- Email: [email protected]
240Pu in Emmental type cheese produced in different regions of Western Europe
Journal of Environmental Radioactivity 72 (2004) 287–298 www.elsevier.com/locate/jenvrad
90
Sr, 238U, 234U, 137Cs, 40K and 239/240Pu in Emmental type cheese produced in different regions of Western Europe
P. Froidevaux a,, J.-J. Geering a, L. Pillonel b, J.-O. Bosset b, J.-F. Valley a a
Institute of Applied Radiophysics, University of Lausanne, Grand Pre´ 1, 1007 Lausanne, Switzerland b Federal Dairy Research Institute (FAM), Liebefeld, 3003 Bern, Switzerland Received 15 April 2003; received in revised form 26 June 2003; accepted 30 June 2003
Abstract A method is presented for the determination of 90Sr and uranium in Emmental type cheese collected in dairy plants from different European countries. Results display a significant correlation (r ¼ 0:708, Student t-test ¼ 6:02) between the 90Sr content of the cheese and the altitude of grazing. The highest 90Sr activity is 1.13 Bq kg1 of cheese and the lowest is 0.29 Bq kg1. Uranium activity is very low with a highest 238U value of 27 mBq kg1. In addition, 234U/238U ratio shows a large enrichment in 234U for every location. Without any significant indication of the geographic origin of the cheese, this enrichment is believed to be due to the geological features of the pasture, soil and underground water. These results tend to prove that the contamination of milk by uranium originates principally from the water that the cows drink instead of the forage. This finding may have a great importance in models dealing with dairy food contamination by radionuclides following a nuclear accident. Also, the 90Sr content and to a lesser extent the 234U/238U ratio could be used to trace the authenticity of the origin of the cheese. 137Cs activity is lower than the detection limit of 0.1 Bq kg1 in all the samples collected (n ¼ 20). Based on natural 40K activity in cheese (15–21 Bq kg1), the decontamination factor for the alkaline cations from milk to cheese is about 20. Plutonium activity stays below the detection limit of 0.3 mBq kg1. # 2003 Elsevier Ltd. All rights reserved. Keywords: Strontium-90; Cheese; Uranium isotopes; Milk-to-cheese transfer; Food authenticity
Corresponding author. Tel.: +41-21-623-34-81; fax: +41-21-623-34-35 E-mail address: [email protected] (P. Froidevaux).
0265-931X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0265-931X(03)00179-6
288
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
1. Introduction As a result of the atmospheric fallout from nuclear bomb tests and of the Chernobyl nuclear power plant accident, 90Sr and 137Cs have been widely deposited all over the world. However, the fallout of the 1960s shows large variations depending on the geomorphology of the region, while the Chernobyl deposition depends essentially on the meteorological conditions over Europe a few days after the accident. Switzerland is a good example of such a situation due to the presence of natural barriers such as the Jura mountains chain in the north-west (maximum altitude: 1650 m) and the Alps in the south. The geomorphology of the country has led to very different fallouts during the 1960s. For instance, plutonium deposition can reach 300 Bq m2 in the Jura mountains, while the average in the Swiss lowland (average altitude: 300 m) is about 75 Bq m2 (Froidevaux et al., 2001; Bundt et al., 2000). For 90Sr, the average values are Jura: 3500 Bq m2; Swiss lowland: 700 Bq m2 (reference date: 1 June 2002). Still higher values are observed in alpine pastures (Geering et al., 1995). On the other hand, 137Cs deposition due to the Chernobyl accident is essentially observed in the south of the Alps (Tessin state) (Bosset et al., 2001) and to a lesser extent in the north east of Switzerland (mainly Thurgovie state) where heavy rainfalls were recorded a few days after the accident. Immediately after the Chernobyl accident, special attention was given to milk and milk products in Switzerland. Typically, the 90Sr value for milk increased by a factor of two, passing from 0.1 to 0.2 Bq l1. Furthermore, the results showed that almost all the 90Sr passes from milk to cheese by the precipitation of the casein (Geering et al., 1990). As most of the Swiss cheese is produced in small dairy units with milk collected from cows grazing in Alpine and sub-alpine pastures, it was soon recognised that the main contributor to the 90Sr dose would be dairy products. In fact, Gatsberger et