Water Research 37 (2003) 813–822

Study of heavy metal pollution and speciation in

Buyak Menderes and Gediz river sediments
H. Akcaya,*, A. Oguzb, C. Karapirec
a
Chemistry Department, College of Education of Buca, Dokuz Eylul University, 35150 Buca-Izmir, Turkey
b
The Ohio State University, Columbus, OH 43210-1172, USA
c
Chemistry Department, Celal Bayar University, 45030 Muradiye-Manisa, Turkey

Abstract

In this study, two economically important rivers of Turkey, Gediz and Buyuk Menderes (BM) are studied to
determine their environmental pollution levels. An old analytical procedure involving sequential chemical extraction is
improved and used for the partitioning of particulate trace metals (Cu, Co, Cr, Mn, Fe, Zn, Pb and Ni). Cationic and
anionic Mn and Cr species with different phases are also determined by using leaching, extraction and ion exchange
speciation processes.
The sediment samples are analysed using graphic-furnace atomic absorption spectroscopy and differential pulse
anodic stripping voltammeter. Experimental results obtained on five replicate samples of fluvial bottom surface
sediments at the sampling points demonstrate that the relative standard deviation of the sequential extraction procedure
is generally better than 710%. The accuracy, determined by comparing total metal concentrations with the sum of the
five sequential chemical extractions, is proved to be satisfactory. The detection limits established for three standard
deviations of blank for different metals are identical and found to be 0.1 mg/kg for sediment samples and 1 ppb for
water samples. The results show that the pollution levels are significant especially for Pb, Cr, Mn and Zn in the Gediz
river and Co, Mn, and Zn in the BM river. Comparison between our results and the measurements outlined before
industralization and the beginning of the intensive pesticide applications in agricultural fields (In: Broekaert et al.,
editors, Metal speciation in the environment, vol. 23. 1989. p. 601–11; A research on the environmental pollution in the
agricultual fields and watering in the Aegean region, Ege University Research Project, No. 127 1988 (in Turkish); DSI
Water Analyses Report, Ankara, 1985 (in Turkish)) show that the pollution in these rivers is probably originated from
industrial, agricultural and domestic waste discharges.
r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Trace metals; Chemical speciation; Sediment; River pollution; Water pollution; Environment

1. Introduction Cu(II), Pb(II), Mn(II) and Cr(VI), have special im-

The determination of extractable trace metals in soil
or sediment is often used to gain an insight into chemical
speciation. The toxicity of metals depends especially on
their chemical forms rather than on their total elemental
contents, and therefore, speciation studies increasingly
gain importance. Examination of toxic heavy metals, i.e.
*Corresponding author. Tel.: +90-232-420-4882; fax: +90-
232-420-4895.
E-mail address: husamettin.akcay@deu.edu.tr (H. Akcay).
portance in environmental samples [1,2]. Thus, there is a
crucial need to develop trace element analysis methods
that allow separation of the different element species
prior to trace element analysis.
River sediments are basic components of our environ-
ment as they provide nutrients for living organisms and
serve as sinks for deleterious chemical species. Unfortu-
nately, industrial and household waste discharges—
directly or indirectly,through leakages in the sewage
systems—into water sources cause excessive pollution of
surface and underground water. Consequently, water

