Annali di Chimica, 97, 2007, by Società Chimica Italiana 983

METHOD VALIDATION FOR THE DETERMINATION OF LEAD IN RAW

COW’S MILK BY ELECTROTHERMAL ATOMIC ABSORPTION
SPECTROMETRY

Deniz YURTSEVER SARICA1, Ali Rehber TÜRKER(°)2

1 TÜBİTAK/ATAL Scientific and Technological Research Council of

Turkey/Ankara Test and Analysis Laboratory, Beşevler, Ankara 06530, Turkey
2 Department of Chemistry, Science and Art Faculty, Gazi University, 06500,
Ankara, Turkey

Summary - In this study, lead in raw cow’s milk has been determined by validated
electrothermal atomic absorption spectrometry (ETAAS) with Zeeman-effect
background correction. Maximum pyrolysis and optimum atomization temperatures
of lead were determined in the presence of modifiers. Pd + Mg(NO3)2 has been found
a powerful modifier mixture for the determination of lead. The analytical parameters
of the method such as limit of detection, limit of quantification and the effect of
interfering ions have been investigated. The detection limit (3σ) achieved by the
method was calculated to be 0.62 ng/mL for Pb. Repeatability of the method
evaluated as the relative standard deviation of 16-17 replicates using 5 ng/mL, under
optimum experimental conditions were about 1.5 % for synthetic sample solution
and about 3 % for real sample (N=5). The described method has been validated by
analyzing certified reference material (BCR-CRM 150) and by comparing the results
with those obtained by inductively coupled plasma mass spectrometry (ICP-MS).
The validated method was applied to raw cow’s milk samples produced in 7 different
regions of Turkey in 2003-2004. Raw cow’s milk contained a mean (range) of 31.4
(2.5 – 313) µg kg-1 lead with a relative error below 2%.

INTRODUCTION

Rapid industrialization, use of agricultural chemicals and automobile exhaust fumes has
increased the toxic heavy metals in the environment and their direct effect to the air, water and
food. Lead is a common polluting heavy metal mainly because of its widespread industrial use,
especially fuel additive and natural abundance.1 The main sources of lead contamination are
smelting works, application of wastewater treatment sludge to soil, transportation, rain, snow, hail
and other. Also sources of lead pollution can be extended by paints, lead wastes, cell batteries and
lead solders.2 Ingested lead accumulates in the different organs of the body. It is especially
dangerous because it can damage the brain and peripheral nerves. Lead affects everyone, but
children are at a higher risk as they are still at growing stage. Lead may also affect brain function by

(°)
Corresponding author; fax: (+90)-312-212 2279; e-mail: aturker@gazi.edu.tr
984 SARICA and TÜRKER

