Czech J. Anim. Sci.

, 48, 2003 (11): 481486

Original Paper

Determination of metals in cows milk by flame atomic absorption spectrophotometry

M. SIKIRI1, N. BRAJENOVI1, I. PAVLOVI2, J. L. HAVRANEK2, N. PLAVLJANI2
1 2

Ruer Bokovi Institute, Zagreb, Croatia Faculty of Agriculture, University of Zagreb, Zagreb, Croatia

ABSTRACT: e essential (Ca, Cu, Fe, Mg, Mn, Zn) and toxic metals (Pb and Cd) were determined by flame atomic absorption spectrophotometry (FA-AAS) in raw cows milk. Milk was sampled four times in March 2000 on fifteen farms near Kri in the Zagreb region from bulk milk directly into plastic bottles (to avoid contamination). e raw cows milk samples were digested in an MLS-1200 Mega Microwave Digestion System with MDR Technology, and the metals were determined directly by FA-AAS in the solutions of digested samples. Statistical analyses were performed using statistical software SAS v. 8.0. Contents of essential minerals (Ca, Cu, Fe, Mg, Mn, Zn) in milk from Croatian farms are on the normal levels in comparison with the data from other countries. e level of toxic metals (Pb and Cd) was lower than the tolerance level defined by Croatian regulations (Pb < 100 g/l and Cd < 10 g/l). Keywords: lead; cadmium; copper; iron; manganese; magnesium; calcium; raw cows milk; flame atomic absorption spectrophotometry

Increased awareness of the influence of diets on human health has prompted producers to produce food of higher quality, rich in nutrients and nutraceuticals, in environmentally friendly manner. On the other hand, raw materials for food production are getting poor in essential minerals and vitamins at the same time. Milk and dairy products have been recognized all over the world for a long time as good for their sensory properties, but also for their beneficial influence on human health (Steijns, 2001). On the other hand, milk is an expensive raw material produced by animals (Boland et al., 2001). To produce milk or to make its products of good quality and to reduce their price, a good raw material is needed. erefore constant control of raw milk bought from farms is necessary. Milk should be controlled not only for fat and proteins as it is usual, but also for mineral content whenever possible. Milk is known as an excellent source of Ca, and it can supply moderate amounts of Mg, smaller amounts of Zn and very small amounts of Fe and Cu (Levy et al., 1985; Pennigton et al., 1987,

1995a,b). erefore the essential metals are interesting for determination of their adequate daily intake by the organism. On the other hand, due to the growing environmental pollution it is also necessary to determine and monitor the levels of toxic metals in milk, such as lead and cadmium, because they can significantly influence the human and animal health (Steijns, 2001). e most frequently used technique to determine metals in food and biological materials is atomic absorption spectrophotometry (AAS). If the concentrations of metals are high enough, AAS determinations are usually done by flame atomic absorption spectrophotometry (FA-AAS). When the concentrations of metals are low, electrothermal atomic absorption spectrophotometry (EA-AAS) is applied. Both techniques should give the same results if the concentration of metals is above the detection limit of FA-AAS (Jorhem, 1993). Because of low industrial production Croatia has a unique opportunity to produce milk rich in nutritional elements and with very low levels of toxic substances at the same time. In order to determine
481

Original Paper

Czech J. Anim. Sci., 48, 2003 (11): 481486

the quality of raw material bought by the Croatian dairy industry one-month research was conducted on 15 family farms in the region of Ivanic grad, the region of biggest milk producers in Croatia. Our attention was focused on the mineral composition of milk.

