Cd(II) and Cu(II) complexes of polydentate Schiff base ligands: Synthesis, characterization, properties and biological activity

Article abstract of DOI:10.1016/j.ica.2004.11.026

We report the synthesis of the Schiff base ligands, 4-[(4-bromo- phenylimino)-methyl]-benzene-1,2,3-triol (A₁), 4-[(3,5-di-tert-butyl- 4-hydroxy-phenylimino)-methyl]-benzene-1,2,3-triol (A₂), 3-(p-tolylimino-methyl)-benzene-1,2-diol (A₃), 3-[(4-bromo- phenylimino)-methyl]-benzene-1,2-diol (A₄), and 4-[(3,5-di-tert- butyl-4-hydroxy-phenylimino)-methyl]-benzene-1,3-diol (A₅), and their Cd(II) and Cu(II) metal complexes, stability constants and potentiometric studies. The structure of the ligands and their complexes was investigated using elemental analysis, FT-IR, UV-Vis, ¹H and ¹³C NMR, mass spectra, magnetic susceptibility and conductance measurements. In the complexes, all the ligands behave as bidentate ligands, the oxygen in the ortho position and azomethine nitrogen atoms of the ligands coordinate to the metal ions. The keto-enol tautomeric forms of the Schiff base ligands A₁-A ₅ have been investigated in polar and non-polar organic solvents. Antimicrobial activity of the ligands and metal complexes were tested using the disc diffusion method and the strains Bacillus megaterium and Candida tropicalis. Protonation constants of the triol and diol Schiff bases and stability constants of their Cu²⁺ and Cd²⁺ complexes were determined by potentiometric titration method in 50% DMSO-water media at 25.00 ± 0.02°C under nitrogen atmosphere and ionic strength of 0.1 M sodium perchlorate. It has been observed that all the Schiff base ligands titrated here have two protonation constants. The variation of protonation constant of these compounds was interpreted on the basis of structural effects associated with the substituents. The divalent metal ions of Cu²⁺ and Cd ²⁺ form stable 1:2 complexes with Schiff bases. The Schiff base complexes of cadmium inhibit the intense chemiluminescence reaction in dimethylsulfoxide (DMSO) solution between luminol and dioxygen in the presence of a strong base. This effect is significantly correlated with the stability constants K_CdL of the complexes and the protonation constants K _OH of the ligands; it also has a nonsignificant association with antibacterial activity.

Full text of DOI:10.1016/j.ica.2004.11.026

Inorganica Chimica Acta 358 (2005) 1785–1797
www.elsevier.com/locate/ica
Cd(II) and Cu(II) complexes of polydentate Schiff base ligands:
synthesis, characterization, properties and biological activity
a
Aysegul Golcu , Mehmet Tumer , Havva Demirelli , R. Alan Wheatley
a
b
c,
*
a
Department of Chemistry, Faculty of Science and Arts, University of Kahramanmaras Sutcu Imam, 46100 Kahramanmaras, Turkey
b
Department of Chemistry, Faculty of Education, University of Gazi, Ankara, Turkey
c
Department of Chemistry, University of Hull, Cottingham Road, Hull HU6 7RX, UK
Received 15 April 2004; accepted 28 November 2004
Abstract
We report the synthesis of the Schiff base ligands, 4-[(4-bromo-phenylimino)-methyl]-benzene-1,2,3-triol (A1), 4-[(3,5-di-tert-
butyl-4-hydroxy-phenylimino)-methyl]-benzene-1,2,3-triol (A2), 3-(p-tolylimino-methyl)-benzene-1,2-diol (A3), 3-[(4-bromo-phe-
nylimino)-methyl]-benzene-1,2-diol (A4), and 4-[(3,5-di-tert-butyl-4-hydroxy-phenylimino)-methyl]-benzene-1,3-diol (A5), and their
Cd(II) and Cu(II) metal complexes, stability constants and potentiometric studies. The structure of the ligands and their complexes
1
was investigated using elemental analysis, FT-IR, UV–Vis, H and 13C NMR, mass spectra, magnetic susceptibility and conduc-
tance measurements. In the complexes, all the ligands behave as bidentate ligands, the oxygen in the ortho position and azomethine
nitrogen atoms of the ligands coordinate to the metal ions. The keto-enol tautomeric forms of the Schiff base ligands A1–A5 have
been investigated in polar and non-polar organic solvents. Antimicrobial activity of the ligands and metal complexes were tested
using the disc diffusion method and the strains Bacillus megaterium and Candida tropicalis.
Protonation constants of the triol and diol Schiff bases and stability constants of their Cu2+ and Cd2+ complexes were determined
by potentiometric titration method in 50% DMSO–water media at 25.00 0.02 ꢀC under nitrogen atmosphere and ionic strength of
0.1 M sodium perchlorate. It has been observed that all the Schiff base ligands titrated here have two protonation constants. The
variation of protonation constant of these compounds was interpreted on the basis of structural effects associated with the substit-
uents. The divalent metal ions of Cu2+ and Cd2+ form stable 1:2 complexes with Schiff bases.
The Schiff base complexes of cadmium inhibit the intense chemiluminescence reaction in dimethylsulfoxide (DMSO) solution
between luminol and dioxygen in the presence of a strong base. This effect is significantly correlated with the stability constants KCdL
of the complexes and the protonation constants KOH of the ligands; it also has a nonsignificant association with antibacterial
activity.
ꢁ 2004 Elsevier B.V. All rights reserved.
Keywords: Schiff base ligands; Metal complexes; Antimicrobial activity; Stability constants; Potentiometric studies; Luminol chemiluminescence
1. Introduction
transition metal complexes of Schiff bases [3–5]. This is
due to the fact that Schiff bases offer opportunities for
inducing substrate chirality, tuning the metal centred
electronic factor, enhancing the solubility and stability
of either homogeneous or heterogeneous catalysts [6–
11]. Schiff base complexes have been amongst the most
widely studied coordination compounds in the past
few years, since they are becoming increasingly impor-
tant as biochemical, analytical and antimicrobial re-
agents [12]. Schiff bases derived from a large number
The condensation of primary amines with carbonyl
compounds yields Schiff bases [1,2]. In the recent years,
there has been considerable interest in the chemistry of
*
Corresponding
+441482466416.
author.
Tel.:
+441472852338;
fax:
E-mail address: rawheatley@canadalane.demon.co.uk (R.A.
Wheatley).
0020-1693/$ - see front matter ꢁ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.11.026