al. (2001) demonstrated recently that the transfer rate of 90Sr from vegetation to milk is higher in highland pastures than in lowland pasture. This fact is partly explained by a lower dietary calcium intake mainly due to lower calcium concentration of the vegetation in Alpine pastures. Of all the actinides present naturally or artificially in soil, only uranium is susceptible to be transferred to some extent to the vegetation due to its partial solubility, especially as uranyl cation. In an investigation on the uranium phytoremediation of contaminated sites, Ebbs et al. (2001) demonstrated that uranium was more readily extracted when present predominantly as the free uranyl cation. Moreover, while it was clear that some plants preferentially accumulate uranium, none accumulated this latter to more than 3.5 lg/plant except in the presence of soil supplemented with citric acid, an efficient chelator. One of the major contributions of radiological exposure to man from uranium is mining and milling tailings, where enhanced concentrations of uranium and thorium series are observed (Rayno, 1983). Thus, enhanced accumulation of uranium in forage or in drinking water could lead to enhanced uranium content in the milk and beef food chain (Lapham et al., 1989). This paper aims to quantify 90Sr in Emmental type cheese produced in different Western European countries to assess the role of geomorphology in the deposition
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
289
and contamination of the dairy food chain by radionuclides in case of accidental fallout. Another goal is to estimate the role of the geology in the presence of uranium in milk products and to get information about the uranium pathway from feeds to the milk by measuring the 234U/238U ratio in cheese samples. Furthermore, analysis of the 90Sr and uranium isotope content of cheese has been conducted to see if these radioactive components could be used to determine the authenticity and traceability of the origin of dairy products.
2. Experimental 2.1. Sampling Twenty Emmental cheese samples were collected from six European regions (Fig. 1). Three samples were taken from each region (except for Switzerland with six and Finland with two samples). Three regions are adjacent to Switzerland (Savoie (F), Allga¨u (D), Vorarlberg (A)), the two remaining regions (Bretagne (F) and the central region of Finland) are quite distant. Considered together, these six regions account for approximately 77% of European Emmental production. The samples were supplied directly by the cheese factories (Austria, Finland, Switzerland) or were collected with the help of a national dairy research centre (France and Germany). The altitude of production is the altitude of the dairy farms. For alpine regions, the true altitude of pasture might be a little bit different due to the v steepness. All the samples were stored at 20 C until analysis.
Fig. 1. Geographic origins of the collected samples. Al ¼ Allg€au; BR ¼ Bretagne; CH ¼ Switzerland; FI ¼ Finland; SA ¼ Savoie; VO ¼ Vorarlberg.
290
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
2.2. Reagents and equipment All reagents used were of analytical grade (Merck, Darmstadt, Germany or Fluka, Buchs, Switzerland). The cheese samples were oven dried and ashed at 600 v C in a Heraus furnace. Digestion of ash from cheese samples was carried out in an MLS lltraCLAVE microwave apparatus or in an MLS Ethos Plus microwave apparatus (Milestone Inc., Monroe, CT). Chromatographic separations were carried out on a Bio-Rad AG 50W 8 (100–200 mesh) cationic exchanger (90Sr), or on an Eichrom UTEVA resin (U). An anion exchanger (Bio-Rad AG1 2, 100– 200 mesh) was used to isolate plutonium and remove thorium isotopes. The solutions were loaded onto the chromatographic column by a peristaltic pump (Ismatec IPS-8, Zu¨rich, Switzerland). Yttrium and strontium yields as well as calcium content were determined on a Perkin Elmer AS 4100 atomic absorption spectrometer (Y: 410.2 nm; Sr: 460.7 nm, Ca: 422.7 nm). Yttrium oxalate sources were measured on a Tennelec LB 4100 low-level proportional gas counter (Canberra Electronique, Paris, France). Alpha spectra were recorded using a Canberra Alpha Analyst spectrometer and PIPS detector of 450 mm2 in diameter (Canberra Electronique, Paris, France). c-spectra were recorded using a Canberra HPGe well detector GC4523 in a 4 ml geometry (ash) or in a 250 ml geometry (grated cheese) and the analytical software ‘‘Genie 2000’’. 2.3.