MATERIAL AND METHODS

Sampling strategy. Samples of raw cows milk were collected four times in March 2000 from fifteen farms located near Kri in the Zagreb region. e cows on each farm were fed dry hay and mineral premixes and drank water from the pump. Raw milk was sampled from the bulk milk of the farms directly into carefully washed plastic bottles because of the relatively great possibility of sample contamination, especially by lead and cadmium. Precautions against contamination. All chemicals were of the highest available purity. Deionised water was used. e laboratory ware was cleaned with chromosulphuric acid and rinsed with deionised water. e laboratory ware was dipped in 10% HNO3 for 24 h, rinsed with deionised water, dipped in 3% EDTA solution for 24 h and rinsed with deionised water again. Anakytical procedure. e fat and protein content (%) was determined on Bently 150 Infrared milk analyser. To determine the concentrations of metals chemicals of the highest available purity were used. HNO3 (Riedel) was used for the dilution and mineralization of raw milk samples, H 2O2 (9.791 mol/dm3, Riedel) was used for digestion of all milk samples. Lead nitrate (3.019 10 3 mol/dm3, Merck) and cadmium nitrate (4.230 103 mol/dm3, Merck) solutions were used as calibration standards. All solutions were prepared with deionised water. e metals (calcium, copper, iron, magnesium, manganese, zinc, lead and cadmium) were analysed by FA-AAS in a total of 60 milk samples. To eliminate the organic part of milk, 0.500 g of raw cows milk was treated with 6 ml of 65% HNO3 and 1 ml 30% H2O2, mixed and digested in an MLS-1200 Mega Microwave Digestion System with MDR Technology. Lead and cadmium were determined with ammonium salts pyrrolidine-1dithiocarboxylic acid (APDC) into methyl-isobutyl ketone by FA-AAS. e other investigated metals were determined directly by FA-AAS.
482

Instrumentation. PU 9100X FA-AAS equipped with Epson FX-850 printer was used. e instrument was controlled by Philips P 3120 PC. e air/ acetylene flame was used for the analysis of metals. Instrumental conditions for the determinations of metals in raw cows milk are given in Table 1. Statistics. Data were analysed by General Linear Models procedure. Mean differences were separated by Duncans Multiple Range Test. e level of significance was set at P < 0.05. All analyses were performed using statistical software SAS v. 8.0.

RESULTS
Average mineral concentration of raw cows milk is shown in Figure 1. e results of ANOVA are summarized in Table 2. Significant correlations between the investigated parameters are shown in Figure 2.

Fat and protein

In the period of investigations fat content in raw cows milk from 15 different farms was significantly different, ranging from 2.55 to 4.62%. Similar results were obtained for protein content that ranged from 2.81 to 4.1%. e correlation between fat and protein content was not statistically significant.

Calcium and magnesium

Differences between farms were significant for calcium content, ranging from 1 125.76 to 2 019.04 mg/kg, but not for magnesium content 136.02196.67 mg/kg. Calcium content correlated with fat content (R = 0.647).

Iron, zinc, copper and manganese

Iron, zinc, copper and manganese contents in milk from different farms were statistically different. Relatively high coefficients of correlation of manganese and copper with the other three minerals were obtained (Mn Fe: R = 0.641, Mn Zn: R = 0.731, Mn Cu: R = 0.618, Cu Fe: R = 0.613, Cu Zn: R = 0.629) while the contents of iron and zinc were not so highly correlated. In addition, correlations existed between copper and calcium content (R = 0.629) as well as between

Table 1. Instrumental conditions for lead, cadmium, calcium, copper, iron, manganese, magnesium and zinc determination in raw cows milk by FA-AAS Cadmium 228.8 0.5 1 1.01.3 0.032 0.91.2 0.09 0.91.2 0.029 0.91.2 0.003 0.91.2 0.013 422.7 0.5 1 279.5 0.5 1 285.2 0.5 1 213.9 0.5 1 Calcium Copper Iron Manganese Magnesium Zinc

Lead

Primary wavelength (nm) Band-pass (nm) Injection volume (ml) Flame type Fuel flow rate (l/min) Sensitivity (mg/l)

217.0 0.5 1

0.91.2 0.10

324.8 248.3 0.5 0.2 1 1 air/acetylene 0.81.1 0.81.0 0.041 0.060

Table 2. ANOVA summary of the farm effect on mineral levels in raw cows milk. Means with the same letter are not significantly different Ca (%) 0.0011 (mg/kg) 1 151.62 2 019.04b 1 463.57ab 1 314.40a 1 384.77ab 1 588.46ab 1 401.71ab 1 094.49a 1 478.34ab 1 309.12a 1 279.91a 1 364.10ab 1 389.10ab 1 694.72ab 1 125.76a
a