interfering with neurotransmitter release and synapse formation.3 Lead is known to damage the
kidney, liver and reproductive system, basic cellular processes and brain function4. Known as one
of the ubiquitous harmful metals, lead is found usually in very low levels in milk and dairy
products, except when animals have consumed contaminated feed.5 Milk contains both essential and
nonessential trace elements. Lead is nonessential, potentially toxic heavy metal even at very low
concentrations6. Concentration of lead in milk is a matter of special concern because in Turkey cow
milk is one of the major dietary constituents for infants from the fourth-sixth month of life onwards.
For public health, the concentrations of Pb in milk need to be monitored. In Turkey, milk is
consumed both raw and in pasteurized products in large quantities, and it is important that
contaminant levels remain low to minimize exposure.
The most commonly used methods for Pb determination in milk and in milk products,
among which at least 100 of them are produced by the International Dairy Federation, IDF.
(Falomir et al.), determine lead in human milk by electrothermal atomic absorption spectrometry
directly.1 Eleven minor and trace elements including lead were determined in thirty two breast milk
samples by ICP-MS.7 Karadjova et al.8 used combined analytical procedures consisting of wet
digestion step followed by instrumental determination – differential pulse cathodic stripping
voltammetry (DPCSV) or electrothermal atomic absorption spectrometry (ETAAS) - for the
determination of Cd, Co, Cr, Cu, Fe, Ni and Pb in milk, cheese and chocolate. Jeng et al.9
determined lead in raw milk by graphite furnace atomic absorption spectrophotometer (GFAAS)
and the mean Pb content was 2.03 µg L-1. Barreira and Arribas10 described the determination of Pb
in cow’s milk by differential pulse striping voltammetry (DPSV). Zorica and Liljana11 also
described DPSV for the determination of Pb in milk. According to EU legislations,12 by considering
the possible LOD levels obtained, ET-AAS should be the preferred technique for determinations,
and be used in many countries for this purpose.9, 13 However, probably due to highly contaminated
sampling districts, flame AAS for Pb content was used as well14, 15 not only in milk but also for
different foodstuffs including sweets, cheese and chocolate.8, 16
In this study, optimized electrothermal atomic absorption spectrometry (ET-AAS) method
was validated by using certified reference materials. The validated method was applied for the
determination of lead in raw cow’s milk of 7 different dairy farms in Turkey. The accuracy of the
method was also checked by ICP-MS determination. Samples of bulk milk were analyzed after
microwave digestion for their lead contents in relationship to various periods of the year.

EXPERIMENTAL

Apparatus
All atomic absorption measurements were carried out with a Perkin Elmer AAnalyst 800
model graphite furnace atomic absorption spectrometer (GF-AAS) equipped with a longitudinal
Zeeman Effect background corrector and THGA tube, an auto sampler (AS 800) and an automatic
data processor. Single element hollow cathode lamp of lead was used as radiation source. Argon
was used as a carrier gas during all the stages except for atomization. The instrumental operating
conditions and working conditions for the determination of lead in raw cow’s milk samples were
shown in TABLES 1 and 2, respectively. For the detection of lead in milk as a method confirmation
Agilent 7500a Model inductively coupled plasma mass spectrometry (ICP-MS) was used. For
sample digestion, a Milestone MLS-1200 Mega Microwave digestion system with MDR
Technology was used both before and after separation of the fat of whole milk by centrifugation
using Heraeus Sepatech Labofuge 200 model centrifuge.
 Determination of lead in milk by ETAAS 985

TABLE 1. - Instrumental operating conditions applied for the determination of lead

Wavelength, nm 283.3
Slit width, nm 0.5
Lamp Pb Hollow Cathode Lamp at 10 mA
Carrier/purge gas Ar (99.999 % (v/v))
Integration time, s 5
Graphite tube Pyro/Platform (THGA)
Matrix modifiers 0.003 mg Mg(NO3)2 and 0.005 mg Pd mixture
Volume of sample, µL 20
Volume of modifier, µL 10 (5 µL from each)
Signal mode Peak area
Background correction system Zeeman effect

TABLE 2. - Instrumental operating conditions applied for the determination of lead

Steps Temperature, Ramp time, s Hold Gas (Ar)

°C time, s flow,
mL min-1
Drying 1 110 15 20 250
Drying 2 130 15 40 250
Ashing 975 25 20 250
Atomization 1600 0 5 0
Clean up 2400 1 3 250