90
Sr measurements
Details of the separation technique used for determining the 90Sr in cheese samples have been described in an earlier paper (Geering et al., 1990). Briefly, to determine 90Sr, 90Y activity was measured. The higher b- energy of 90Y (2.3 MeV) led to a better counting efficiency and the shorter half-life (64.5 h) of 90Y made a radionuclidic purity control possible. The separation of strontium was achieved successively on two ion exchange columns (Dowex 50w, 50 ml, then 8 ml) using first a (trans-1,2-cyclohexylene-dinitrilo) tetraacetic acid (CyDTA) solution and then a sodium citrate solution to elute strontium. After adding an yttrium carrier, the latter solution was allowed to stand for about 10 days to let 90Y grow partially. Yttrium was then separated from strontium on another ion exchange column (Dowex 50w, 8 ml) by elution with a sodium malonate solution. Finally, yttrium oxalate was precipitated and a thin source was prepared by filtration on a 0.24 lm v Millipore cellulose ester filter (n GSWOP2400). To ensure a low background, the source activity was measured in anticoincidence in a Tennelec LB 4100 w proportional counter. Under these conditions, the background was found to be 0.004 dps. The source was measured typically for 100 h and intermediate results were automatically taken every 4 h to check the purity of 90Y. A detection limit of 0.05 Bq/kg can be achieved. 2.4. Measurements of actinides Five gram of cheese ash, 20 ml of 8 M HNO3 and 232U tracer (52:1 0:6 mBq) were introduced into a flask suitable for microwave digestion. In the case of plu-
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
291
tonium determination, 242Pu (44 0:4 mBq) was added. The digestion of ash was v v carried out using the following programme: 5 min to 100 C, 5 min to 170 C, 5 v v v min at 170 C, 5 min to 200 C and 5 min at 200 C. After cooling, the solution was filtered and pumped onto an anion exchange column (Bio-Rad AG1 2, 6 ml, 100–200 mesh, NO 3 form) and the effluent was transferred onto an Eichrom1 U/TEVA extraction cartridge (U/TEVA, Eichrom Europe, Paris, France, 2 ml, 50–100 mesh, NO 3 form). The stacked columns were washed with 10 ml 8 M HNO3, then with 40 ml 3 M HNO3. The two columns were separated and the U/TEVA column washed with 10 ml of 9 M HCl to remove thorium. Uranium was eluted from the U/TEVA column by 20 ml of 0.01 M HCl. In the case of plutonium determination, the anion exchange column was washed with 20 ml of concentrated HCl to remove thorium, then the plutonium was eluted by reduction of Pu4+ to Pu3+ with 50 ml of an hydroxylamine solution. The uranium fraction was evaporated and uranium electrodeposited on a stainless 2 steel disk in a HSO 4 =SO4 buffer at pH 1.9, over 75 min at 1.2 A. Plutonium was electrodeposited in the same way.
3. Results and discussion 3.1. Measurements of
90
Sr