Czech J. Anim. Sci., 48, 2003 (11): 481486

Protein 0.5022 141.53 196.67 162.34 161.21 157.59 180.52 159.35 153.61 196.18 159.70 150.44 162.88 163.55 193.60 136.02 0.10 0.14b 0.12ab 0.13ab 0.14b 0.13ab 0.13ab 0.12ab 0.13ab 0.12ab 0.12ab 0.12ab 0.13ab 0.16b 0.13ab
a

Fat 0.0050 0.31 0.83b 0.49a 0.41a 0.54a 0.64b 0.50a 0.47a 0.48a 0.39a 0.38a 0.52a 0.63b 0.63b 0.49a
a

Mg <0.0001 0.20 0.69d 0.40bc 0.32a 0.41bc 0.39bc 0.43bc 0.37bc 0.41bc 0.34bc 0.31a 0.33bc 0.31a 0.48c 0.32a
a

Fe <0.0001

Zn

Cu

Mn 0.0069 0.026 0.055b 0.043ab 0.035ab 0.040ab 0.048ab 0.040ab 0.040ab 0.044ab 0.034ab 0.033ab 0.039ab 0.043ab 0.052b 0.039ab
a

Pb 0.0007 (g/kg) 23.06a 70.56b 41.36ab 29.61ab 40.36ab 43.50ab 37.53ab 47.22ab 58.59ab 32.51ab 39.98ab 36.32a 36.08a 56.43b 38.58ab

Cd <0.0001 3.02a 5.47bcd 6.09cd 6.47d 5.65bcd 5.65bcd 4.06ab 6.12cd 5.63bcd 5.80cd 5.46bcd 5.16bcd 4.49abc 4.91bcd 5.67bcd

<0.0001

<0.0001

Farm

(%)

I II III IV V VI VII VIII IX X XI XII XIII XIV XV

2.81 3.97e 3.22abc 3.32bc 3.23abc 3.86de 3.23abc 3.17abc 3.41cde 3.06abc 2.97ab 3.14abc 3.45bcd 4.10e 3.50bcd

3.15 5.90c 4.23b 3.81ab 4.08b 4.62bc 3.69ab 3.61ab 4.45b 4.01b 3.89abc 3.54ab 3.79ab 4.38ab 2.55a

ab

Original Paper

483

Original Paper
a)
Protein Fat

Czech J. Anim. Sci., 48, 2003 (11): 481486

Ca Mg

(%) 6
5 4 3 2 1 0

b)

(mg/kg) 2000
1500 1000 500

X I II III IV V VI VIIVIIIIX X XI XIIXIII IVXV Farm

d)

X I II III IV V VI VIIVIIIIX X XI XIIXIII IVXV Farm

c)
(mg/kg)
0.8 0.6 0.4 0.2 0.0

Fe Zn Cu Mn

(g/kg)

Pb Cd

60 40 20 0

X I II III IV V VI VIIVIIIIX X XI XIIXIII IVXV Farm

X I II III IV V VI VIIVIIIIX X XI XIIXIII IVXV Farm

Figure 1. Composition of raw cows milk from different farms: a) protein and fat; b) Ca and Mg; c) Fe, Zn, Cu and Mn; d) Pb and Cd

protein content and zinc (R = 0.669), copper (R = 0.656) and manganese (R = 0.630).

DISSCUSION
e concentrations of different essential minerals in milk from 15 farms included in this investigation can be arranged in the sequence Ca > Mg > Zn > Cu > Fe > Mn. is sequence is similar to that obtained by Rodrguez et al. (1999), the only difference being the sequence of Cu and Fe. It is known that the metal concentration in milk can vary due the factors influencing its secretion from the mammary gland, such as breed of the animal, season of the year, feeding and factors related

Lead and cadmium

Both lead and cadmium contents in milk from different farms were statistically different. No correlation was obtained between these minerals and other investigated parameters.