1786
A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
of carbonyl compounds and amines have been used
[13,14]. It has been shown that Schiff base complexes de-
rived from 4-hydroxysalicylaldehyde and amines have
strong anticancer activity, e.g., against Ehrlich ascites
carcinoma (EAC) [15]. It is well known that some drugs
have greater activity when administered as metal com-
plexes than as free organic compounds [16].
A large number of reports are available on the chem-
istry and the biocidal activities of transition metal com-
plexes containing O,N and S,N donor atoms. The
transition metal complexes having oxygen and nitrogen
donor Schiff bases possess unusual configuration, struc-
tural lability and are sensitive to molecular environment
[16]. The environment around the metal centre ‘‘as coor-
dination geometry, number of coordinated ligands and
their donor group’’ is the key factor for metalloproteins
to carry out specific physiological functions [17].
potentiometric determination of the protonation con-
stants of the Schiff bases and stability constants of the
complexes by the potentiometric titration method in
50% DMSO–water media at 25.00 0.02 ꢀC and ionic
strength 0.10 M sodium perchlorate. Data were evalu-
ated using a computer program BEST [18]. The effects
of substituents on the protonation constants of Schiff
bases and on the stability constants of Schiff base-metal
complexes have also been investigated.
We also report and discuss the inhibition by the Schiff
base complexes of cadmium of the intense chemilumi-
nescence reaction in dimethylsulfoxide (DMSO) solu-
tion between luminol and dioxygen in the presence of
a strong base [19]. This is believed to be the first study
of the inhibition of luminol chemiluminescence to be
carried out in a non-aqueous solvent system.
Among different physicochemical properties of
organic compounds, protonation constants and stability
constants determined in mixed solvents provide an impor-
tant basis for speculation about whether substituent
effects influence their acidity and basicity. It is also
accepted that the knowledge of the stability constants of
such Schiff base-metal complexes may eventually help to
throw some light on the inactivation of essential trace
metals in biological systems. It was therefore, of interest
to determine the protonation and stability constants of
some Schiff base-metal complexes in DMSO–water media
of different compositions potentiometrically.
2. Experimental
2.1. Materials
All chemicals used in this study were obtained com-
mercially and used as supplied unless otherwise stated.
3,5-Di-t-butyl-4-hydroxyaniline was prepared according
to a known procedure [20] and used without further
purification.
2.1.1. Potentiometric titrations
In the present paper, the preparation, characteriza-
tion and antimicrobial properties of the various Schiff
bases and their Cu(II) and Cd(II) complexes shown in
Fig. 1 are reported. The study further deals with the
Stock solutions of these Schiff bases were prepared in
purified DMSO [21]. Doubly distilled conductivity (Mil-
lipore system) water was used as aqueous medium as
well as for preparation of DMSO–water mixtures. All
Y
X
Y
R
R
X
1
3
N
R
OH
R
O
N
Ar
M
Ar:
R
Ar
2
N
O
X
R
3
1
2
Y
M: Cu(II) and Cd(II)
Groups
X
A1
OH
OH
H
A2
A3
OH
H
A4
OH
H
A5
OH
OH
H
Y
OH
R1
C(CH3)3
OH
H
H
C(CH3)3
OH
R2
Br
H
CH3
H
Br
H
R3
C(CH3)3
C(CH3)3
Fig. 1. Proposed structures of the ligands and their metal complexes.

A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
1787
the other chemicals used were of A.R. grade and were
used without further purification. Stock solutions of
0.03 M Cu(ClO4)2 · 6H2O and Cd(ClO4)2 · 6H2O were
standardized using an appropriate indicator by EDTA
titrations [22]. Sodium hydroxide solutions were pre-
pared as 50% aqueous dimethyl sulfoxide solutions
and its concentration and absence of carbonate were
frequently checked by means of Gran plots [23] using
potassium hydrogen phthalate (Merck) as the acid.
0.1 M acid solutions prepared from Merck p.a. per-
chloric acid were titrated against standardized 0.1 M so-
dium hydroxide solution [24]. The ionic strength of each
solution was adjusted to 0.10 M by the addition of Na-
ClO4 as supporting electrolyte.
off, washed with cold ethanol and dried in a vacuum
dessicator.
Caution: As the perchlorate salts are explosive, care is
essential.
2.2.3. Preparation of microbial culture
Fifteen compounds were evaluated for their in vitro
antibacterial activity against Bacillus megaterium DSM
32 and antifungal activity against Candida tropicalis
FMC 23 by the agar-well diffusion method. Bacteria
were inoculated into Nutrient Broth (Difco) and incu-
bated for 24 h and the fungi studied incubated in Malt
Extract Broth (Difco) for 48 h. In the agar-well diffusion
method, Mueller Hinton Agar (Oxoid) for bacteria and
Malt Extract Broth (Difco) sterilized in a flask and
cooled to 45–50 ꢀC was distributed (20 ml) to sterilized
petri dishes after injecting 0.01 ml cultures of bacterium
prepared as mentioned above and allowed to solidify.
The dilution plate method was used to enumerate micro-
organisms (105 bacteria per ml) and fungi (103–104 per
ml) for 24 h [25]. By using a sterilized cork borer (7
mm diameter), wells were dug in the culture plates.
Compounds dissolved in CHCl3 were added (0.2 ll) to
these wells. The petri dishes were left 4 ꢀC for 2 h and
then the plates were incubated at 30 ꢀC for bacteria
(18–24 h) and at 25 ꢀC for fungi (72 h). At the end of
the period, inhibition zones formed on the medium were
evaluated as millimeters (mm) diameter. The control
samples were CHCl3 only.
2.1.2. Effects of cadmium complexes on luminol
chemiluminescence
12 mM Luminol (Sigma–Aldrich) stock solution
was prepared by dissolving luminol in DMSO and
protected from light by a foil wrapper. 1.0 mmol
dmꢀ3 solutions of cadmium acetate (Sigma–Aldrich)
and the cadmium complexes were made up in DMSO
(Avocado), were stored at 4 ꢀC and thawed by gentle
warming before use.
Luminol chemiluminescence in DMSO requires the
presence of only air and a strong base [19]; a reagent
was prepared comprising DMSO saturated with air
and sodium hydroxide and containing 400 mg dmꢀ3
Tween 65 (Fluka). The solvent was stirred for 30 min
with the surfactant and an excess of NaOH pellets while
simultaneously bubbling with air. It appeared that sus-
pended small particles of solid sodium hydroxide were
an essential component of the reagent, for if these were
allowed to settle and the clear solution decanted for use
as reagent, light emission was very feeble. As the results
show, the presence of the surfactant gave tolerable
reproducibility from the same batch of reagent, but
blank values of different batches varied.
2.2.4. Physical measurements
Elemental analyses (C, H, N) were performed by the
¨
TUBITAK Instrumental Analysis Laboratory (Besevier,
Ankara, Turkey) using a Carlo Erber 1106 elemental
analyser. Infrared spectra were obtained using KBr discs
(4000–400 cmꢀ1) on a Shimadzu 8300 FT-IR spectro-
photometer. The electronic spectra in the 200–900 nm
range were obtained on a Shimadzu UV-160 A spectro-
photometer. Magnetic measurements were carried out
by the Gouy method using Hg[Co(SCN)4] as calibrant.
Molar conductances of the Schiff base ligands and their
transition metal complexes were determined in DMSO
(ꢁ10ꢀ3 M) at room temperature using a Jenway Model
4070 conductivity meter. The mass spectral analyses
were carried out on a VG ZabSpec spectrophotometer.
1H and 13C NMR spectra were taken on a Varian XL-
200 NMR instrument. TMS was used as internal stan-
dard and deuterated dimethyl sulfoxide as solvent. The
metal contents of the complexes were determined by
an Ati Unicam 929 Model AA Spectrometer in solutions
prepared by decomposing the compounds in aqua regia
and then subsequently digesting in concentrated HCl.
2.2. Procedures
2.2.1. Preparation of Schiff base ligands (A1–A5)
The Schiff base ligands were prepared by the condens-
ing the amine derivatives (1 mmol) with the carbonyl
compounds (1 mmol) in EtOH (40 ml) by boiling the
mixture under reflux for 5–6 h. The precipitated ligand
was filtered off, recrystallized from acetone/hexane and
dried in a vacuum desiccator.
2.2.2. Preparation of Schiff base complexes
The appropriate quantity of Schiff base ligand
(1 mmol) was dissolved in absolute EtOH (15 ml). To
this solution was added a solution of Cd(ClO4)2 Æ 6H2O
or Cu(ClO4)2 Æ 6H2O (0.5 mmol) in absolute EtOH (15
ml) [1:2 molar ratio, M:L]. The mixture was stirred for
3 h at 80 ꢀC. The precipitated complex was then filtered
2.2.5. Potentiometric titrations
The pH measurement of proton-ligand and metal–
ligand systems of Schiff bases were made with a Orion