0043-1354/03/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 3 - 1 3 5 4 ( 0 2 ) 0 0 3 9 2 - 5
814 H. Akcay et al. / Water Research 37 (2003) 813–822
quality and irrigation value are lost. It is necessary to
know the mechanisms of the trace metals transportation
and their complexes in rivers to understand their
chemical cycles in nature.
A number of studies on the metal distributions in river
sediments and suspended particles, and on speciation of
metals have been performed [1–13].
In this work, two economically important rivers of
Turkey, Gediz and Buyuk Menderes (BM), are studied
to establish their environmental pollution levels for
some major trace elements. Tessier’s extraction proce-
dure is improved to determine the distribution of trace
metals as exchangeable, bound to carbonates, bound to
Fe–Mn oxides, bound to organic sulphurs and bound to
residue. A three steps extraction and ion-exchange
method is used for the partitioning of Cr(III) and
Cr(VI) species. The Mn species are measured as well.
2. Materials and methodes
2.1. Chemicals and reagents
All chemicals used are of analytical grade from E.
Merck, Fluka and Riedel. All solutions are prepared
using ultra-pure or double-distilled water (DDW).
2.2. Sampling
The Gediz and BM Rivers are known to be under
contamination menace by wastes derived from industrial
sources (industrial operations represent approximately
1/5 of the total industrial activity in Turkey), sewage
(there are four big cities discharges) and agricultural
activities which corespond in these regions to 35% of the
total in Turkey. Thus, data obtained from this study
have been found to be particularly suited for a variety of
reasons as the relatively high concentration of trace
metals, the significant changes in the ambiant physico-
chemical conditions, etc. During expeditions between
1996 and 1998, a total of 120 surface sediment samples
are collected from various part of the BM and Gediz
rivers (Fig. 1). Every sample consists of approximately
300 g of surface layer sediment (0–2 cm). Polyethylene
scoops and cans are used exclusively for sampling and
storage. The samples are put in ice bags during the
transportation to the laboratory and stored in a deep-
freeze unit until the drying procedure [14]. All devices
are cleaned by rinsing with pure water and kept in 0.1 M
HNO3 for several days before sampling. Samples are
dried in an oven at 551C for 2 days. In order to
normalize the variations in grain size distributions, the
dried sediment samples are sieved (EFL 2mk 3 Test
Sieve) through a 73 mm screen for the analysis. A Model
NF 1215 centrifuge is used for the solid/liquid separa-
tion procedure.
2.3. Measurement of the environmental parameters
The pH and redox potential of solutions are
performed by a Metrohm 636 pH—potentiometer and
pH 320/et-1 WTW pH meter. For the pH measurement,
the pH meter and potentiometer are calibrated with
three buffer solutions (pH=4.0; 7.0 and 9.0) and a redox
buffer solution (Eh ¼ 465 mV). River water conductivity
in samples is checked with a EDT BA 380 type of
conductometer.
Total organic matter (TOM), water temperature,
dissolved oxygen (DO), biological oxygen demand
(BOD), chemical oxygen demand (COD) and nitrogen
of ammonia are measured [15].
2.4. Speciation process
The most stable metal species have been widely
studied because of their possible toxic effects. Different
types of samples have been analysed depending on
ecological interest such as waste waters, river waters and
sediments.
Five steps sequential leaching-extraction operation is
applied to determine the trace metal species in various
environmental conditions.
The first fraction that contains exchangeable species
of the trace metals is obtained by the following
procedure: each solid sample is extracted at room
temperature for an hour with 10 ml of either magnesium
chloride solution (1 M MgCl2 buffered at pH 7.0) or
sodium acetate solution (1 M NaOAc buffered at pH
8.2) with continuous shaking. The trace metals adsorbed
on sediments are exchangeable and may be in equili-
brium with the waters ionic composition, which is
typical for sorption–desorption processes.
The second fraction contains the species bound to
carbonates. The residue from the first extraction is
leached for 5 h at room temperature with 10 ml 1 M
NaOAc adjusted to pH 5.0 with acetic acid (HOAc)
maintained.
The residue from the second extraction is extracted
with 20 ml 0.004 M NH2OH  HCl dissolved in 25% (v/v)
HOAc. The latter experiment is performed at 96731C
with occasional agitation and the time, needed for
complete dissolution of free iron oxides is determined
and used. This third fraction includes trace metal species
bound to Fe–Mn oxides (third fraction).
Three milliliters of 0.02 M HNO3 and 5 ml of 30%
H2O2 adjusted to pH 2.0 with HNO3 are added to the
residue from the third extraction. The mixture is heated
at 85721C for 2 h with occasional agitation. A second
3 ml aliquot of 30% H2O2 adjusted to pH 2.0 with
HNO3 is then added and the sample is heated again to
85721C for 3 h with intermittent agitation. After
cooling, 5 ml of 3.2 M NH4OAc in 20% (v/v) HNO3 is
added and the sample is diluted to 20 ml and agitated
 H. Akcay et al. / Water Research 37 (2003) 813–822
Fig. 1. Working area and sampling points for BM and Gediz Rivers.
continuously for 30 min. The addition of NH4OAc is
designed to prevent adsorption of the extracted metals
onto the oxidized sediment. The fourth extract thus
involves the trace metal species bound to organic matter
(fourth fraction).
Finally, the residue from the fourth extraction is
digested with a mixture of a ratio 4/1/1/1 of HNO3/HF/
HClO4/HCl for 3 h in a water-bath. The residue
dissolved entirely and it is diluted with 100 ml DDW.
The resultant solution is then analysed by spectro-
photometer to determine Fe content and flame-AAS,
graphic-furnace atomic absorption spectroscopy (GF-
AAS), and differential pulse anodic stripping voltam-
meter (DPASV) for trace metals using standard addition
techniques.
The selective extractions are conducted in centrifuge
tubes (polypropylene, 25 ml) to minimize losses of solid
materials. Between each successive extraction, separa-
tion is effected by centrifuging at 5000 rpm for an hour.
The supernatant is removed with a pipette and analysed
for trace metals, whereas the residue is washed with 8 ml
of DDW; after centrifugation for an hour, this second
supernatant is discharged.
Trace metal concentrations (Cr, Mn, Zn, Pb and Ni)
are analysed by the AAS and voltammetric methods
(GF-AAS, Perkin-Elmer, model 2380 and a GBC 933
AAS; Metrohm 746 Trace analyser, Model 384) [16,17].
Voltammetry is suited for the trace analyses of
numerous ecotoxic metals. The highly sensitive anodic-
stripping voltammetry in the differential pulse mode
(DPASV) is applicable for trace metal speciation in
natural waters [18].
815
The chromium speciation in the sediment samples
is carried out separately by three-step extraction
methods, too. The first step is the extraction with
distilled water at 901C serve to define operationally
water soluble Cr(VI). The second step is the extraction
with a phosphate buffer solutions (pH 7.0; 5.0 mM
K2HPO4 in 5.0 mM KH2PO4) that defined water soluble
plus exchangeable Cr(VI). Finally, the last step is the
extraction with a carbonate/hydroxide solution: (pH
11.8–12.3; 0.25 M NaCO3 in 0.5 M NaOH) to obtain
the total Cr(VI).
Preconcentration and recovery of Cr(VI) is carried
out with anion exchange resin column of Amberlite LA-
1. The HNO3 solution eluent is used for different
molarities (0.2–2  104 M). The pH value for retention
of Cr(VI) is important and it varies from highly to
slightly acidic. Cr(III) preconcentration carried out
using a column with polystyrene–divinylbenzene resin
in the pH range from 5 to 8.5. 0.1 M HNO3 solution is
used as an eluent.
Retention of both species, Cr(VI) and Cr(III), in
the same place is achieved using an ion-exchange
column of Dionex ASH and Dionex AG4A. Cr(III)
elution is carried out in diluted HCl solution while
Cr(VI) is removed by 0.2 M hydoxylammonium
chloride. The pH values were 1.0–2.0 for retention of
Cr(III) and 6.0–8.0 for retention of Cr(VI). The
stability evaluation of the different oxidation states
of Mn in river water is realized by using Pourbaix-
stability diagram [26]. Chromium determination was
carried out both voltammetric and atomic absorption
methods [19].
816 H. Akcay et al. / Water Research 37 (2003) 813–822