Reagents and solutions

All reagents were of analytical grade. Milli-Q water (Millipore, MA, USA) was used
throughout. HNO3 (65 % J.T. Baker) and H2O2 (30 % Sigma) were used. Working solutions were
prepared by appropriate dilution of the lead stock solution (1003 ± 2 µg mL-1, Inorganic Ventures
Inc.) with 0.2 % (m/m) HNO3 immediately before use. Matrix modifier stock solutions were
prepared by dissolving ammonium dihydrogen orthophosphate (Inorganic Ventures Inc. USA) as
40000 µg mL-1 of PO43- in water, palladium nitrate (Inorganic Ventures Inc. USA) as 10000 µg mL-1
of Pd in 10 % HNO3 and Mg(NO3)2.6H2O (MERCK) as 6000 µg mL-1 Mg(NO3)2 in 0.2 % HNO3.
All stock solutions were stored in polyethylene flasks in the dark and at 4 oC. Certified reference
material (BCR-CRM 150 skimmed milk powder) and two reference materials (FAPAS TO 739 and
FAPAS TO 752) were used for validation.
Solutions were prepared in calibrated borosilicate volumetric flasks and immediately
transferred into high density polyethylene bottles. All used laboratory ware was cleaned by
successively soaking in 10 % and 20 % (v/v) HNO3 for 24 h and 12 h, respectively, then rinsed with
water obtained from Milli-Q for several times in order to avoid any signal suppression problem due
to HNO3 residue.17 Then all acid-cleaned labware was dipped into 3% EDTA solution for overnight
and rinsed with deionised water several times.

Collection and preparation of raw cow’s milk samples

Raw cow’s milk samples were collected from 7 dairy farms located in different regions of
Turkey 8 times during the years 2003-2004. All collections were made directly from the trucks in
986 SARICA and TÜRKER

pre-acid cleaned high density polyethylene bottles with screw top because of the relatively great
possibility of sample contamination regarding lead.
As soon as the milk samples arrived at the laboratory they were first filtered by gravity
through Whatman No 4 filter paper prior to analysis to remove particulate material. Then filtered
samples were homogenized and agitated with vortex, and 2/3 of samples were centrifuged via
Heraeus Sepatech Labofuge 200 model centrifuge at 5000 rpm for 15 min at room temperature to
separate fat and to get skimmed milk. 15-20 mL of skimmed milk sample was frozen at -18 °C and
kept freeze-dried at least 12-14 hours prior to analysis. When the samples were to be analyzed, the
milks were placed at room temperature and thawed for overnight.

Digestion of samples
Due to the microwave digestion recovery results obtained via spiked blank samples
skimmed milk were used throughout the study. Skimmed milk prepared by a procedure given in the
section of “Collection and preparation of raw cow’s milk samples” was digested in a MILESTONE
MLS-1200 Model Microwave digestion system. In order to achieve complete digestion several
solvents (acid mixtures) were experimentally tried to dissolve CRM BCR 150 milk powder by
microwave digestion. Higher recovery values were obtained for the procedure described below and
it was applied for digestion in subsequent experiments.

TABLE 3. - Microwave programme used for the mineralization of the milk samples

Step Power, W Time, min

1 250 1
2 0 2
3 250 2
4 700 7.5
5 0 15

In order to get successful digestion and removal of organic parts, homogenized samples (0.5 g
and 1.0 g) were placed into a PTFE digestion vessel and 2.5 mL of nitric acid (65 % (m/m)) and 2.5
mL of 30% hydrogen peroxide were added. The vessels were immediately closed after the addition
of the oxidants. The microwave program used is detailed in Table 3. Microwave assisted
mineralized samples were brought to a volume of 25.0 mL by deionized (Milli-Q) water. Then,
these digested samples were spiked with lead for recovery experiments. It was necessary for the
diluted samples to wait for at least 30 minutes so that all species were equilibrated with the solution.
Then the same digestion procedure was applied for CRM and RMs.

Procedure
The digested samples were analyzed directly by electrothermal atomic absorption spectrometry
using calibration graph method. Calibration against working standard solutions of lead was
performed in the analytical range of 5–180 ng mL-1. Modifier components at the optimum mass and
mass ratio were added to the working standard and sample solutions. Type of matrix modifier and
volume of sample were assayed to select the best analytical conditions. Magnesium nitrate,
ammonium phosphate, palladium chloride and ascorbic acid, citric acid and their mixtures are the
most common matrix modifiers for ET-AAS and they bring considerably high ashing (pyrolysis)
temperature possibility by changing the atomization chemistry.18 Twenty microliters of lead
standard solution (0.5–100 ng mL-1) or sample solution and totally 10 µL of modifier solution
 Determination of lead in milk by ETAAS 987

(0.003 mg of Mg(NO3)2, 0.005 mg of Pd) were injected into the graphite furnace. The instrumental
conditions applied are reported in TABLE 1.