1788
A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
EA 940 pH meter, equipped with a Mettler Toledo In-
lab 412 combined glass electrode and Orion 960 auto-
matic titrator, containing carbonate-free sodium
hydroxide at a known (ꢁ0.1 mol dmꢀ3) concentration
at 25.00 0.02 ꢀC with ionic strength 0.10 M (Na-
ClO4). Temperature was maintained constant inside
the cell at 25.00 0.02 ꢀC, by the circulating water
by a Haake thermostatted bath (precision 0.02). All
potentiometric measurements in this study were car-
ried out in water–DMSO mixtures containing 50%
DMSO because of low solubility of Schiff bases and
possible hydrolysis in aqueous solutions.
The potentiometric cell was calibrated before each
experiment to obtain ꢀlog[H+] values (pH) for the titra-
tion medium [26]. The ion products (Kw = [H+][OHꢀ])
were calculated at a constant ionic strength of 0.10 M
with NaClO4 in 50% aqueous DMSO solutions based
on measurements of [OHꢀ] and pH in several series of
experiments. The standardization of the combined pH
electrode was also checked in the alkaline range by addi-
tion of excess NaOH. By assuming the E0cell value deter-
mined in the acidic range to be reliable and the [OHꢀ]
concentration of a base added in excess, we calculated
the reproducible values of pKw for the examined 50%
aqueous dimethyl sulfoxide solution [27,28]. The pKw
value obtained is 15.48 in this medium.
closed and auto-shutter off (i.e., no photoexcitation).
The instrument photomultiplier was set for high sensi-
tivity. Solutions under examination were held in dispos-
able polystyrene cuvettes (Fisher, Loughborough, UK),
4.0 ml capacity. In the interests of reproducibility, the
following protocol for measuring the light emission
was observed strictly for each determination.
The chemiluminescence reagent as described above
was inverted and then shaken five times in its flask to
give reasonable homogeneity; 2.0 ml was pipetted into
the cuvette and the determination begun (time = 0 s)
by adding 200 ll of the DMSO solution under investiga-
tion (or, for blanks, 200 ll of DMSO alone). The cuvette
was inverted and shaken five times. At the 60 s point,
100 ll of 12 mM luminol in DMSO solution was added,
followed by two inversions and 10 shakes. The cuvette
was then installed in the cuvette holder of the spectroflu-
orimeter and at the 90 s point, the first luminescence
reading was recorded. Readings were then recorded
every 15 s, until the 10th reading has been recorded at
the 225 s point. Light emission was measured in the arbi-
trary units of the spectrofluorimeter data output.
2.2.6.1. Data processing. An index of light emission was
obtained by summing the 10 readings taken. Means and
standard deviations of this measure were calculated for
each solution examined. The means were ranked and
the statistical significance of the differences between
means adjacent in the series was evaluated using the Stu-
dent t-distribution [29].
Linear correlation between light emission and other
properties was measured by Pearsonꢀs correlation coeffi-
cient, computed by MSExcel. The significance of the
correlation coefficients was tested at p = 0.10 using sta-
tistical tables [30].
Potentiometric titrations were carried out at constant
temperature and in an inert atmosphere of nitrogen with
CO2-free standardized 0.1 M NaOH in a 50.0 ml solu-
tion containing 0.1 M NaClO4: (i) 2.5 · 10ꢀ3 M HClO4
(for
cell
calibration);
(ii)
3.0 · 10ꢀ3
M
HClO4 + 1.5 · 10ꢀ3 Schiff base (for the protonation
constant of the Schiff base); (iii) 3.0 · 10ꢀ3
M
HClO4 + 1.5 · 10ꢀ3 M Schiff base + 7.5 · 10ꢀ4 M Cu
(ClO4)2/Cd(ClO4)2 (for the stability constant of the
complex).
2.2.5.1. Data processing. The protonation and stability
constants of A1–A5 Schiff bases were evaluated by itera-
tive non-linear least squares fit of potentiometric equi-
librium curves through mass balance equations for all
the components expressed in term of known and un-
known equilibrium constants using a computer program
BEST [18]. All the models converged at r < 0.03 pH units
of the observed pH values, which is considered to be an
acceptable fit. The equilibrium constants reported in this
paper were obtained as averaged values of three titra-
tions. Selection of the equilibrium models was based
on critical evaluation of the least squares fitting results,
namely analysis of the statistical parameters.
3. Results and discussion
3.1. Spectroscopy
Condensation of the aldehydes with primary amines
readily gives rise to the corresponding imines, which
were easily identified by their IR, 1H and 13C NMR
spectra, where replacement of the carbonyl by the imine
group results in: (i) lowering of the energy of the
m(C@O) stretch in the IR spectrum and (ii) a shift to
higher field of the CH@N proton signal in the 1H
NMR spectrum. The imines prepared in this way are
formed in nearly quantitative yields and are of high pur-
ity. All compounds, except 3,5-di-t-butyl-4-hydroxyani-
line compound, are very stable at room temperature in
the solid state. On the other hand, the electron with-
drawing or – attracting groups at the ortho and para
positions of the ligands are able to produce different
electronic and steric effects. Therefore, it may be possi-
2.2.6. Effects of cadmium complexes on luminol
chemiluminescence
All chemiluminescence measurements are determina-
tions of emission at 480 nm (10 nm bandwidth) using a
Shimadzu RF-150X spectrofluorimeter, with the shutter