2.5. Precision and accuracy with other literature data obtained from other rivers
[10]. The concentration of Zn in BM sediment samples is
The analytical precision and quality assurance for the 85–185 ppm, whereas in Gediz the range is between 140
overall procedure is tested by five sub-samples for each and 180 ppm. If these values are compared with
of the sediment and water samples. The precision the average Zn content of surface soil, there is an
defined as the percent of variation at 5, 10 and 15 times apparent Zn pollution risk for both rivers. The chemical
the detection limit of every metal is 30, 12 and 10,
respectively. The recovery of the leach process and
digestion is checked by using standard solutions and
blanks. According to the literature, the recovery of the
HNO3 digestion is generally in the range 60–90% of the
total concentrations with the exception of Cr for which
55% is found. The experiments showed that all studied
metals are successfully extracted in the liquid phase,
when HNO3/HF/HClO4/HCl is used in 4/1/1/1 propor-
tions. The mean yield is approximately 9076%. Using
higher sediment/solution ratios could lower the detec-
tion limit, defined as twice the base-line noise. As
anticipated, the precision is generally low when the
concentration approaches the detection limit but im-
proves for higher concentrations; the coefficients of
variation as 10% or lower are typically observed for
metal concentrations amounting to five times or more of Fig. 2. Mean percent distribution of the trace metals in the
the detection limit. The standard deviations are between Gediz River’s sediments.
4% and 6% relative to those in ion-exchange column.

3. Results and discussion

The trace metal levels in the water and sediment of the

BM and Gediz Rivers are outlined in Tables 1, 3a, 3b, 4,
6a, 6b and Figs. 2 and 3. The basic environmental
parameters are summarized in Table 2. The working
area and sampling points are shown in Fig. 1. Validation
of methodology is achieved by analytical performance
characteristics using statistical control. The accuracy
and confidence of the experimental results were assured
by using certified reference materials and control
samples.
The experimental results showed that Zn mostly
occurs in the first three extraction fractions of the BM Fig. 3. Mean percent distribution of the trace metals in the BM
and Gediz river sediments. These finding is compatible River’s sediments.