The results of 90Sr determination are displayed in Table 1. Calcium content is used as an internal standard and the activities are reported as Bq g1 Ca. Table 1 shows that the calcium content of fresh cheese is almost constant and the average value is 9:64 0:5 g kg1 of cheese. 90Sr activities range from 0.034 to 0.121 Bq g1 Ca. The limited number of samples selected for this study makes it difficult to interpret the 90Sr activity in terms of the geographic origin of a given cheese sample but individual results clearly show a significant correlation (r ¼ 0:708, Student t-test ¼ 6:02) between the altitude of production of the cheese and the 90Sr content (Fig. 2), which makes it possible to distinguish Emmental type cheese produced in Bretagne and in Finland from those produced in Alpine regions (Switzerland, Savoie, Allga¨u and Vorarlberg). A study conducted by Schuller et al. (2002) on the fallout from global weapons of 137Cs in soils of south-central Chile demonstrated that the 137Cs deposition found in different locations correlates with altitude (r ¼ 0:714) and annual rainfall rate (r ¼ 0:791). Therefore, rainfall and snowfall can be considered as the driving forces for soil contamination by fallout. The 90Sr deposition following the Chernobyl accident has been very low in Western Europe. In particular, Friedli et al. (1991) observed that virtually all the Chernobyl 90Sr seems to have vanished from soil, grass and milk a few months after the deposition. Therefore, all the 90Sr measured in this study can be considered as radiostrontium from weapons. The results show that the geomorphology plays a critical role in the distribution of 90Sr from nuclear bomb test fallouts in soils and, consequently, in milk and cheese. In Bretagne, virtually no relief is present to generate strong precipitations leading to a high contamination of soil by 90Sr. A similar
96.72 96.60 96.47 95.88 96.97 96.93 96.36 92.28 96.70 96.79 96.62 96.21 96.73 96.61 96.87 96.87 96.33 96.85 96.72 96.53 96.54 96.50
AL1 AL2 AL3 BE1 BE2 BE3 BR1 BR2 BR3 SA1 SA2 SA3 SG1 SG2 SG3 VO1 VO2 VO3 FI1 FI2 FI1 FI2
760 800 900 480 470 640 100 100 100 540 350 540 750 610 450 950 710 690 200 300 200 300
Altitude of production 9:37 0:3 9:67 0:3 9:45 0:3 9:32 0:3 9:36 0:3 9:49 0:3 10:26 0:3 10:60 0:3 10:63 0:3 9:69 0:3 9:96 0:3 10:4 0:3 9:63 0:3 9:76 0:3 9:27 0:3 9:48 0:3 9:35 0:3 9:21 0:3 8:83 0:3 9:03 0:3 8:83 0:3 9:03 0:3
g Ca (kg fresh weight)1
Sr (Bq g1 Ca)
0:121 0:008 0:066 0:005 0:088 0:006 0:078 0:007 0:066 0:006 0:034 0:003 0:053 0:004 0:028 0:002 0:039 0:003 0:058 0:004 0:036 0:004 0:068 0:005 0:072 0:006 0:071 0:005 0:042 0:007 0:105 0:006 0:120 0:008 0:102 0:007 0:058 0:005 0:036 0:003
90
1:07 0:17 1:03 0:12 1:51 0:11 3:54 0:17 2:56 0:11 1:47 0:10 2:05 0:15 2:17 0:12 0:79 0:07 1:34 0:2 1:10 0:10 1:46 0:15 1:90 0:10 2:35 0:30 2:91 0:30 1:37 0:10 2:03 0:20 2:82 0:11 0:32 0:09 0:40 0:07 0:56 0:10 0:18 0:04
U (Bq g1 Ca)
234
0:79 0:17 0:82 0:12 1:09 0:15 2:90 0:20 2:14 0:11 1:05 0:10 1:27 0:10 1:51 0:09 0:52 0:09 1:24 0:20 0:83 0:14 1:23 0:13 1:35 0:10 1:33 0:20 2:59 0:20 1:05 0:10 1:82 0:21 2:34 0:10 0:27 0:06 0:28 0:05 0:38 0:10 0:17 0:04
U (Bq g1 Ca)
238
Al: Allga¨u (D), BE: Bern (CH), BR: Bretagne (F), SA: Savoie (F), SG: St-Gall (CH), VO: Vorarlberg (D), FI: Finland.
Percentage lost on ignition
Sampling area 1:35 0:25 1:26 0:15 1:38 0:14 1:22 0:12 1:20 0:12 1:40 0:15 1:61 0:18 1:44 0:10 1:53 0:15 1:08 0:20 1:31 0:15 1:19 0:14 1:41 0:15 1:77 0:20 1:12 0:15 1:30 0:15 1:12 0:15 1:18 0:10 1:18 0:4 1:42 0:4 1:47 0:3 1:1 0:2
U/238U
234
1:49 0:2 1:68 0:2 1:59 0:2 1:68 0:2 1:68 0:2 1:96 0:3 1:49 0:2 1:59 0:2 1:87 0:3 1:96 0:3 1:87 0:3 1:87 0:3 1:77 0:3 1:59 0:2 1:68 0:2 2:05 0:3 2:15 0:3 1:31 0:2 n.m. n.m. n.m. n.m.