A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
1789
ble that the t-butyl groups on the aniline ring and the
hydroxy or methoxy groups meta to OH in the salicylid-
ene moiety lead to a decrease in stability of the com-
plexes. The yields of the complexes are lower than
those of the ligands. Further stirring and heating did
not increase the yield of the complexes containing the
t-butyl groups. The low yields may be due to the steric
hindrance around the coordination centre. All the li-
gands and their complexes are soluble in common or-
ganic solvents such as CHCl3, EtOH, MeOH, THF,
etc. Solution conductivity measurements were per-
formed to establish the charge of the complexes. The re-
sults show that all compounds are non-electrolyte [31].
The results of the elemental analyses, given in Table 1,
are in accord with the composition suggested for the li-
gands and their metal complexes.
The infrared spectral data of the ligands (A1–A5) and
their metal complexes are given in Table 2. In the li-
gands, the bands in the 3500–3420 cmꢀ1 range may be
assigned to m(O–H) stretching. For the free ligands, the
broad bands in the 2800–2700 cmꢀ1 range are assigned
to the OH group vibration (ortho position) associated
intramolecularly with the nitrogen atom of the CH@N
group [32]. These bands disappear in the complexes, as
a result of proton substitution by cation coordination
to oxygen. For the ligands, the strong bands observed
in the 1640–1615 cmꢀ1 range are assigned to the azome-
thine group vibration. These bands are slightly shifted
towards lower frequencies in the complexes, and this
change in the frequencies shows that the imine nitrogen
atom coordinated to the Cd(II) and Cu(II) ions. The
medium intensity bands observed for all ligands in the
1350–1275 cmꢀ1 range can be attributed to the phenolic
stretch. These bands are observed for the complexes at
lower wave number by ca. 10–20 cmꢀ1 relative to the
free ligands suggesting involvement of the oxygen atom
of the C–O moiety in coordination [33]. In all of the
present complexes a medium and/or weak band ob-
served in the 510–415 cmꢀ1 range can be attributed to
the m(M–N) and m(M–O) [34] modes.
In the case of the ligands and their metal complexes,
it is particularly important to establish whether the mol-
ecules retain the imine character of their phenol precur-
sor. The most useful techniques to investigate the
tautomeric forms (Fig. 2) of these ligands are UV and
NMR spectroscopy, while IR seems of limited value
here because location of the m(C@O) and m(C–O)
stretches in the spectra is obscured by the abundance
of aromatic skeletal modes. In order to investigate the
keto-enol tautomeric forms of the free ligands, the elec-
tronic spectra were measured in heptane, chloroform
and ethanol. In heptane, the ligands exhibit maxima in
320–278 nm range. However, in chloroform and etha-
nol, new bands in the 464–337 nm range were observed.
The former set (ꢁ278 nm) in heptane have been assigned
to the enolimine tautomer and the latter to the keto-
amine tautomer of the Schiff bases [35].
Electronic spectra have been measured in EtOH and
the numerical data are given in Table 2. Hydrogen-bond
forming solvents (EtOH) thus favour the formation of
the ketoamine. The interaction of enolimine with a
hydrogen bond forming solvent would presumably re-
duce the O–H bond strength and facilitate proton trans-
fer to the nitrogen centre. In like manner, the bands in
the 464–337 nm range are also assigned to the n–p* tran-
sition of the azomethine group. In the spectra of the
complexes, the bands of the azomethine chromophore
n–p* transition are shifted to lower frequencies indicat-
ing that the imine nitrogen atom is involved in coordina-
tion to the metal ion. The bands at higher energies
Table 1
Some analytical and physical data for the imine ligands and their complexes
a
KM
Compound
Colour
leff (as B.M.)
Yield (%)
m.p. (ꢀC)
Found (Calc.) (%)
a
KM
C
H
M
A1
orange
brown
84
74
70
85
60
70
87
68
79
90
65
65
86
70
68
201
>250
>250
204
50.67(50.65)
46.35(46.32)
42.95(42.91)
70.53(70.58)
65.03(64.99)
61.16(61.13)
74.04(74.00)
65.20(65.17)
59.50(59.52)
53.47(53.42)
48.37(48.33)
44.89(44.92)
73.87(73.90)
67.81(67.78)
63.58(63.60)
4.49(4.54)
4.17(4.18)
3.87(3.85)
3.88(3.92)
3.64(3.61)
3.36(3.40)
6.15(6.17)
5.46(5.43)
5.01(4.96)
4.86(4.79)
4.32(4.34)
4.04(4.03)
4.16(4.11)
3.74(3.77)
3.57(3.53)
3.22(3.24)
2.63(2.64)
2.51(2.48)
7.58(7.56)
6.74(6.71)
6.34(6.31)
5.68(5.73)
4.70(4.66)
4.28(4.25)
3.47(3.42)
2.83(2.79)
2.64(2.59)
7.88(7.92)
7.02(6.99)
6.60(6.56)
1.8
7.4
7.0
1.2
6.8
9.0
1.5
6.9
8.0
1.0
8.5
10.0
1.9
7.3
7.6
Cu(A1)2
Cd(A1)2
A2
1.79
Diamag.
9.48(9.43)
15.55(15.49)
yellow
yellow
Cu(A2)2
Cd(A2)2
A3
dark brown
yellow
orange
1.8
Diamag.
>250
>250
156
8.27(8.19)
13.72(13.64)
Cu(A3)2
Cd(A3)2
A4
light brown
orange
orange
1.81
Diamag.
>250
238
184
12.40(12.32)
19.99(19.93)
Cu(A4)2
Cd(A4)2
A5
dark brown
yellow
yellow
1.8
Diamag.
>250
201b
163
9.90(9.84)
16.27(16.20)
Cu(A5)2
Cd(A5)2
a
dark brown
orange
1.81
Diamag.
>250
124b
8.65(8.55)
14.27(14.19)
Xꢀ1 cm2 molꢀ1
.
b
Decompose.

1790
A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
Table 2
Infrareda and electronic spectral data for the Schiff base ligands and their metal complexes (cmꢀ1
)
Compound
m(OH)x
m(OH)
m(CH3)y
m(O–Hꢂ ꢂ ꢂN)
m(CH@N)
m(C–OH)
m(M–N)
m(M–O)
kmax (nm)
A1
A2
3450 br
3380 br
3400 br
3375 br
3410 br
2800 m
2790 m
2700 m
2745 m
2740 m
1620 s
1620 s
1641 s
1633 s
1625 s
1605 m
1610 m
1614 s
1620 s
1613 s
1612 s
1618 s
1613 s
1624 s
1610 s
1290
1280
1283
1300
1265
1270
1276
1275
1294
1260
1275
1273
1270
1250
1262
420, 375, 339, 280, 240
425, 348, 294, 255
359, 320, 281, 226
3610 s
2980 s
2910 s
A3
A4
464, 368, 317, 279, 271
430, 275, 225
418, 354, 287, 238
A5
3600 s
3608 s
2975 s
503 w
505 w
497 w
500 w
503 w
510 w
490 w
480 w
495 w
497 w
485 w
434 w
428 w
415 w
434 w
444 w
424 w
448 w
442 w
420 w
430 w
418 w
Cd(A1)2
Cd(A2)2
Cd(A3)2
Cd(A4)2
Cd(A5)2
Cu(A1)2
Cu(A2)2
Cu(A3)2
Cu(A4)2
Cu(A5)2
a
2978 s
2910 s
357, 289, 234
389, 334, 271, 245, 220
438, 354, 305, 262, 260
422, 354, 349, 283, 238
624, 380, 324, 275, 216
633, 412, 285, 254, 215
610, 370, 341, 295, 271, 227
612, 363, 325, 294, 270,221
627, 429, 327, 220
3600 s
3609 s
2975 s
2977 s
2909 s
3598 s
2975 s
br (broad), s (strong), m (medium), w (weak). x: Sterically hindered phenol, y: CH3 and CH3 group of the t-butyl group.
Y
X
of the complexes are presented in Table 1. The magnetic
moment of the copper(II) complexes was observed in the
range of 1.79–1.81 B.M. which corresponds to a single
unpaired electron with a very slight orbital contribution.
The geometries of the Cu(II) complexes are square-pla-
nar [40]. The structures of the monomeric complexes are
supported by magnetic moment data. The Cd(II) com-
plexes of the ligands A1–A5 were found to have diamag-
netic character and tetrahedral geometry around the
metal ions.
Y
X
N
R
OH
R
NH
O
3
R
1
R
3
R
1
2
R
2
Additional structural information can be deduced
1
1
from the H and 13C NMR spectra. The H- and 13C-
chemical shifts for the Schiff base ligands are given in
Table 3. It is noteworthy that the influence of substitu-
tion on the chemical shifts of the aniline moiety is weak.
When a bromine atom is present on the aniline residue, a
significant deshielding of the proton signals can be ob-
served, due to the strong electron-withdrawing effect of
this substituent. However, the OH groups on the salicy-
lidene moiety increases the electron density of the aro-
matic rings, due to the resonance or mesomeric effect.
In Schiff base ligands A1–A5, there is the proton donor
OH group and one proton acceptor group in the ortho
position. Due to presence of the OH groups in ortho po-
sition to imine group, formation of a few intramolecular
hydrogen bonds is possible. It is important to emphasize
Enolimine
Ketoamine
Fig. 2. Tautomeric forms of the ligands.
(ꢁ290–206 nm range) are associated with benzene p–p*
transitions. The spectra of the complexes show intense
bands in the high-energy region at 302–372 nm which
can be assigned to charge transfer L ! M bands [36].
Although the precise nature of this transition involving
the phenolate group is not fully clear [it has been consid-
ered to be an O(phenolate) ! metal(II) LMCT [37] or as
a metal(II) ! p* (phenolate) MLCT [38] transition], it is
agreed that a higher energy component should exist near
330 nm [39]. t-Butyl groups on the aniline moiety occupy
a more significant region of space in the vicinity of the
metal centre due to the presence of the branched carbon.
Therefore, it is quite plausible that a sterically induced
increase in the average metal–ligand separation results
in a hypsochromic shift in the c.t. absorption maximum.
The spectra of the Cu(II) complexes contain one absorp-
tion band in the 633–610 nm range which may be as-
signed to the d–d transition of the Cu(II) ion
suggesting that they are four-coordinated complexes.
Magnetic measurements were recorded at room tem-
perature and the effective magnetic moment (leff) values
1
the H resonance of the O–H group in the 10.10–12.30
ppm range. The signal due to the OH proton disappears
in D2O solution.
The formulation of the ligands are deduced from
analytical data, 1H and 13C NMR and further sup-
ported by mass spectroscopy. The relatively low
intensities of the molecular ion peaks, [M]+, are indic-
ative of the ease of fragmentation of the compounds,
and this may reflect the number of heteroatoms pres-
ent in each structure. The spectra of the ligands A1–
A5 show peaks at m/e 308, 358, 227, 292 and 342