Table 1
Trace metal concentrations (mg/l) in BM and Gediz Rivers water

River Metal (Fe (%), others in ppm)

Co Cr Cu Fe Mn Ni Pb Zn

BM
Summer 0.005 0.011 0.012 0.65 0.090 0.010 0.022 0.056
Winter 0.010 0.013 0.010 0.58 0.098 0.009 0.020 0.053

Gediz
Summer 0.007 0.020 0.013 1.20 0.053 0.013 0.048 0.080
Winter 0.011 0.028 0.011 1.10 0.050 0.011 0.040 0.076
 H. Akcay et al. / Water Research 37 (2003) 813–822 817

Table 2
The physical and chemical parametrs of BM and Gediz Rivers water

twater (1C) pH EC (M/cm) TOM DO BOD COD N-NH3

BM
June 16 8.3 1460 7.9 12.7 5.6 38.5 0.05
February 22 8.0 1830 6.2 10.6 6.8 46.3 0.55

Gediz
June 17 8.2 1450 8.5 13.0 5.5 41.5 0.05
February 22 7.9 1730 7.9 10.5 6.7 48.0 0.46

Table 3

Metal Sampling points

Sarayk.oy Kuyucak Yenipazar Koc-arl"y Didim

(a) Mean trace metal concentrations in BM River’s water

Cu 0.011 0.015 0.014 0.011 0.014
Zn 0.100 0.084 0.056 0.030 0.044
Fe 0.620 1.640 0.540 1.900 3.970
Mn 0.050 0.060 0.070 0.320 0.130
Cr 0.011 0.008 0.009 0.010 0.008
Pb 0.026 0.022 0.021 0.010 0.017

(b) Mean trace metal concentrations in Gediz River’s water

Alasehir Ahmetli Nif Emiralem Maltepe

Cu 0.013 0.015 0.019 0.022 0.024

Zn 0.156 0.185 0.175 0.255 0.275
Fe 0.092 0.084 0.151 0.093 0.097
Mn 0.062 0.060 0.068 0.091 0.096
Cr 0.009 0.011 0.009 0.0096 0.015
Pb 0.023 0.027 0.024 0.021 0.032

speciation data showed that Zn is in an unstable form
bounded to carbonates and Fe/Mn oxides. Therefore, it
can be concluded that Zn exchanged easily. The most
important cause of the Zn pollution is industrial wastes
(Table 3).
The Ni concentration of Gediz and BM river
sediments varies between 101–129 and 168–418 ppm,
respectively. The fractionation pattern of Ni for both
river samples indicates that Ni bounded to organic and
sulphide compounds and in the strong acid soluble
(residual) fraction. Although, these results contradict
some researches in literature, which generally show that
Ni is found abundantly in the first four fractions [9–11],
they are similar with a study conducted in northern
Greece [27]. According to that study, it is normal to find
Ni particularly in the silicate phase when its compounds’
properties are considered.
The Pb content of BM and Gediz rivers’ sediments are
between 35–75 and 105–140 ppm, respectively. In BM,
sediment Pb is found in the last two fractions (F4+F5).
However, in the Gediz river sediment, it is found
abundantly in the first three fractions
(F1+F2+F3X%52) which means that Gediz has a Pb
pollution risk (Tables 3 and 4).
While the Cu concentration of Gediz river sediment
ranges between 108 and 152 ppm, it is between 130 and
150 ppm in BM river sediment. Speciation data indicates
that although there is no pollution risk (from the percent
distribution of the fractions), the values are greater in
the Gediz river sediments. The fractionation pattern is in
correlation with other literature data [10,12].
Speciation by an ion exchange–extraction process that
allows partitioning of the total manganese concentration
into each extraction fractions (total Mn, Mn(II), total
cationic Mn, total anionic Mn, total organic bound Mn)
is summarized in Table 5. Speciation studies shows that
Mn compounds found primarily in the first three
fractions (67%). The Mn contents of these rivers
818 H. Akcay et al. / Water Research 37 (2003) 813–822

Table 4
Mean distribution of the trace metals in the BM and Gediz Rivers sediments

Co Cr Cu Fe Mn Ni Pb Zn

BM
F1 271 271 772 1071 271 1272 271 0.570.4
F2 271 271 1572 1071 5075 1672 471 271
F3 1273 8575 1872 3900750 200710 8874 1572 6675
F4 672 3073 4273 3500746 2072 5573 871 3974
F5 672 4073 5273 95007110 11076 12275 2172 1072
Mt 2974 16577 13775 1850071000 388715 315725 5478 120710

Gediz
F1 170.5 1171 371 2072 4073 772 271 170.5
F2 170.5 2271 1671 25075 1872 1272 2673 1572
F3 2573 6673 2271 67007100 310715 2273 3573 5575
F4 872 1071 4572 2200770 1072 2073 2273 6375
F5 271 8374 5272 151007100 10875 5074 3573 1272
Mt 3874 20076 14073 2550071000 510725 106710 128715 160715

F: sequential extraction fraction; Mt: total metal concentration in the sediment measured after acid leach; metal concentration has been
defined as mg-metal/kg-dry sediment.