K (Bq g1 Ca)
40
Table 1 Results of the radiochemical analyses carried out on different samples of Emmental type cheese of different origins (uncertainties of the measurements given in 2r)
292 P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
293
situation is encountered in Finland. The Emmental type cheese produced in these two countries displayed the lowest 90Sr activities. In Alpine regions, 90Sr cheese activity is clearly a matter of the altitude of production (Fig. 2). Gatsberger et al. (2001) demonstrated that not only 137Cs but also the 90Sr uptake from soil to plant is enhanced in Alpine pastures compared to lowland pastures. The results of 90Sr determination in grass in Switzerland during the years 2000–2002 (Froidevaux et al., 2003) display the same trends, and values of 2.8–6.4 Bq kg1 dry weight (dw) are obtained in lowland Switzerland (mainly Swiss lowland), 7.5–11.8 Bq kg1 in the Jura mountains and 8.7–15 Bq kg1 in Alpine and sub-Alpine pastures (Table 2). Furthermore, the results of 90Sr in grass normalised to the Ca content show that the mineral matrix (essentially Ca and K) of grass from Jura and the Alps is enriched in 90Sr compared to lowland pastures. In addition, 90Sr (Bq g1 Ca) content of milk is enhanced in Alpine regions compared to the Swiss lowland (average 0.05 Bq g1 Ca in the Swiss lowland vs. 0.20 Bq g1 Ca in the Davos region). This fact is supported by the larger 90Sr content of the mineral matrix of cheese produced in Alpine locations. 3.2. Measurements of
137
Cs and
40
K
137
Cs and 40K were determined by c-spectrometry on cheese ash (4 ml) or grated cheese (250 ml). 137Cs activity in cheese stays under the detection limit of 0.1 Bq kg1 for all the samples. Natural 40K activity is low compared to the activity measured in milk (average 40K in Swiss milk was 44 5 Bq g1 Ca, n ¼ 15, data from Geering et al., 2001, personal communication) and ranges in cheese between
Fig. 2. 90Sr activity (Bq g1 Ca) as a function of the altitude of cheese sampling location. Error bars are given in 2r.
294
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
Table 2 Results of the radiochemical analyses carried out on different samples of grass for 90Sr determination in Switzerland in 2002 (data from Froidevaux et al., 2003). Uncertainties of measurements are reported as 2r Sr Bq (kg dry weight)1
Sr Bq g1 Ca
Location of sampling
Swiss region
90
90
Givrine St-Cergue Intragna Sessa Rodi-Fiesso Coire Davos Gene`ve Go¨sgen Mu¨hleberg Villigen Beznau Grangeneuve
Swiss Jura Swiss Jura Tessin (south Alps) Tessin (south Alps) Tessin (south Alps) Est sub-Alps Est Alps Swiss lowland (west) Swiss lowland (centre) Swiss lowland (centre) Swiss lowland (north) Swiss lowland (north) Swiss lowland (west)
11:8 0:5 3:3 0:1 16:8 0:4 11:5 0:3 6:7 0:4 6:7 0:3 8:7 0:4 2:8 0:1 6:4 0:3 2:7 0:1 2:6 0:4 4:8 0:2 4:9 0:5
1:1 0:03 0:20 0:009 1:12 0:02 1:12 0:03 0:38 0:02 0:21 0:02 0:52 0:02 0:77 0:04 0:24 0:01 0:19 0:01 0:46 0:07 0:20 0:01 0:40 0:04
1.31 and 2.15 Bq g1 Ca. Based on the 40K in milk and in cheese samples, the decontamination factor for alkaline cations is about 20, which means that most of the alkaline cations are in a soluble form in which they are transferred to the whey during the cheese making process. Cheese making has been widely used as a milk decontamination process for 137Cs in areas contaminated by the Chernobyl accident (Patel and Prasad, 1993). 3.3. Measurements of uranium Uranium activity is low in all the cheese samples. The lowest values are found in Finland cheese (0.17–0.38 mBq 238U g1 Ca) and the highest ones in Switzerland cheese (1.07–2.90 mBq 238U g1 Ca). Measurements of uranium by a-spectrometry makes it possible to determine both 234 and 238 uranium isotopes. Surprisingly, depletion in 238U is observed in all the cheese samples, the highest depletion level being found in Bretagne cheese (234 U=238 U ¼ 1:53) and the lowest in cheese produced in the Savoie area (234 U=238 U ¼ 1:19). Depletion of 238U in water is a well known phenomenon. This reflects the process of a-recoil during the decay of 238U which leads to high 234U/238U ratios in river waters, and a flux of 234U into the oceans from marine sediment pore-water (Henderson, 2002). On the other hand, the soils present an equilibrium between the parent (238U) and the daughter (234U) or a slight depletion in 234U, due to the preferential weathering of 234U. In closed systems older than 1 106 y, such as the earth’s mantle, uranium decay chains are at equilibrium. Uranium analysis of grass samples collected in different regions of Switzerland shows a large range of activities (0.25–14.3 Bq kg1 dw) but no depletion in 238U. Instead, an equilibrium of uranium isotopes is observed in three samples and a slight depletion in 234U is observed in one case (Table 3 and Fig. 3), reflecting the situation taking place in the soil, e.g. no
Grass Grass Grass Grass
Rodi-Fiesso (CH)a Leibstadt (CH)a Davos (CH)a Mu¨hleberg (CH)a
Froidevaux et al. 2003.