RESULTS AND DISCUSSION

In order to achieve the accurate determination of lead in raw milk using ET-AAS, selection
of the relevant matrix modifier and its concentration together with the ashing and atomization
temperatures were all experimentally optimized.

Optimization of analytical conditions

The graphite furnace program (drying, ashing, atomization temperatures, and ramp and hold
time) were optimized in order to obtain the maximum integrated absorbance signals (TABLE 2).
Effect of matrix modifiers on the ashing temperature and characteristic mass (mo)20 of lead were
shown in TABLE 4 (Characteristic mass is defined as the mass of analyte, which yields a signal
equals to a 0.0044 absorbance). As can be seen from the table the matrix modifier makes significant
contribution to the sensitivity. By considering the tolerance to interferences, thermal stabilization,
characteristic mass and the analysis of digested milk samples, the analytical performances of the
matrix modifiers were compared and selected. The best result was obtained with a mixture of 0.003
mg Mg(NO3)2 and 0.005 mg Pd and this modifier combination was used throughout the studies.
Other modifiers either gave too high backgrounds or non sufficient performance at high ashing
temperatures above 950 °C. It is suitable to use an ashing temperature as high as possible to remove
the matrix efficiently and a temperature at least 900 °C should be aimed for analytes in food and
biological samples.19 Therefore, the pyrolysis temperature was used as 975 °C for subsequent
experiments.
The ashing temperature was kept constant whereas the atomization temperature was scanned
with 100 °C increments at first and in order to get the exact temperature, experiments were carried
out with 25 °C increments around the most probable temperature at which Pb peak is obtained as
sharp enough. On the other hand, above 1700 °C signal was splitted. Accordingly, the atomization
temperature was adjusted as 1600 °C for the rest of the analyses.

Analytical features
The determination of lead in sample solutions was performed with the instrumental
parameters recommended by the manufacturer and using optimum experimental conditions
determined (TABLES 1 and 2) by using calibration on the base of single element solutions.
Calibration graph for lead was linear and correlation coefficient was about 0.9994.
The sensitivity of the proposed ETAAS method is expressed by means of the limit of
detection (LOD) and characteristic mass. LOD signal is calculated by taking the mean of blank
signals plus three times the standard deviation of the blank signals and was found as 0.62 ng/mL for
Pb. Limit of detection (LOD) may be formulated as follows:21

LOD = XLOD/m =[Xbl + 3σbl]/m

Where m is the slope of calibration curve. Limit of quantification (LOQ) is accepted that it
usually equals about three times the LOD value. Therefore, LOQ of Pb is 1.86 ng/mL.
988 SARICA and TÜRKER

TABLE 4. - Effect of matrix modifiers on the ashing temperature and characteristic

mass of lead

Matrix modifier Amount of matrix Ashing Characteristic mass, mo,

modifier temperature pg
°C

None --- 600 32

Mg(NO3)2 0.003 mg Mg(NO3)2 850 41

NH4H2PO4 0.050 mg NH4H2PO4 950 36

Mg(NO3)2 + 0.050 mg NH4H2PO4 and 1000 34

NH4H2PO4 0.003 mg Mg(NO3)2

Pd +Mg(NO3)2 0.003 mg Mg(NO3)2 975 28

and
0.005 mg Pd

Pd + NH4H2PO4 0.005 mg Pd 900 44

and
0.050 mg NH4H2PO4

Effect of diverse ions

The effects of major common ions (some alkaline, alkaline earth and transition metals) that
are already present in cow’s milk samples on the recovery of lead have also been studied.22 The
interference study results were shown in TABLE 5. As can be seen from the table, tested elements
did not show any interfering effect at mentioned concentrations.
 Determination of lead in milk by ETAAS 989