Table 5 The Co content of BM river sediment is changed

The distribution of the Cr and Mn species in the BM and Gediz between 30 and 38 ppm, while this range is between 32
Rivers sediments and 40 ppm for Gediz river. The chemical speciation
Species Gediz (mg/kg BM (mg/kg data indicates an insignificant pollution level for both
dry sediment) dry sediment) rivers since in the first three fractions (F1+F2+F3)
amount X57% for Gediz and X68% for BM (Figs. 2
Cr (III) 6177 5577
and 3 and Tables 4, 6a and b). The Co pollution
originates from industrial waste discharges and if
Cr (VI)
Soluble 4975 2673 necessary steps not taken it can increase.
Exchangeable 100710 7077 The most stable Cr species, Cr(III), Cr(VI) and total
Total 151714 98710 Cr content in BM and Gediz rivers’ sediments are widely
studied in this research due to their possible toxic effects
Total Cr [Cr (III)+Cr (VI)] 210715 155713 [20,21]. Special attention must be paid to the sample pre-
treatment step to ensure that the original Cr species do
Mn (II) 450710 298710 not change during the analytical procedure. The Cr
Total cationic Mn 465710 298710 concentrations of the Gediz and BM river sediments are
Total anionic Mn 4775 2275
in the ranges 170–220 and 160–195 ppm, respectively.
Total Mn bound to organic 873 572
Sequential extraction studies show that in Gediz River
matter
sediments Cr species are present in the first three
Total Mn 500720 320715 fractions ((F1+F2+F3)X55%), while in the BM river
sediments they are found in higher proportions in F4
and F5. Table 5 shows the three-step extraction results
for Cr(VI). Cr(VI), which is extracted by using
carbonate-hydroxide method, is incompliance with
sediments ranges between 380 and 420 ppm for the BM column extraction results from different steps of the
and 480 and 550 ppm for the Gediz. The values are high five-step extraction. In the BM River sediments, the
for Gediz River sediments and mostly resulted from distilled water and phosphate buffer extracted essentially
anthropogenic activities. Moreover, Mn is rather poorly no measurable Cr(VI) concentration. This result indi-
absorbed from organic matter like a mechanism similar cates on the one hand that the Cr(VI) is soluble and
to Fe [5]. It might be suggested that bioavailability of reduced to Cr(III) in the presence of reductant agents as
Mn in organic matter of sediments is lower. Further- sulphides, and on the other hand that Cr(VI) pollution
more, because of industrial and agricultural activities it which originates from the human activities is not
can be concluded that exchangeable Mn species are significant [13,20–25]. However, the high Cr(VI)
found abundant in Gediz river sediments. concentration in the Gediz River sediment is very
 H. Akcay et al. / Water Research 37 (2003) 813–822 819

Table 6

Metal Fraction Sampling points

HASANKOY MALTEPE NIF EMIRALEM

(mg kg) (%) (mg/kg) (%) (mg kg) (%) (mg kg) (%)

(a) Trace metal distribution in the Gediz River’s sediments according to sampling points
Cu 1 3 2.2 3 2.9 6 4.2 6 4.1
2 16 11.9 9 8.6 11 7.7 12 8.2
3 20 14.8 12 11.4 32 22.5 34 23.3
4 45 33.3 35 33.3 40 28.2 38 26.0
5 51 37.8 46 43.8 53 37.3 56 38.4
Mt 130 113 150 152

Co 1 1 2.8 2 5 0.5 1.5 0.8 2.5

2 0.5 1.4 0.6 1.5 0.8 2.5 3 9.4
3 24 67.6 25 63.1 19 58.8 13 40.9
4 8 22.5 8 20.2 7 21.7 6 18.9
5 2 5.6 3 7.6 5 15.5 9 28.3
Mt 34 40 33 34

Zn 1 1 0.7 1 0.6 1.3 0.9 1.2 0.9

2 13 8.8 10 6.5 12 8.2 8.5 6.1
3 56 38.1 72 46.5 70 47.8 69 49.4
4 64 43.5 69 44.5 49 33.5 50 35.8
5 13 8.8 3 1.9 14 9.6 11 7.9
Mt 151 158 150 143