Tap water Tap water Drinking water Drinking water Milk Milk Milk
Ullava (Fi) Lapinlahi (Fi) Swiss lowland Swiss Alps Swiss lowland 1 Swiss lowland 2 Swiss lowland 3
a
Type of sample
Site of sampling
0:25 0:01 1:29 0:04 14:3 0:3 2:71 0:08
234
U (Bq kg1 dw)
0:63 0:06 0:45 0:06 3:3 0:3
1:30 0:16 1:7 0:3
234
U (mBq l1)
0:25 0:01 1:33 0:04 14:3 0:3 3:08 0:08
U (Bq kg1 dw)
238
1:3 0:16 1:4 0:2 4–33 4–196 0:58 0:06 0:38 0:06 2:9 0:3
U (mBq l1)
238
0:85 0:02
0.0140.0005
U (Bq g1 Ca) 238
1:10 0:10 1:20 0:10 1:12 0:10
1:0 0:2 1:20 0:2
U/238U
234
1:00 0:06 0:97 0:04 1:00 0:03 0:88 0:05
U/238U
234
This work This work Bosshard (1992) Bosshard (1992) This work This work This work
Reference
Table 3 Results of the radiochemical analyses carried out on different samples of water, milk and grass. Uncertainties of measurements are reported as 2r
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298 295
296
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
Fig. 3. 234U/238U ratio in cheese and grass samples. Error bars are given in 2r.
fractionation of uranium isotopes during the transfer of uranium from the soil to the plant. One hypothesis which could explain the depletion in 238U in cheese is that the contamination of milk by uranium arises through drinking water instead of forage. It would mean that the metabolisms of drink and food are different. Taking into account an ingestion of 40 kg d1 of dry forage containing the lowest activity, e.g. 0.25 Bq 238U by the cow, the daily intake of 238U would be 10 Bq, with a 234 U/238U ratio of 1:00 0:06. With a milk yield of 30 l d1, the uranium excretion (based on cheese and milk content) through milk would however be at most 0:060 0:009 Bq d1 with a 234U/238U ratio close to 1:35 0:10 (Swiss cheese and milk average). This is only a very slight fraction of the intake that cannot be attributed to forage. On the other hand, if we take into account an ingestion of 60 l of water at 10 mBq l1 of 238U (Swiss average in drinking water, data from Bosshard et al., 1992), the daily intake would be only 0.6 Bq through drinking, with a 234U/238U ratio being close to 1.20. The uranium content of Finnish cheese was found to be significantly lower than the uranium content of the cheese produced in the other locations. As Finnish soils contain large deposits of uranium, this was an unexpected result. Therefore, samples of the drinking water of cows were collected in the two farms from which the two cheese samples originated and analysed for uranium. The results show that the two water samples have a very low uranium content (Table 2). These results seem to indicate that the uranium contamination of milk and cheese is due to the drinking water rather than to the forage. 3.4. Measurements of plutonium 238
Pu and 239/240Pu activities were determined on one sample from each location, except Finland (AL1, BE1, SA1, VO1, SG1, BR1), and were found to be less than
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
297
the detection limit of 0.3 mBq kg1 of cheese. The results confirm that weapon grade plutonium is strongly bound to the soil particles and is not transferred to a significant extent to the plant. 4. Conclusions The study of Emmental type cheese produced in six different Western European countries suggests that the 90Sr content of dairy products could be used to distinguish between the production zones. Clearly, Bretagne and Finland cheeses can be distinguished from Alpine cheeses on the base of 90Sr content. Geomorphology played a strong role in the distribution of weapons grade 90Sr, which is reflected in the significant correlation measured between 90Sr activity and the altitude of grazing. Furthermore, the results show that uranium isotope measurements in cheese can be used to get information on the mechanism of uranium transfer from feed to milk. In particular, this investigation throws some light on the role of water on the milk contamination by uranium. In the next