A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
1791
Table 3
The 1H (13C) NMR data (as ppm) for the Schiff base ligands using CDCl3 as solvent
Compound
CH3
t-Bu
Ar
CH@N
OH
A1
A2
A3
A4
A5
7.20–7.74 (112.81–152.45)
7.14–7.19 (110.10–156.66)
7.18–7.60 (116.13–157.17)
7.32–7.72 (113.40–154.61)
7.04–7.29 (115.20–155.00)
8.94 (163.15)
8.38 (166.75)
8.90 (162.52)
8.92 (165.05)
8.39 (169.90)
10.45–11.70
10.10–12.15
10.60–11.05
10.55, 10.30
10.10–12.30
1.51 (32.00)
2.28 (27.30)
1.62 (32.95)
CH3 coupling constant J ꢁ 3 Hz.
[M]+, respectively. All the ligands decompose in a
similar way.
been obtained. In this way, it would be possible to deter-
mine the coordination of the ligands.
3.2. Antimicrobial activity
3.4. Protonation constants of Schiff bases
Synthesized compounds were tested for in vitro anti-
microbial activity by the agar-well diffusion method.
The antibacterial and antifungal activities of the ligands
and their complexes against one bacterium (B. megate-
rium) and one fungus (C. tropicalis) are presented in Ta-
ble 4. Also included is the activity of the solvent CHCl3.
It was observed that all the compounds tested showed
antibacterial and antifungal activity. The complex
Cu(A1)2 of the ligand A1 does not show activity against
C. Tropicalis. On the other hand, the complex Cu(A2)2
has most activity against the same fungi. However, the
ligands and their Cu(II) and Cd(II) complexes are effec-
tive against the bacterium B. megaterium.
The study of complex formation by the studied Schiff
bases cannot be carried out in aqueous solution because
of the nature of the compounds involved. These metal
complexes as well as the ligands themselves are insoluble
in water. This solvent has been most widely used for
potentiometric determination of stability constants.
The mixture DMSO–water 50:50% was the chosen sol-
vent for our study. In such a medium, the studied Schiff
bases and their metal complexes are soluble giving stable
solutions. The use of this mixed solvent has some advan-
tages over pure DMSO. Thus, pure DMSO is very
hygroscopic and controlling its water content is difficult
[41]. This fact would affect reproducibility of our exper-
iments. However, DMSO–water 50:50% mixture has
only a small hygroscopic character. A further advantage
is its compatibility with the standard glass electrode, so
that the pH measurements may be carried out in a sim-
ilar way to that employed in a purely aqueous solution.
In contrast, the glass electrode has a slow response and
irreproducible behaviour in pure DMSO (10). For these
reasons, the use of pure DMSO is not recommended for
potentiometry. Another advantage of the DMSO–water
50:50 mixture is its large acidity range (pKw = 15.48)
which allows the investigation of deprotonation equilib-
ria of weak acids which could be hardly studied in water
[42,43].
The stoichiometric protonation constants of the
investigated Schiff bases (A1–A5) were determined in
50% DMSO–water mixture at 25 ꢀC and these constants
are tabulated in Table 5. As the titration curve of the li-
gand in Fig. 5, which is drawn based on A1 Schiff base, it
can be seen that there are two end-points at a = 1 and
a = 2. According to the results obtained from this titra-
tion curve it can be concluded that the Schiff bases stud-
ied here have two protonation constants. The log KOH
and log KNH values are related to the protonation of
no. 2 phenolic oxygen and imine nitrogen, respectively,
as Fig. 4.
3.3. Crystallization
We have already made numerous efforts to crystallize
any complex of the compounds for X-ray diffraction
studies. However, up till now, no single crystals have
Table 4
Antimicrobial effects of the ligands and their metal complexesa
Compound
Inhibition zone
(mmb) C. tropicalis
Inhibition zone
(mmb) B. megaterium
Control (CHCl3)
A1
A2
15
10
7
18
14
12
13
15
23
21
20
18
21
19
22
17
20
18
A3
A4
13
12
18
17
11
13
24
15
26
21
23
26
A5
Cd(A1)2
Cd(A2)2
Cd(A3)2
Cd(A4)2
Cd(A5)2
Cu(A1)2
Cu(A2)2
Cu(A3)2
Cu(A4)2
Cu(A5)2
a
This is also illustrated in the species distribution of
the A1 ligand in Fig. 3. At pH < 4, the ligand exists in
the fully protonated form H2L+ and in the form HL.
Conc. of compounds is 50 lg/cm3.
Including diameter of disc (6 mm).
b

1792
A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
Table 5
Successive protonation constants of A1–A5 Schiff bases in 50%
DMSO–water mixture (l = 0.100 M NaClO4, 25.00 0.02 ꢀC)
(3)
(2)
OH
OH
(3)
(2)
OH
OH
Schiff bases
Log KOH
Log KNH
N
O
(1)
A1
A2
A3
A4
A5
7.65 0.02
8.24 0.02
11.55 0.02
11.50 0.02
8.54 0.02
1.75 0.02
2.54 0.02
1.80 0.02
1.53 0.02
2.84 0.02
LogK
OH
N
OH
(1)
+
+
H
Br
Br
As the pH is increased, the ligand loses its protons from
imine nitrogen to become HL, which is the predominant
species in pH range of 3–6. As conditions become more
alkaline the second phenolic hydroxyl group begins
deprotonation to the free ligand Lꢀ anion which is pre-
dominant at pH > 9.
(3)
OH
OH
(3)
(2)
OH
OH
(2)
+
N
OH
(1)
LogK
NH
(+)NH OH
(1)
+
H
Previous studies about Schiff bases report that the log
protonation constants of imine nitrogen of Schiff bases
are about 3–5 [26,4]. In this study, however, the proton-
ation constant logarithms of the Schiff bases studied are
to be found between 1.5 and 2.84. This can be explained
by the fact that there is an aromatic triol or diol at-
tached to the imine nitrogen and such substituted phenyl
groups inductively reduce the electron density of imine
nitrogen.
From Table 5, log KOH of A2 Schiff base is 8.24 and
that of A1 Schiff base is 7.65 and log KNH of A2 Schiff
base is 2.54 and that of A1 Schiff base is 1.75. Substitu-
tion by hydroxyl instead of bromine causes a 0.59 unit
increase for log KOH and a 0.79 unit increase for
log KNH, which means a lowering of the acidity. This
can be explained by electron release of the hydroxyl
group in the benzene ring with resonance. Since tertiary
methyl groups are attached at the meta position relative
to the imine nitrogen, it is assumed that they have no ef-
fect on the value of the protonation constant of the
imine nitrogen.
Br
Br
Fig. 4. The log KOH and log KNH equilibrium reactions of the ligand
A1.
Another remarkable situation about protonation
constant values of the Schiff bases is that values of log -
KOH protonation constants of the A3 and A4 Schiff bases
are 3 units greater than others. This may be because of
the attachment at the meta position of the OH group,
which is inductively an electron acceptor and an electron
donor with resonance. At the meta position, OH cannot
donate electrons with resonance, so it just acts as an
electron acceptor.
In the case of A3 Schiff base and A4 Schiff base, the
log KNH values are found to be 1.80 and 1.53, respec-
tively, and the log KOH values are found to be 11.55
and 11.50, respectively. It can be seen that there is a rela-
tionship between alkyl substitution on the amino nitro-
gen and phenolic oxygen and increasing log KNH and
log KOH values. The electron-releasing effect of the alkyl
groups is the main factor in these examples. However,
the log KOH value of A4 Schiff base is only slightly smal-
ler than the log KOH value of A3 Schiff base because the
substituents are located far from the reaction site.
100
90
L-
80
70
60
50
40
30
20
10
0
HL
3.5. Stability constants of the Schiff base complexes
The potentiometric titration curves of A1 Schiff base
with equivalents of ligand to metal ion for Cd2+ and
Cu2+, are shown in Fig. 5. All metal ions depress the
titration curve of the free ligand by the release of pro-
tons according to the abilities of the metal ions to bind
to the ligand Schiff bases. As the titration curves of the
complexes formed by cadmium and copper together
with A1 Schiff base are examined, two inflection points
can be observed at a = 2 and a = 4 for Cd2+, and at
H2L+
2
4
6
8
10
12
pH
Fig. 3. Species distribution diagram for the systems A1 Schiff base (L)
as
function of pH. l = 0.100 mol lꢀ1 NaClO4, t = 25.0,
TL = 1.5 · 10ꢀ3 mol lꢀ1, % = percentage concentration of species.
a