Fe 1 9 0.04 160 0.04 20 0.08 13 0.05

2 376 2 247 1 270 1 310 1
3 5640 24 6842 28 5776 2350 6708 26
4 5405 23 1482 6 221 0.9 1135 4
5 12455 52 15981 65 18233 74 17634 63
Mt 26290 28650 26600 26700

Ni 1 6 6 5 4 8 7 5 0.2
2 9 9 8 7 18 14 12 12
3 20 20 20 20 26 21 23 21
4 18 18 22 22 19 16 19 18
5 48 47 47 46 53 43 47 44
Mt 110 106 129 113

Mn 1 39 8.1 40 7.6 40 7.5 32 6.6

2 19 3.9 22 4.2 20 3.8 30 6.2
3 302 62.9 320 65.1 325 61.2 303 62.5
4 15 3.1 34 6.5 28 5.3 16 3.3
5 105 21.9 108 20.6 118 22.2 104 21.4
Mt 490 535 540 500

Cr 1 10 5.9 11 6.0 13 6.3 10 5.9

2 21 12.4 22 12.0 27 12.98 22 12.9
3 56 32.9 66 36.1 75 36.1 60 35.3
4 10 5.9 17 9.3 15 7.2 11 6.5
5 73 42.9 66 36.1 78 37.5 67 39.4
Mt 175 190 212 175

Pb 1 2 1 3 3 17 1 4 3
2 24 18 15 14 30 21 25 19
3 52 40 31 30 40 28 41 31
4 22 17 20 20 25 17 22 17
5 31 24 35 34 41 29 40 30
Mt 139 107 149 137
820 H. Akcay et al. / Water Research 37 (2003) 813–822

(b) Trace metal distribution in the BM River’s sediments according to sampling points

SARAYKOY KUYUCAK YENIPAZAR KOCARLI DIDIM

(mg/kg) (%) (mg/kg) (%) (mg/kg) (%) (mg/kg) (%) (mg/kg) (%)

Cu 1 7 5.4 6 4.5 8 5.9 7 4.9 8 5.8

2 13 10.0 15 11.4 16 11.8 17 11.97 16 11.6
3 19 14.6 18 13.6 19 13.97 20 14.1 19 13.8
4 40 30.8 42 31.8 43 31.6 45 31.7 44 31.9
5 51 39.2 51 38.6 50 36.8 53 37.3 51 36.9
Mt 134 135 138 145 140

Co 1 2 6.8 2 5.7 1 3.4 2 5.5 1 3.1

2 1.5 5.1 2 5.7 2 6.9 2 5.5 2 6.2
3 13 44.1 15 62.9 13 44.8 15 41.7 13 40.6
4 6 20.3 8 22.9 8 27.6 8 22.2 8 25.0
5 7 23.7 8 22.9 5 17.2 9 25.0 8 25.0
Mt 31 37 31 38 33

Zn 1 0.20 0.18 0.24 0.26 0.25 0.18 0.23 0.13 0.21 0.17
2 0.48 0.44 3 3 8 6 10 5 1 0.98
3 67 62 53 59 71 52 89 49 71 56
4 28 26 27 30 53 39 65 35 45 36
5 13 12 7 8 5 3 19 10 8 7
Mt 112 93 132 181 132

Fe 1 9 0.05 8 0.05 8 0.04 8 0.06 7 0.08

2 9 0.05 0 0 8 0.04 313 2 7 0.08
3 3099 19 3109 20 4903 26 3745 27 5221 28
4 3570 23 3404 34 4164 22 203 1 236 1
5 9274 68 8577 57 9772 52 9535 70 12984 70
Mt 23,880 21,412 25,476 24,871 24,637

Ni 1 13 3 10 5 12 4 12 5 11 4
2 18 4 15 7 20 7 18 7 18 6
3 79 17 84 36 122 40 95 36 122 44
4 79 17 27 12 30 10 24 9 24 9
5 169 37 94 41 122 40 114 43 106 38
Mt 418 168 338 279 315

Mn 1 2 0.5 3 0.8 3 0.7 2 0.5 3 0.8

2 48 12.5 51 13.0 55 13.4 52 13.0 55 14.4
3 200 52.1 195 49.6 210 51.3 205 51.1 190 49.7
4 22 5.7 25 6.4 28 6.8 26 6.5 24 6.3
5 112 29.2 119 30.3 113 27.6 116 28.9 110 28.8
Mt 390 400 415 405 385