step towards the modelling of the contamination of the dairy food chain by radionuclides, we intend to improve our knowledge of 90Sr and uranium pathways from the cow’s feed and drink by measuring 90Sr and uranium isotope activity in cow’s excreta such as urine and faeces. Acknowledgements F. Barraud and T. Schmittler are acknowledged for their technical help. This v work was supported by the Federal Office of Public Health under contract n 3189.001.4. References Bosshard, E., Zimmerli, B., Schlatter, C., 1992. Uranium in the diet: risk assessment of its nephro- and radiotoxicity. Chemosphere 24, 309–321. Bosset, J.-O., Albrecht, B., Badertscher, R., Dalla Torre, M., Gauch, R., Imhof, R., Isolini, D., Lavanchy, P., Meyer, J., Spahr, U., Wismer, M., Burger, M., Bontognali, R., Rezzonico, E., Quadroni, G., 2001. Caracte´risation microbiologique, chimique et sensorielle de laits, de caille´s et de fromages de che`vre tessinois de types Formaggini (Bu¨scion, Robiola) et Formaggella. Travaux de chimie alimentaire et d’hygie`ne 92, 546–580. Bundt, M., Albrecht, A., Froidevaux, P., Blaser, P., Flu¨hler, H., 2000. Impact of preferential flow on radionuclide distribution in soil. Journal of Environmental Science and Technology 34, 3895–3899. Ebbs, S., Brady, D., Norvell, W., Kochian, L., 2001. Uranium speciation, plant uptake, and phytoremediation. Practice Periodical of Hazardous, Toxic and Radioactive Waste Management 5, 130– 135. Friedli, C., Geering, J.-J., Lerch, P., 1991. Some aspects of behaviour of 90Sr in the environment. Radiochimica Acta 52, 237–240. Froidevaux, P., Geering, J.-J, Schmittler, T., Barraud, F., Valley, J.-F, 2001. Mesures de plutonium et d’ame´ricium dans l’environnement. Environmental radioactivity and radiation exposure in Switzer-
298
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
land. Feral Office of Public Health, Division of Radiation Protection, Chemin du Muse´e 3, 1700 Fribourg, Switzerland ((Chap. 7.1) (ISBN 3-905235-40-4)). Froidevaux, P., Geering, J.-J., Friedrich-Be´net, K. , Schmittler, T., Barraud, F., Valley, J.-F., Volkk, H., Jungok, M., Burger, M., Schmidt, E., Byrde, F., Jakob, A., Astner, M., 2003. Radioactivite´ du sol et de l’herbe. Environmental radioactivity and radiation exposure in Switzerland, chap. 4.3, Federal Office of Public Health, Division of Radiation Protection, Chemin du Muse´e 3, 1700 Fribourg, Switzerland. Gastberger, M., Steinhau¨sler, F., Gerzabeck, M.H., Hubmer, A., 2001. Fallout strontium and caesium transfer from vegetation to cow milk at two lowland and two Alpine pastures. Journal of Environmental Radioactivity 54, 267–273. Geering, J.-J., Friedli, C., Lerch, P., 1990. Method of determination of 90Sr in the environment. Journal of Trace Microprobe Techniques 8, 211–230. Geering, J.-J., Carlone, F., Valley, J.-F, 1995. Mesures de 90Sr dans le sol, l’herbe, le lait, le froment, les verte`bres et les dents. Environmental radioactivity and radiation exposure in Switzerland. Feral Office of Public Health, Division of Radiation Protection, Chemin du Muse´e 3, 1700 Fribourg, Switzerland ((Chap. B.3.11 et B3.12) (ISBN 3-905235-21-8)). Henderson, G., 2002. Seawater (234U/238U) during the last 800 thousand years. Earth Planetary Science Letters 199, 97–110. Lapham, S.C., Millard, J.B., Samet, J.M., 1989. Health implications of radionuclide levels in cattle raised near U mining and milling facilities in Ambrosia Lake, New Mexico. Health Physics 56, 327–340. Patel, A.A., Prasad, S.R., 1993. Decontamination of radioactive milk: a review. International Journal of Radiation Biology 63, 405–412. Rayno, D.R., 1983. Estimated dose to man from uranium milling via the beef/milk food chain pathway. Science of the Total Environment 31, 219–241. Schuller, P., Voigt, G., Handl, J., Ellies, A., Olivia, L., 2002. Global weapon’s fallout 137Cs in soils and transfer to vegetation in south-central Chile. Journal of Environmental Radioactivity 62, 181–193.