TABLE 5. - Effect of various ions present in cow’s milk on recovery of lead (lead
concentration, 50 µg L-1)

Interfering Concentration Recovery a) of

ions (µg g-1) Pb, %

Ca2+ - 96
1100 93
2+
Mg - 94
100 90

Fe3+ - 98
5 96
Zn2+ - 98
3 98
Cu2+ - 102
0.5 97
Na+ - 96
500 95
K+ - 98
1500 97
a)
Mean of three replicates

Validation of the method

In order to check the accuracy and to validate the method, a certified reference material
(BCR-CRM 150 skimmed milk powder) and two reference materials (FAPAS T 0739 and FAPAS
T 0752) were analyzed. The standard addition calibration method was carried out with five replicate
measurements. The lead contents in milk powder (BCR-CRM 150 and two RMs) determined and
shown in TABLE 6 were close to the certified values. During analyses, a blank and a known lead
standard were run after every 7 samples for accuracy assurance. To assess laboratory and method
bias in combination, the certified reference material (CRM) was analyzed. The estimate of the
combined bias is the difference between the mean result and the certified value. It is an acceptable
level for these concentrations.
The accuracy of the proposed method has also been checked by analyzing spiked skimmed
milk samples and by using ICP-MS as another method of analysis. As can be seen from the TABLE
6, determination of lead could be performed accurately (relative error < 5 %) and precisely (relative
standard deviation < 3 %) in real samples.
The relative standard deviation of 16-17 replicates on a 5 ng mL-1 solution, used to evaluate
the repeatability of the method, was ~1.5 % for Pb in synthetic sample solution. After every 7
samples, two pre-analyzed sample was rechecked randomly for repeatability evaluation. Results
showed that the percentage of repeatability was less than 3 % which is recognized internationally
within the range of quality control standards.
990 SARICA and TÜRKER

TABLE 6. - Determination of lead in certified reference material (BCR-CRM 150

Skim Milk Powder) and in reference materials (FAPAS T 0739 and
FAPAS T 0752) by standard additions method

Specified value Founda Relative error,

µg kg-1 µg kg-1 %

BCR-CRM 150
Skim Milk Powder 1000 ± 40 998 ± 8 - 0.2
FAPAS T 0739*
Milk Powder 126 125 ± 1 -0.8

FAPAS T 0752*
Milk Powder 102 104 ± 17 +2.0
a
Results are calculated as x ± s for N=5
*
Assigned values of FAPAS T 0739 and 0752.

TABLE 7. - Determination of Pb in skimmed milk samples by ETAAS

Sample code Added, Founda), Relative

ng mL-1 ng mL-1 error, %

A
Oct 2003 - 3 - 5.3 ± 0.1 -
10 16.0 ± 0.2 +4.6

20 24.9 ± 0.8 -1.6

D
Oct 2003 - 3 - 17.2 ± 0.6 -
10 28.1 ± 0.7 +3.3

20 37 ± 1 -0.5
a)
Results are calculated as x ± s for N=5

Application of the method to real samples

In order to demonstrate the applicability of the proposed method and to investigate lead
contamination in raw cow milk in Turkey, the optimized technique was applied to real samples. For
this purpose, 54 raw cow’s milk samples, collected between January 2003 and December 2004 from
7 different dairy farms of Turkey, were used. Raw cow’s milk collected in May 2004 was also
analyzed by ICP-MS for the determination of lead content. Summary of the obtained results were
given in TABLE 8. There is a good harmony between the results obtained by the proposed method
and by ICP-MS. In this study the 54 samples, 21 milk samples containing more than 20 µg kg-1 of
lead were found. The highest lead concentration was 313 µg kg-1. It is an extreme value and could
 Determination of lead in milk by ETAAS 991