Cr 1 2 1.3 2 1.0 2 1.1 2 1.1 2 1.1

2 2 1.3 2 1.0 2 1.1 3 1.7 3 1.6
3 80 50.3 85 44.3 85 46.4 85 46.96 89 48.6
4 30 18.9 35 18.2 38 20.8 35 19.3 33 18.0
5 45 28.3 68 35.4 56 30.6 56 30.9 56 30.6
Mt 160 195 185 190 188

Pb 1 2 4 2 4 0.98 2 4 5.1 0.57 1

2 3 8 3 6 3 7 8 10.3 2 6
3 15 34 12 31 15 31 35 44.9 11 25
4 7 16 6 15 9 19 17 21.8 9 21
5 17 39 17 42 19 40 14 17.9 20 47
Mt 46 36 45 75 44
 H. Akcay et al. / Water Research 37 (2003) 813–822
representative to justify the pollution that is originated
from human activities. The difference of the Cr(VI)
concentration of five-step sequential and of three-step
extraction processes (analysis) showes that the reduction
tendency for Cr(VI) is considerable under ambient
conditions in the river. These results indicate that
Cr(VI), which has high toxicity, is present in consider-
able levels in the Gediz river sediments. This situation is
significant because of the wastewater discharges of
leather facilities, located nearby the river.
In speciation studies of Fe in these river sediments, no
pollution risk is determined. The Fe content of Gediz
and BM Rivers sediments range between 26,000–28,500
and 21,000–25,500 ppm, respectively. These results show
a relationship with the geochemical structure of the
Menderes massif [4]. According to the results of Fe(II)/
Fe(III) separation and speciation, different Fe(II) and
Fe(III) forms are located especially in bound to Fe and
Mn oxides (F3) and bound to organic matter (F4). This
fact depends on the redox potential, electrical conduc-
tivity (EC), (pH), i.e.
4. Conclusion
In natural media, trace metals undergo numerous
changes during their transport due to dissolution,
precipitation and sorption phenomena. Trace element
concentrations of river basins depend on not only
industrial and household waste inputs but also on the
geochemical composition of the area. High concentra-
tions of Zn, Mn and Cr are thought to have resulted
from anthropogenic influences, practically from industry
and pesticides used in agriculture, and are found to
present a pollution risk. Fe concentrations in the BM
and Gediz rivers depend on the geochemical structure of
the Menderes massif, which consists of Fe-rich meta-
morphic rocks. Pb enrichment in Gediz river sediments
has an exchangeable character and represents pollution
potential in this river. As in the Ni speciation study, this
metal is found bound to silicates. Thus, this can be
concluded that both rivers have no anthropogenic
source of Ni pollution. The Cu contents of Gediz River
is higher than the Cu content of BM River, especially in
Kemalpasa-Manisa region and this is a potential
pollution risk for this region. Speciation studies prove
that the industrial wastes may cause this pollution.
Leaching, extraction and ion-exchange studies shows
that Mn compounds which are pollution indicators
occur primarily in the first three fractions in Gediz river.
It might be suggested that bioavailability of Mn in
organic matter of sediments is lower; on the other hand
exchangeable manganese species are abundant especially
in the Gediz river sediments. This result shows that Mn
pollution might be originated from a kind of pesticide,
which contains Mn and is used widely in this region.
821
Cr analyses indicate the pollution presence in Gediz
River. Especially, high Cr(VI) value confirmes that the
pollution originated from industrial activities is crucial.
However, in the BM River sediments Cr species are
located mainly in the fourth and fifth fractions, which
may originate from the geochemical composition of this
region. The speciation data for Co suggests a weak
pollution risk in both rivers.
Acknowledgements
The authors wish to express their gratitude to the
Turkish Scientific and Technical Council (Project no.
DEBCAG-570), Dokuz Eylul University Research Fund
(Project no. 0922.97.01.42) who hold a share in the
partial funding of the indicated studies.
References
[1] Novotry K, Turzikova A, Komarek J. Speciation of
copper, lead and cadmium in aquatic systems by circulat-
ing dialysis combined with flame AAS. Fresenius J Anal
Chem 2000;366:209–12.
[2] Abollino O, Aceto M, Sarzanini C, Mentasti E. The