A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
1793
Fig. 5. Potentiometric titration curves for A1 Schiff base (L) and 1:2 stoichiometries of Cu2+ and Cd2+ to A1 Schiff base (ML2) as a function of added
NaOH. (a = Moles of base added per mole of metal ion (or ligand) present.)
a = 3 and a = 4 for Cu2+. The presence of an inflection
point on the titration curves at a = 3 and a = 2 for
Cu2+ and Cd2+, respectively, can be explained by the
formation of protonated complexes. This is also illus-
trated in the species distribution of the ligand in Figs.
6 and 7. In order to investigate change with pH in the
concentration of the complexes, which Schiff bases
formed with Cu2+ and Cd2+, the stability constant val-
ues in Tables 6 and 7 are evaluated, using SPE computer
program (1) and the species distribution curves are
drawn. For example, in Fig. 6, if the distribution dia-
gram for Cu–A1 system is examined it will be seen that
the complex form with CuL2H2 proton is dominant in
the region up to pH 3.5. This species forms 93% at pH
2. Between pH 4 and 8 CuL2H complex is dominant
and above pH 9 CuL2 complex forms. This CuL2 com-
plex forms 95.6% ratio at pH 11.6. Moreover, from the
distribution diagram it can be seen that very few CuL
and CuLH complexes form. Similarly, as the distribu-
tion diagram in Fig. 7 is examined for Cd–A1 system,
complex form CdL2H2 seems to be dominant at
pH < 6. This complex forms 42.5% at pH 3.9. Above
pH 9, CdL complex is dominant.
The stoichiometric stability constants of metal(II)
(Cu2+ and Cd2+) complexes of the investigated Schiff
Bases were determined in 50% DMSO–water mixture
at 25 ꢀC and these constants are tabulated in Tables 6
and 7. All the metal ions were found to combine
easily with A1, A2, A3, A4 and A5 Schiff bases to form
deprotonated (ML2) and multiprotonated (MHnLn) in
100
90
CuL2H2
CuL2H
CuL2
80
70
60
%
50
40
30
20
10
Cu
3.00
CuL
7.00
- Log H+
0
A
2.00
4.00
5.00
6.00
8.00
9.00
10.00 11.00 12.00
Fig. 6. Species distribution diagram for the systems Cu–A1(L) Schiff base system in 1:2 molar ratio as a function of pH. l = 0.100 mol lꢀ1 NaClO4,
t = 25.0, TL = 3.0 · 10ꢀ3 mol lꢀ1, TCu = 1.5 · 10ꢀ3 mol lꢀ1, % = percentage concentration of species.

1794
A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
100
90
80
70
60
CdL
Cd
%
A
50
40
30
20
10
0
CdL2H2
CdLH
CdL2
2.00
4.00
6.00
8.00
- Log H+
10.00
12.00
Fig. 7. Species distribution diagram for the systems Cd–A1(L) Schiff base system in 1:2 molar ratio as a function of pH. l = 0.100 mol lꢀ1 NaClO4,
t = 25.0, TL = 3.0 · 10ꢀ3 mol lꢀ1, TCd = 1.5 · 10ꢀ3 mol lꢀ1, % = percentage concentration of species.
concentrations depending on the pH of the solution.
When the values in the table are examined, the following
order is obtained for stability constants of mononuclear
complexes (ML) with the Schiff bases: A3 > A4 > A5 >
A2 > A1. When the rank orders of stability constants
and of protonation constants are compared, it is seen
that there is correlation between them. The linear corre-
lation coefficients (r) between the logarithms of the sta-
bility constants of the complexes and the protonation
constants of the ligands are set out in Table 8. For both
cadmium and copper complexes, correlation between
the complex stability constants and the ligand proton-
ation constant KOH attains or approaches statistical sig-
nificance [30] in almost every case and is greatest for the
complexes of the type MH2L2, i.e., stability of these
complexes is greatest when the ligand is protonated un-
der most conditions. Unsurprisingly, this does not apply
to complexes of the ML2-type, which display a negative
correlation between the two constants. None of the sta-
bility constants correlates significantly with KNH. The
magnitude of stability constants for all ligands was
found to be Cu2+ > Cd2+, in agreement with increasing
acidity of the metal ion.
3.6. Effects of the cadmium complexes on luminol
chemiluminescence
In DMSO solution, luminol reacts with dissolved
oxygen in the presence of a strong base with intense
chemiluminescence [19]. Because of the relatively simple
reaction conditions, this phenomenon has found great
favour as a demonstration of chemiluminescence [44],
but analytical applications have almost universally made
use of the related reaction in aqueous solution.
Each of the five cadmium complexes studied here had
the effect of significantly (p < 0.01) diminishing the
chemiluminescence of luminol in DMSO solution, as
shown in Fig. 8. Chemiluminescence diminished along
the series A1 > A2 > A4 > A3, in each case the signal
being significantly less than that obtained in the pres-
ence of the preceding member of the series (p < 0.01
for A2 > A4, but p < 0.05 for A1 > A2 and A4 > A3).
The observed difference in emission between the two
complexes with the most powerful inhibitory effect
(A3 > A5) were not significant at the concentration used,
in part because of the imprecision of relatively low
signals.
Table 6
Logarithms of the stability constants of Cu(II) complexes of A1–A5 Schiff bases, in 50% DMSO–water mixture (l = 0.100 M NaClO4, 25.00 0.02
ꢀC)
Schiff Bases
Log KML
Log bMHL
Log bML
Log bMHL
Log bMH L
2 2
2
2
A1
A2
A3
A4
A5
8.09 0.02
8.25 0.02
8.49 0.02
8.38 0.02
8.33 0.02
6.25 0.02
6.44 0.02
7.05 0.02
6.98 0.02
6.32 0.02
13.94 0.02
13.73 0.02
13.67 0.02
13.65 0.02
13.83 0.02
21.83 0.02
21.76 0.02
22.01 0.02
21.96 0.02
22.06 0.02
25.44 0.02
26.03 0.02
28.25 0.02
28.12 0.02
26.35 0.02
Definitions of stability constants: KML = [ML]/[M2+][L]; bMHL = [MHL]/[M2+][L][H+]; bML = [ML2]/[M2+][L]2; bMHL = [MHL2]/[M2+][L]2 [H+];
bMH L = [MH2L2]/[M2+][L]2 [H+]2 (L = ligand in all definitions).
2
2
2
2