not be explained without any further monitoring of the sample or region. Therefore this should be
treated as an outlier even no needs to apply Q-test and so forth. Actually among the high Pb
containing samples, 14 of them are too high regarding the upper limit (The CE Regulation
no:2001/46612 establishes a maximum residue limit (MRL) value for Pb in milk as 0.020 µg kg-1).
The results found for lead were compared with those given by other authors. Munoz and
Palmero23 reported that lead concentration of milk is 27.35 µg kg–1. In another study,24 the
concentration of lead of milk sample from distinct regions of Turkey was found between 22.1 –
59.2 µg L–1. Tüzen and Soylak25 reported that the concentration of lead in milk was found below the
limit of detection of the method (LOD, 0.25 µg L–1). Lead concentration of the cow’s milk and raw
sheep milk collected from Turkey has been reported as 18 ±2.8 and 20 ±3 µg L–1, respectively by
Yaman.26 In another study,27 the lead level of the milk powder samples was found in the range of
0.98-4.45 µg L–1.

CONCLUSION

The proposed procedure provides a simple, selective, accurate and precise method for the
determination of lead in milk matrices. The proposed ET-AAS technique was successfully applied
for the determination of lead in raw cow’s milk samples. From the values of the analytical
parameters of the proposed method it can be concluded that it is useful for determining lead in raw
cow’s milk. It provides reliable determination of lead in skimmed milk samples (relative error and
relative standard deviation are below 5 %). The detection limit achieved was also satisfactory for
the samples studied.

TABLE 8. - Resultsa) obtained for the determination of Pb in skimmed milk by

ETAAS using the optimum conditions and confirmed with the ICP-
MS method

Codes A B C D E F G

Sample
Dates
Aug 2003- 10.9 ± 0.3 10.3 ± 0.2 11.4 ± 0.2 20.3 ± 0.4 21.5 ± 0.7 16.0 ± 0.6 No
1 sample
Sep 2003- 2.52 ± 28.8 ± 0.7 5.1 ± 0.2 7.7 ± 0.1 5.2 ± 0.2 32.2 ± 0.7 No
2 0.03 sample
Oct 2003- 5.3 ± 0.1 15.2 ± 0.4 13.4 ± 0.1 17.2 ± 0.6 4.1 ± 0.1 19.5 ± 0.4 16.8 ±
3 0.2
Nov 2003- 6.2 ± 0.2 18.1 ± 0.3 29.9 ± 0.3 15.1 ± 0.4 10.3 ± 0.2 22.6 ± 0.6 21.0 ±
4 0.8
Jan 2004- 49 ± 1 52 ± 1 36.4 ± 0.7 16.4 ± 0.2 51 ± 1 38.1 ± 0.3 42.5 ±
5 0.5
Feb 2004- 6.8 ± 0.2 9.5 ± 0.2 19.1 ± 0.4 13.7 ± 0.6 17.9 ± 0.4 11.0 ± 0.3 8.9 ± 0.2
6
Apr 2004-7 47 ± 1 52 ± 1 60 ± 2 39.9 ± 0.5 54 ± 2 49 ± 1 11.9 ±
0.2
May 2004- 84 ± 1 26 ± 1 64 ± 1 15.4 ± 0.3 15.5 ± 0.8 26 ± 1 313 ± 12
8
992 SARICA and TÜRKER

TABLE 8. - (Continued)

ICP-MS b)
results 83.3 ± 0.8 26.9 ± 0.2 64.1 ± 0.1 15.59 ± 16.0 ± 0.1 25.8 ± 0.3 307 ± 9
0.09
Mean ( x ) 26.6 26.5 29.9 18.2 22.5 26.8 69.1

Median 8.9 22.2 24.5 15.9 16.7 24.3 18.9

Range (w) 2.5 - 84 9.5 - 52 5.1 - 64.3 7.8 - 39.9 4.1 - 54.5 11 - 49 8.9 - 313
a)
Results are calculated as x ± s for N=3 in µg/kg; for each one of 3 parallel samples.
b)
For sample of May 2004. Not included in mean, median and range calculations.