Table 3. Average mineral composition of raw cows milk from different countries Country Italy Spain Pakistan Croatia
484

Ca (mg/kg)

Mg (mg/kg)

Fe (mg/kg) 0.65

Cu (mg/kg) 0.21 0.16 0.02 0.45 0.38 0.12

Zn (mg/kg) 3.82 3.70 0.22 4.20 0.51 0.16

References Del Petere, Di Stanislao (1984) Zuerera-Cosano et al. (1994) Bano et al. (1985) this paper

1 251 48

116 3

0.46 0.10 0.60

1 403.94 322.33 165.01 37.97

0.13 0.02

Czech J. Anim. Sci., 48, 2003 (11): 481486

a)
b)

Original Paper

1.0 0.8

Zn Cu Mn

2500 2000

Ca

mg(mg/kg) kg-1

mg kg-1

(mg/kg)

0.6 0.4 0.2 0.06 0.04 0.02 0.00 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4

1500 1000 500 0 1 2 3 4 5 6 7

Protein (%)

Fat (%) d)
1.0 0.8
Fe Zn Cu

c)
2500 2000 1500 1000 1.0 0.8 0.6 0.4 0.2 0.0 0.1

Ca Fe Zn

mg (mg/kg) kg-1

(mg/kg)
0.2 0.3 0.4 0.5 0.6 0.7 0.8

mg kg-1

0.6 0.4 0.2 0.0 0.02 0.03 0.04 0.05 0.06 0.07

Cu (mg/kg)

Mn (mg/kg)

Figure 2. Correlations between: a) Zn, Cu, Mn and protein; b) Ca and fat; c) Ca, Fe, Zn and Cu; d) Fe, Zn, Cu and Mn in raw cows milk from different farms

to animal handling by humans (Moreno-Rojas et al., 1993; Zurera-Cosano et al., 1994). In this study the animals on the farms were of the same breed, samples were taken in a relatively short period, cows were fed the same concentrate but different dry hay. erefore the statistically significant differences in the content of essential minerals between the farms can be ascribed to different dry hays. Lead and cadmium are toxic minerals often associated with traffic pollution, therefore variations of their levels in milk from different farms are likely due to the location of meadows in relation to roads. Tolerance limits of these two toxic metals according to Croatian legislation are: Pb < 100 g/l, Cd < 10 g/l (Narodne Novine, 1994). Lead and cadmium levels in milk from all investigated farms were below these limits. In comparison with the average mineral composition of raw milk obtained by investigations in Italy (Del Petre and Di Stanislao, 1984), Spain (ZuereraCosano et al., 1994) and Pakistan (Bano et al., 1985), milk from Croatian farms has more Ca and Mg, similar content of Cu and lower content of Fe

and Zn (Table 3). However, it should be noted that deviations are much larger in our investigation. Relatively low correlation coefficients are in accordance with previous studies (Rodrguez et al., 1999). is was explained by the relatively narrow range of metal concentrations found in certain types of milk.

CONCLUSIONS
Contents of essential minerals (Ca, Mg, Fe, Zn. Cu, Mn) in milk from Croatian farms are on the normal levels in comparison with the data from other countries. e statistically significant difference in their contents in milk obtained from different farms implies the necessity of controlling mineral levels in milk by the dairy industry. Lead and cadmium levels in milk from all investigated farms are below the limits as defined by the Croatian legislation, indicating a possibility of producing organic products.
485