A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
1795
Table 7
Logarithms of the stability constants of Cd(II) complexes of A1–A5 Schiff bases, in 50% DMSO–water mixture (l = 0.100 M NaClO4 t = 25.00 0.02
ꢀC)
Schiff Bases
Log KML
Log bMHL
Log bML
Log bMHL
Log bMH L
2 2
2
2
A1
A2
A3
A4
A5
3.38 0.02
3.69 0.02
3.86 0.02
3.78 0.02
3.73 0.02
6.77 0.02
7.03 0.02
7.12 0.02
7.09 0.02
6.86 0.02
5.31 0.02
5.18 0.02
5.09 0.02
5.13 0.02
5.26 0.02
13.54 0.02
13.41 0.02
13.87 0.02
13.78 0.02
13.65 0.02
20.45 0.02
20.72 0.02
22.15 0.02
22.03 0.02
21.01 0.02
Definitions of stability constants as in Table 6.
Metal ions released by the breakdown of the complex
might have interfered in some way with the chemilumi-
nescence process. So luminol chemiluminescence was
also measured in the presence of 100 lM cadmium ace-
tate (Fig. 9). Although all the cadmium complexes sig-
nificantly diminished the luminol emission, cadmium
acetate at the same metal ion concentration as the com-
plexes did not. There have been numerous previous
observations of the effects on luminol chemilumines-
cence of cadmium ions, but only in aqueous solution.
Thus, cadmium(II) ions enhance luminol-hydrogen per-
oxide chemiluminescence [45] in the presence of 8-hy-
droxy-5-quinolinesulfonic acid. In contrast, a detection
reaction for cadmium(II) ions in an electrophoresis buf-
fer has been based on their reduction of luminol-hydro-
gen peroxide chemiluminescence signals [46] by
displacement of cobalt(II) ions on which the light emis-
sion depends. However, the emission being inhibited in
that case was dependent on the presence of a metal
ion, whereas luminol chemiluminescence in DMSO is re-
garded as being independent of the d-block catalysts
that are effective in aqueous solutions [19]. So the same
mechanism would not be expected to be operative in the
present study.
Table 8
The linear correlation coefficients (r) between the logarithms of the
stability constants of the complexes and the protonation constants of
the ligands (n = 5 in each case)
Cadmium
Copper
Log KOH
Log KNH
Log KOH
Log KNH
Log bMH L
Log bMHL
Log bMHL
Log bML
Log KML
0.9942***
0.8810**
0.8286*
ꢀ0.4827
ꢀ0.4988
ꢀ0.2857
0.3400
0.9957***
0.5284
0.9840***
ꢀ0.8446*
0.8637*
ꢀ0.4800
0.0529
2
2
2
ꢀ0.5916
0.2398
ꢀ0.8818**
2
0.7708
0.0860
ꢀ0.0843
*
r Significant at p < 0.10.
r Significant at p < 0.05.
r Significant at p < 0.01.
**
***
2000
1500
1000
500
0
Blank
A1
A2
A3
A4
A5
The absence of effect of cadmium ion suggests that
the light-emitting reactions are blocked either by the
whole complexes or by released ligands. Luminol chemi-
luminescence proceeds by a radical mechanism [47] and
is known to be inhibited if radicals are removed from the
reaction medium [19]. The Schiff base ligands A1–5 are
Cadmium complex
Fig. 8. Effect of cadmium complexes each at a final concentration of
100 lM on the chemiluminescence of luminol in DMSO solution
(mean + SD; n = 3 or 4).
The reagent used in this experiment was saturated
with sodium hydroxide. The novel complexes studied
were all water-labile, i.e., subject to nucleophilic attack,
and in DMSO solution they would have been subject to
attack by the sodium hydroxide present; base would be
consumed and metal ions would dissociate from the li-
gands to which they were initially co-ordinated. The
concentration of sodium hydroxide in the reagent was
not accurately known but the limit of solubility is more
than 100 mM (4 mg mlꢀ1). Only a small proportion of
this would be consumed by the postulated reaction be-
tween the complexes and sodium hydroxide. Therefore
the likely fall in base concentration would not seem to
be sufficient to explain the inhibition of chemilumines-
cence that occurred.
2500
2000
1500
1000
500
0
Fig. 9. Effect of cadmium acetate at a final concentration of 100 lM
on the chemiluminescence of luminol in DMSO solution (mean + SD;
n = 3).