retention of metal species by different solid sorbents.
Mechanisms for heavy metal speciation by sequential
three-column uptake. Anal Chim Acta 2000;411:223–37.
[3] Chen MH, Wu HT. Copper, cadmium and lead in
sediments from the Kauhsiung River and its harbour area.
Mar Pollut Bull 1995;30(12):879–84.
[4] Ergin M, Bodur MN, Ediger V, Yemenicioqlu ! S, Okyar M,
Kubilay NN. Sources and dispersal of heavy metals in
surface sediments along the eastern Aegean shelf. Boll
Oceanologia Teor Appl 1993;XI(1):27–44.
[5] Deutsch F, Hoffmann P, Ortner HM. Analytical char-
acterization of manganese in rainwater and snow samples.
Fresenius J Anal Chem 1997;357:105–11.
[6] Kindler FM, Sevim HE. Heavy metals in sediments of
Turkish River systems. In: Broekaert JAC, et al., editors,
Metal speciation in the environment, NATO ASI Series,
vol. 23. 1989. p. 601–11.
[7] Trocine RP, Thefry JH. Metal concentrations in sediment
water and clams from Indian river lagoon of Florida. Mar
Pollut Bull 1996;32(10):754–9.
[8] Tessier A, Campbell PGC, Bission M. Sequential extrac-
tion procedure for the speciation of particulate trace
metals. Anal Chem 1979;51(7):844.
[9] Bodog I, Polyak K, Hlavay JC. Determination of heavy
metals in lake and river sediments. Int J Environ Anal
Chem 1996;79:79–94.
[10] Ure M, Thomas R, Littlejohn D. Ammonium acetate
extracts and their analysis for the speciation of metal ions
in soils and sediments. Int J Environ Anal Chem 1993;
51:65–8.
[11] Pardo R, Barrado R, Castrillejo Y, Velasco MA, Vega M.
Study of the contents and speciation of heavy metals in
river sediments by factor analysis. Anal Lett 1993;1719–39.
822 H. Akcay et al. / Water Research 37 (2003) 813–822
[12] Boughriet A, Ouddane B, Fisher JC, Wartel M, Leman G.
Variability of dissolved Mn and Zn in the seine estuary and
chemical speciation of these metals in suspended matter.
Water Res 1992;26(10):1359–78.
[13] Rowlatt SM, Lovelli DR. Lead, Zinc and Chromium in
sediments around England and Wales. Mar Pollut Bull
1994;28(5):324–9.
[14] UNEP, Reference methods for marine pollution studies,
United Nations Environment Programme. Regional Seas
1985; 31–9.
[15] Radojevic M, Bashkin VN. Practical environmental
analysis. Bodmin, UK: RSC, 1999.
[16] Rengaraj S, Yeon K, Kang S, Lee J, Kim K, Moon S.
Studies on adsorptive removal of Co(II), Cr(III) and Ni(II)
by IRN 77 cation-exchange resin. J Hazardous Mater
2002;B92:185–98.
.
[17] Gruter UM, Kresimon J, Hirner AV. A new HG/LT-GC/
ICP-MS multi-element speciation technique for real sam-
ples in different matrices. Fresenius J Anal Chem 2000;
368:67–72.
[18] Henze G. Applications of polarographic, voltammetric
techniques in environmental analysis. In: Metal speciation
in the environment, NATO ASI series, Ecological Sciences,
vol. 23, 1990. p. 391–408.
[19] Korolczuk M, Grabarczyk M. Selective voltammetric
determiantion of chromium(VI) with DTPA and nitrate.
Fresenius J Anal Chem 1988;363:421–3.
[20] Wilson BE, Gondy A. Effects of chromium supplementa-
tion on fasting insulin levels and lipid parameters in
healthy, non-obse young subjects. Diabetes Res Clin
Practice 1995;28:179–84.
[21] Vitale RJ, Mussoline GR, Rinehimer KA, Petura JC,
James BR. Extraction of sparingly soluble chromate form
soils: evaluation of methods and Eh–pH effects. Environ
Sci Technol 1997;31:390–4.
[22] Marqu!es MJ, Salvador A, Morales-Rubio AE, de la
Guardia M. Analytical methodologies for chromium
speciation in solid matrices: a survey of literature.
Fresenius J Anal Chem 1998;362:239–48.
[23] Baraj B, Martinez M, Merkoci A, Niencheski LFH, Soares
JA. Comparison of chromium speciation by CZE and ion
exchange followed by AAS. Fresenius J Anal Chem 2000;
367:12–6.
[24] DSI Annual Water Analyses Report, Ankara, 1985 (in
Turkish).
[25] Saatc"y F, Hakerlerler H, Tuncay H, Okur B. A research on
the environmental Pollution in the agricultural fields and
watering in the Aegean region, Ege University Research
Project, No. 127 1988 (in Turkish).
[26] Pourbaix M. Atlas d’!equilibres e! lectrochimiques. Paris:
Gauthier-villars.
[27] Sakellariadau F. Proceedings of the International Con-
ference Chemicals in the Environment, 1986, p. 578.