.Y. Jeter, R. Whyman, E.R. Jones Jr., J. Am. Chem.
Soc. 92 (1970) 4982–4984;
(c) W.E. Marsh, K.C. Patel, W.E. Hatfield, D.J. Hodgson, Inorg. Chem. 22 (1983)
515–544;
[25] (a) T. Maki, Y. Araki, Y. Ishida, O. Onomura, Y. Matsumura, J. Am. Chem. Soc. 123
(2001) 3371–3372 (and references therein);
(d) E. Colacio, M. Ghazi, R. Kiveka1s, J.M. Moreno, Inorg. Chem. 39 (2000) 2882–
2890.
(b) M. Roberti, D. Pizzirani, M. Roberti, D. Pizzirani, D. Simoni, R. Baruchello, C.
Bonora, F. Buscemi, S. Grimaudo, M. Tolomeo, J. Med. Chem. 46 (2003) 3546–
3554;
(c) K. Azuma, M. Nakayama, M. Koshioka, Ki Ippoushi, Y. Yamaguchi, K. Kohata, Y.
Yamauchi, H. Ito, H. Higashio, J. Agric. Food Chem. 47 (1999) 3963–3966;
(d) V. Dinoiu, Rev. Roum. Chim. 51 (2006) 1141–1152.
[35] (a) B.J. Hathaway, A.A.G. Tomlinson, Coord. Chem. Rev. 5 (1970) 1–43;
(b) B.J. Hathaway, D.E. Billing, Coord. Chem. Rev. 6 (1970) 143–207;
(c) I.M. Procter, B.J. Hathaway, P. Nicholls, J. Chem. Soc. A: Inorg. Phys. Theor.
(1968) 1678–1684.
[36] U. Sakaguchi, A.W. Addison, J. Chem. Soc., Dalton Trans. (1979) 600–608.
[37] (a) A.A. Isse, A. Maye, E. Vianello, J. Electroanal. Chem. 431 (1997) 249–255;
(b) J.E. Kuder, H.W. Gibson, D. Wychick, J. Org. Chem. 40 (1975) 875–879;
(c) B. Ward, C.K. Chang, R. Young, J. Am. Chem. Soc. 106 (1984) 3943–3950;
(d) A.J. Bard, L.R. Faulkner (Eds.), Electrochemical Methods: Fundamentals and
Applications, second ed., Wiley, New York, NY, 2001.
[26] J.M. Fernandez, E.A. Aruaz, R.C. Rosado, R.A. Toscano, Transition Met. Chem. 24
(1999) 18–24.
[27] (a) A.D. Garnovski, A.P. Sadimenko, M.C. Sadimenko, D.A. Garnovski, Coord.
Chem. Rev. 173 (1998) 31–77;
(b) S. Yamada, Coord. Chem. Rev. 190–192 (1999) 537–555.
[28] A.F. Marinovich, R.S. O’Mahony, J.M. Waters, T.N.M. Waters, Croat. Chem. Acta 72
(1999) 685–703.
[29] J.U. Ahmed, M.T. Raisanen, M. Nieger, P.J. Figiel, M. Leskela, T. Repo, Polyhedron 38
(2012) 205–212.
[30] (a) V.T. Kasumov, F. Ko¨ksal, A. Sezer, Polyhedron 24 (2005) 1203–1211;
(b) V.T. Kasumov, F. Ko¨ksal, A. Kutluay, Spectrochim. Acta, A 76 (2010) 99–106;
(c) V.T. Kasumov, F. Ko¨ksal, R. Ko¨seog˘lu, J. Coord. Chem. 57 (2004) 591–603.
[31] (a) G.C. Percy, D.A. Thorpton, J. Inorg. Nucl. Chem. 35 (1973) 2319–2327;
(b) S.J. Gruber, C.M. Harris, E. Sinn, J. Inorg. Nucl. Chem. 30 (1968) 1805–1830.
[32] (a) M. Fores, M. Duran, M. Sola, Chem. Phys. 234 (1998) 1;
(b) G.L. Estiu, A.H. Jubert, J. Costamagna, J. Mol. Struct. 367 (1996) 97–110;
(c) J.M. Downes, J. Whelan, B. Bosnich, Inorg. Chem. 20 (1981) 1081–1086.
[33] (a) A.C. Braithwaite, T.N. Waters, J. Inorg. Nucl. Chem. 35 (1973) 3223–3229;
(b) S. Di Bella, I. Fragala, I. Ledoux, T.J. Marks, J. Am. Chem. Soc. 117 (1995)
9481–9485;
[38] (a) J.A. Halfen, B.A. Jazdzewski, S. Mahapatra, L.M. Berreau, E.C. Wilkinson, L. Que,
W.B.J. Tolman, J. Am. Chem. Soc. 119 (1997) 8217–8227;
(b) Sokolowski, H. Leutbecher, T. Weyhermu¨ller, R. Schnepf, E. Both, E. Bill, P.
Hildebrandt, K. Wieghardt, J. Biol. Inorg. Chem. (1997) 444–453.
[39] (a) H.G. Petering, H.H. Buskirk, G.E. Underwood, Cancer Res. 24 (1964) 367–372;
(b) L.A. Saryan, E. Ankel, C. Krishnamurti, D.H. Petering, H. Elford, J. Med. Chem.
22 (1979) 1218–1221;
(c) M. Christlieb, H.S.R. Struthers, P.D. Bonnitcha, A.R. Cowley, J.R. Dilworth, J.
Chem. Soc., Dalton Trans. (2007) 5043–5054;
(d) W. Herna´ndez, J. Paz, F. Carrasco, A. Vaisberg, J. Manzur, E. Spodine, L. Hennig,
J. Sieler, L. Beyer, Naturforschung 65b (2010) 1271–1278.
[40] J.F. Larrow, E.N. Jacobsen, Y. Gao, Y. Hong, X. Nie, C.M. Zepp, J. Org. Chem. 59
(1994) 1939–1942.
[41] R.S. Drago, Physical Methods in Chemistry, W.B. Saunders Company, Philadel-
phia, London, Toronto, 1977.
[42] Bruker, SADABS, Bruker AXS Inc., Madison, WI, 2007.
[43] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112–122.
[44] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997), 565–565.
(c) A. Bencini, I. Bertini, D. Gatteschi, A. Scozzafava, Inorg. Chem. 17 (1978)
3194–3197.