1796
A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
all polyphenolic in character and the radical scavenging
potential of such compounds is well-known [48,49]. The
complexes containing benzene-1,2,3-triol ligands (A1,
A2) suppressed luminol chemiluminescence less than
those containing benzene-1,2-diol ligands (A3, A4); the
single complex (A5) containing a benzene-1,3-diol ligand
gave the greatest signal loss.
In addition to their direct radical scavenging effect,
the different ligands might have effects which depend
on the stability of their complexes. So we examined
the relationship between the stability constants of the
complexes and their effect on the chemiluminescence
reaction. There is a significant (p = 0.10) correlation
(r = 0.8310, n = 5, critical value 0.805 [30]) between the
effect of the complexes Cd(A1–5)2 on luminol chemilumi-
nescence and the logarithm of the formation constant
KCdL (i.e., they correlate negatively with intensity of
the signal). The more stable the complex the more it
inhibits the chemiluminescence, suggesting that the rad-
ical scavenging capacity is greatest in the intact complex.
So, it seems, although the cadmium ion does not affect
the light emission directly, it does so indirectly through
its coordination of the ligands.
ing. Thus the activities against B. megaterium correlate
with KOH with coefficients of ꢀ0.8608 for complexes
and ꢀ0.8349 for uncoordinated ligands, both significant
at p < 0.10. The corresponding values for C. tropicalis
are ꢀ0.7635 and ꢀ0.4924, which are not significant.
Because of the relationship between the inhibition of
chemiluminescence and stability, we suggest that this
procedure might make a useful screening test for newly
synthesized compounds. It is rapid (4 min per test) and
inexpensive to perform. In addition, it requires no spe-
cial apparatus and both reagents and instrumentation
are widely available. The value of the procedure as a
predictor of antibacterial activity is also worth further
investigation. But the most intriguing question raised
by this study is how the radical scavenging behaviour
of the polyphenolic Schiff bases is affected by coordina-
tion to a metal ion; this mechanism has possible rele-
vance to the design of antioxidants for use in
therapeutics and in food preservation.
References
There are nonsignificant correlations (r = 0.4410–
0.7473, n = 5) with the logarithms of the other formation
constants, but the fact that it is log(KCdL) that most
strongly correlates with the chemiluminescence is consis-
tent with the finding reported above that the CdL com-
plex is dominant at pH > 9, the prevailing conditions in
the medium of the chemiluminescence reaction. The
strongly basic conditions of the chemiluminescence reac-
tion also explain the relatively high (though not signifi-
cant) correlation (r = 0.6354, n = 5) between the effects
on chemiluminescence and the logarithms of the proton-
ation constants KOH (which range from ꢁ8–12) but not
KNH (ꢁ2, r = 0.1772, n = 5). Inhibition of chemilumines-
cence is thus associated with greater values of KOH
which lead to a greater degree of protonation at high
pH.
The effect on chemiluminescence of the cadmium
complexes is negatively correlated to a reasonable de-
gree (but nonsignificantly) with their antimicrobial
activity for B. megaterium (r = ꢀ0.6096, n = 5), but not
at all for C. tropicalis (r = ꢀ0.0316, n = 5). This seems
to arise because the effect on chemiluminescence of the
complexes is correlated to the logarithms of their forma-
tion constants KCdL and of the protonation constants
KOH of the ligands, which are negatively correlated with
antibacterial activity to a greater degree than are the ef-
fects of the complexes on chemiluminescence. Thus,
antibacterial activity of the complexes correlated with
the formation constants KCdL for B. megaterium and
the coefficient (r = ꢀ0.7980, n = 5) closely approached
significance at p = 0.10; this was not the case for the fun-
gus C. tropicalis (r = ꢀ0.3830, n = 5). The correlations
with the protonation constants KOH are even more strik-
[1] U. Casellato, P.A. Vigato, Coord. Chem. Rev. 23 (1977) 31.
[2] K. Dey, A.K. Biswas, A. Roy, Ind. J. Chem. A 20 (1981) 848.
[3] K.S. Murray, Aust. J. Chem. 58 (1978) 203.
[4] H. Doine, Bull. Chem. Soc. Jpn. 58 (1985) 1327.
[5] K. Nakajima, Y. Ando, H. Mano, M. Kojima, Inorg. Chim. Acta
274 (1998) 184.
[6] B. De Clercq, F. Verpoort, Macromolecules 35 (2002) 8943.
[7] T. Opstal, F. Verpoort, Angew. Chem. Int. Ed. 42 (2003) 2876.
[8] T. Opstal, F. Verpoort, Synlett 6 (2002) 935.
[9] S.N. Pal, S. Pal, Inorg. Chem. 40 (2001) 4807.
[10] B. De Clercq, F. Verpoort, Adv. Synth. Catal. 34 (2002) 639.
[11] B. De Clercq, F. Lefebvre, F. Verpoort, Appl. Catal. A 247 (2003)
345.
[12] M. Tumer, H. Ko¨ksal, M.K. Sßener, S. Serin, Trans. Met. Chem.
¨
24 (1999) 414.
[13] J.W. Pyrz;, A.I. Roe, L.J. Stern, J.R. Que, J. Am. Chem. Soc. 107
(1985) 614.
[14] M. Tumer, B. Erdog˘an, H. Ko¨ksal, S. Serin, M.Y. Nutku, Synth.
¨
React. Inorg. Met. Org. Chem. 28 (1998) 529.
[15] W. Zishen, L. Zhiping, Y. Zhenhuan, Trans. Met. Chem. 18
(1993) 291.
[16] J. Chakraborty, R.N. Patel, J. Ind. Chem. Soc. 73 (1996) 191.
[17] R. Klement, F. Stock, H. Elias, H. Paulus, P. Pelikan, M. Valko,
M. Muzur, Polyhedron 18 (1999) 3617.
[18] A.E. Martell, R.J. Motekaitis, The Determination and Use of
Stability Constants, VCH, Weinheim, 1988.
[19] D.F. Roswell, E.H. White, in: M.A. De Luca (Ed.), Methods in
Enzymology, Chemiluminescence of Luminol and Related
Hydrazides, LVII, Academic Press, New York, 1978, pp. 409–423.
[20] A. Rieker, K. Scheffler, R. Mayer, E. Muler, B. Narr. Ann. 10
¨
(1966) 693.
[21] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory
Chemicals, Pergamon, New York, 1992.
[22] G. Schwarzenbach, M. Flascha, Complexometric Titrations,
Methuen, London, 1957.
[23] F.J.C. Rossotti, H. Rossotti, J. Chem. Edu. 7 (42) (1965) 375.
[24] G. Gran, Acta Chem. Scand. 4 (1950) 559.
[25] C.H. Collins, P.M. Lyne, J.M. Grange, Microbiological Methods,
sixth ed., 1989, p. 410.

A. Golcu et al. / Inorganica Chimica Acta 358 (2005) 1785–1797
1797
[26] T. Gunduz, E. Kılıc¸, F. Ko¨seog˘lu, E. Canel, Anal. Chim. Acta
[40] H. Ko¨ksal, M. Dolaz, M. Tumer, S. Serin, Synth. React. Inorg.
¨
Met. Org. Chem. 31 (2001) 1141.
¨
282 (1993) 489.
¨
[27] P. May, D.R. Williams, P.W. Linder, R.G. Torrington, Talanta
29 (1982) 249.
[41] D. Martin, H.G. Hauthal, Dimethyl Sulphoxide, Van Nostrand
Reinhold, Wokingham, UK, 1975.
[28] E.P. Serjant, Potentiometry and Potentiometric Titrations, Wiley,
New York, 1984.
[42] R. Hermandez-Molina, A. Mederos, P. Gili, S. Dominquez, P.
Nunez, G. Germain, T. Debaerdemaeker, Inorg. Chim. Acta 256
(1997) 319.
[29] H.R. Neave, in: Elementary Statistical Tables, George Allen &
Unwin, London, 1979, p. 20.
[43] F. Lloret, M. Molar, J. Moratal, J. Faus, Inorg. Chim. Acta 124
(1986) 67.
[30] H.R. Neave, in: Elementary Statistical Tables, George Allen &
Unwin, London, 1979, p. 38.
[31] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81.
[44] H.W. Schneider, J. Chem. Educ. 47 (1970) 519, A new, long-
lasting luminol chemiluminescent cold light.
[32] M. Dolaz, M. Tumer, Trans. Met. Chem. 29(5) (2004) 516–522.
[45] C.Q. Zhu, L. Wang, Y.X. Li, M.H. Cheng, P.S. Wang, Chin. J.
Anal. Chem. 27 (6) (1999) 640, Chemiluminescent system of
luminol-8-hydroxy-5-quinolinesulfonic acid-cadmium(II)-hydro-
gen peroxide.
¨
[33] M. Tumer, C. C¸ elik, H. Ko¨ksal, S. Serin, Trans. Met. Chem. 24
¨
(1999) 525.
[34] M. Dolaz, M. Tu¨mer, M. Dıg˘rak, Trans. Met. Chem. 29(5) (2004)
528–536.
[46] J.C. Ren, X.Y. Huang, Anal. Chem. 73 (11) (2001) 2663,
Sensitive and universal indirect chemiluminescence detection for
capillary electrophoresis of cations using cobalt(II) as a probe
ion.
[35] A.R. Amundsen, J. Whelan, B. Bosnich, J. Am. Chem. Soc. 99
(1977) 6730.
[36] M. Tumer, Synth. React. Inorg.Met. Org. Chem. 30 (2000)
¨
´
[47] G. Merenyi, J. Lind, T.E. Eriksen, J. Biolumin. Chemolumin. 5
(1990) 53, Luminol chemiluminescence: chemistry, excitation,
emitter.
1139.
[37] R. Atkins, G. Brewer, E. Kokot, G.M. Mockler, E. Sinn, J.
Inorg. Chem. 24 (1985) 127.
[38] A. Garnier-Suillerot, J.P. Albertini, A. Collet, L. Faury, J.M.
Pastor, L.J. Tosi, J. Chem. Soc., Dalton Trans. 44 (1981) 2544.
[39] A.R. Amundsen, J. Whelan, B. Bosnich, J. Am. Chem. Soc. 99
(1977) 6730.
[48] L. Bravo, Nutr. Rev. 56 (11) (1998) 317, Polyphenols: Chemistry,
dietary sources, metabolism, and nutritional significance.
[49] D. Ryan, K. Robards, Analyst 123 (1998) 31R, Phenolic
compounds in olives.