Original Paper

Czech J. Anim. Sci., 48, 2003 (11): 481486

Pennigton J.A.T., Wilson D.B., Young B., Johnson R.D., Vanderveen J.E. (1987): Mineral content of market samples of fluid whole milk. J. Am. Diet. Assoc., 87, 10361042. Pennigton J.A.T., Schoen S.A, Salmon G.D., Young B., Johnson R.D., Marts R.W.J.E. (1995a): Composition of core foods of the U.S. Food Supply, 19821991. II. Calcium, magnesium, iron and zinc. J. Food Comp. Analysis, 8, 129169. Pennigton J.A.T., Schoen S.A., Salmon G.D., Young B., Johnson R.D., Marts R.W.J.E. (1995b): Composition of core foods of the U.S. Food Supply, 19821991. III. Copper, manganese, selenium and iodine. J. Food Comp. Analysis, 8, 171217. Rodrguez E.M., Sanz Alaejos M., Diaz Romero C. (1999): Chemometric studies of several minerals in milks. J. Agric. Food. Chem., 47, 15201524. Steijns J.M. (2001): Milk ingredients as nutraceuticals. Int. J. Dairy Technol., 54, 81. Zurera-Cosano G., Moreno-Rojas R., Amaro-Lpez M. (1994): Effect of processing on contents and relationships of mineral elements of milk. Food Chem., 51, 7578.
Received: 030121 Accepted after corrections:031015

REFERENCES
Bano N., Naeem M., Khan H.H. (1985): Trace metal studies in milk, milk products and eggs. J. Nat. Sci. Math., 25, 6774. Boland M., MacGibbon A., Hill J. (2001): Designer milks for the new millennium. Livest. Prod. Sci., 72, 99109. Del Petere V., Di Stanislao F. (1984): Trace elements in cow milk sold in Marche Region. Nuovi Ann. Ig. Microbiol., 35, 313323. Jorhem L. (1993): Determination of metals foodstuffs by atomic absorption spectrophotometry after dry ashing: NMKL1 interlaboratory study of lead cadmium, zinc, copper, iron, chromium, and nickel. J. AOAC Inter., 76, 798813. Levy Y., Zeharoa A., Grunebaum M., Nitzan M., Steinherz R. (1985): Copper deficiency in infants fed cow milk. J. Pediatr., 106, 786788. Moreno-Rojas R., Zurera-Cosano G., Amaro-Lpez M. (1993): Micronutrients in natural cow, ewe and goat milk. Int. J. Food Sci. Nutr., 44, 3746. Narodne Novine (1994): Regulation of allowed pesticides, toxins, mycotoxins, metals and histamine and similar substances in food. 46, 1583.

ABSTRAKT

Stanoven obsahu kov v kravskm mlce pomoc plamenov atomov absorpn spektrofometrie
Obsahy zkladnch kov (Ca, Cu, Fe, Mg, Mn, Zn) a toxickch kov (Pb a Cd) v syrovm kravskm mlce byly stanoveny plamenovou atomovou absorpn spektrofotometri (FA-AAS). Vzorky mlka byly odebrny tyikrt v beznu v roce 2000 na 15 farmch pobl Krie v zhebsk oblasti ze smsnho mlka pmo do plastovch lahv (aby se pedelo kontaminaci). Vzorky syrovho kravskho mlka byly podrobeny mineralizaci v mikrovlnnm digesnm systmu MLS-1200 Mega pomoc technologie MDR; obsahy kov byly stanoveny pmo pomoc FA-AAS v roztocch vzork, kter proly mineralizac. Ke statistickm analzm byl pouit statistick software SAS v. 8.0. Obsahy zkladnch kov (Ca, Cu, Fe, Mg, Zn) v mlce z chorvatskch farem jsou ve srovnn s daji z ostatnch zem na normln hladin. Hladina toxickch kov (Pb a Cd) byla ni ne povolen limity, kter definuj chorvatsk pedpisy (Pb < 100 g/l a Cd < 10 g/l). Klov slova: olovo; kadmium; m; elezo; mangan; hok; vpnk; syrov kravsk mlko; plamenov atomov absorpn spektrofotometrie

Corresponding Author Dr. Nataa Brajenovi, Ruer Bokovi Institute, P.O.Box 180, 10002 Zagreb, Croatia Tel. +385 1 468 01 25, fax +385 1 468 02 45, e-mail: nbrajen@irb.hr
486