Received April 16th, 2007

REFERENCES

1) P. Falomir, A. Alegria, R. Barbera, R. Farre and M.J. Lagarda, Food Chem. 64,111 (1999)
2) S. Smirjdkova, O. Ondrasovicova, A. Kaskova and K. Laktiiovd, Folıa Veterinaria, 49, 31
(2005)
3) S.C. Mo, D.S. Choi and J.W. Robinson, J. Environ. Sci. Health, 23, 139 (1988)
4) US EPA Cost and benefit of reducing lead in gasoline. Draft Final Report, Office of Policy
Analysis, US EPA 230-03-84-005, Washington, DC, U.S.A. 1984
5) M.H.J. Blanusa and M. Fugas, Lead and cadmium in human diet in lead polluted area. In:
J. Barcelo (ed.): Environmental Contamination. Vol. 2. CEP Consultants Ltd., Edinburgh.,
490–492, 1990
6) A.A. Jensen, Levels and trends of environmental chemicals in human milk. Chemical
Contaminants in Human Milk. Boca Raton, FL: CRC Press; 45–198, 1991
7) L. Yang, M. Zhang, S. Lin, D. Chen and M. Zheng, Microchim. Acta, 142, 85 (2003)
8) I. Karadjova, S. Girousi, E. Iliadou and I. Stratis, Microchim. Acta, 134, 185 (2000)
9) S.L. Jeng, S.J. Lee and S.Y. Lin, J. Dairy Sci., 77, 945 (1994)
10) J.R. Barreira and S. Arribas, An. Quim., 86, 252 (1990)
11) M. Zorica and J. Liljana, Arh. Farm., 39, 181 (1989)
12) CE Regulation no. 2001/466 March 8, (2001) G.U.C.E. L77/1 March 16, 2001
13) I. Pavlovic, M. Sikiric, J.L. Havranek, N. Plavljanić and N. Brajenovic, Czech. J. Animal
Sci., 49, 164 (2004)
14) M. Sikiric, N. Brajenović, I. Pavlovic, J. L. Havranek, and N. Plavljanic, Czech. J. Animal
Sci., 48, 481 (2003)
15) A.R. Rod, S. Borhani and F. Shemira, Eur Food Res Technol., 223, 649 (2006)
16) I.I. Naqvi, Q. Saeed and M.A. Farrukh, Pakistan J. Biol. Sci., 7, 1355 (2004)
17) D.Y. Sarıca, D. Akım and T. Özden, Turk. J. Chem., 26, 263 (2002)
18) Z. Kılıç, O. Acar, M. Ulaşan and M. İlim, Food Chem., 76, 107 (2002)
19) S. Lynch, and D. Littlejohn, J. Anal. At. Spectrom., 4, 157 (1989)
20) O. Acar, Z. Kılıç and A.R. Türker, Food Chem., 71, 117 (2000)
21) D.Y. Sarıca, A.R. Türker and E. Erol, J. Sep. Sci., 29, 1600 (2006)
22) F. Gaucheron, Reprod. Nutr. Dev., 45, 473 (2005)
 Determination of lead in milk by ETAAS 993

23) E. Munoz, S. Palmero, Food Control, 15, 635 (2004)

24) R. Inam, G. Somer, Food Chem, 69, 345 (2000)
25) M. Tuzen, M. Soylak, Anal. Chim. Acta, 504, 325 (2004)
26) M. Yaman, Mikrochim. Acta, 129, 115 (1998)
27) S.L Jeng, S.J Lee, S.Y. Lin, J. Dairy Sci., 77, 945 (1994)