90
Sr, 238U, 234U, 137Cs, 40K and 239/240Pu in Emmental type cheese produced in different regions of Western Europe
P. Froidevaux a,, J.-J. Geering a, L. Pillonel b, J.-O. Bosset b, J.-F. Valley a a
Institute of Applied Radiophysics, University of Lausanne, Grand Pre´ 1, 1007 Lausanne, Switzerland b Federal Dairy Research Institute (FAM), Liebefeld, 3003 Bern, Switzerland Received 15 April 2003; received in revised form 26 June 2003; accepted 30 June 2003
Abstract A method is presented for the determination of 90Sr and uranium in Emmental type cheese collected in dairy plants from different European countries. Results display a significant correlation (r ¼ 0:708, Student t-test ¼ 6:02) between the 90Sr content of the cheese and the altitude of grazing. The highest 90Sr activity is 1.13 Bq kg1 of cheese and the lowest is 0.29 Bq kg1. Uranium activity is very low with a highest 238U value of 27 mBq kg1. In addition, 234U/238U ratio shows a large enrichment in 234U for every location. Without any significant indication of the geographic origin of the cheese, this enrichment is believed to be due to the geological features of the pasture, soil and underground water. These results tend to prove that the contamination of milk by uranium originates principally from the water that the cows drink instead of the forage. This finding may have a great importance in models dealing with dairy food contamination by radionuclides following a nuclear accident. Also, the 90Sr content and to a lesser extent the 234U/238U ratio could be used to trace the authenticity of the origin of the cheese. 137Cs activity is lower than the detection limit of 0.1 Bq kg1 in all the samples collected (n ¼ 20). Based on natural 40K activity in cheese (15–21 Bq kg1), the decontamination factor for the alkaline cations from milk to cheese is about 20. Plutonium activity stays below the detection limit of 0.3 mBq kg1. # 2003 Elsevier Ltd. All rights reserved. Keywords: Strontium-90; Cheese; Uranium isotopes; Milk-to-cheese transfer; Food authenticity
Corresponding author. Tel.: +41-21-623-34-81; fax: +41-21-623-34-35 E-mail address: [email protected] (P. Froidevaux).
0265-931X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0265-931X(03)00179-6
288
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
1. Introduction As a result of the atmospheric fallout from nuclear bomb tests and of the Chernobyl nuclear power plant accident, 90Sr and 137Cs have been widely deposited all over the world. However, the fallout of the 1960s shows large variations depending on the geomorphology of the region, while the Chernobyl deposition depends essentially on the meteorological conditions over Europe a few days after the accident. Switzerland is a good example of such a situation due to the presence of natural barriers such as the Jura mountains chain in the north-west (maximum altitude: 1650 m) and the Alps in the south. The geomorphology of the country has led to very different fallouts during the 1960s. For instance, plutonium deposition can reach 300 Bq m2 in the Jura mountains, while the average in the Swiss lowland (average altitude: 300 m) is about 75 Bq m2 (Froidevaux et al., 2001; Bundt et al., 2000). For 90Sr, the average values are Jura: 3500 Bq m2; Swiss lowland: 700 Bq m2 (reference date: 1 June 2002). Still higher values are observed in alpine pastures (Geering et al., 1995). On the other hand, 137Cs deposition due to the Chernobyl accident is essentially observed in the south of the Alps (Tessin state) (Bosset et al., 2001) and to a lesser extent in the north east of Switzerland (mainly Thurgovie state) where heavy rainfalls were recorded a few days after the accident. Immediately after the Chernobyl accident, special attention was given to milk and milk products in Switzerland. Typically, the 90Sr value for milk increased by a factor of two, passing from 0.1 to 0.2 Bq l1. Furthermore, the results showed that almost all the 90Sr passes from milk to cheese by the precipitation of the casein (Geering et al., 1990). As most of the Swiss cheese is produced in small dairy units with milk collected from cows grazing in Alpine and sub-alpine pastures, it was soon recognised that the main contributor to the 90Sr dose would be dairy products. In fact, Gatsberger et al. (2001) demonstrated recently that the transfer rate of 90Sr from vegetation to milk is higher in highland pastures than in lowland pasture. This fact is partly explained by a lower dietary calcium intake mainly due to lower calcium concentration of the vegetation in Alpine pastures. Of all the actinides present naturally or artificially in soil, only uranium is susceptible to be transferred to some extent to the vegetation due to its partial solubility, especially as uranyl cation. In an investigation on the uranium phytoremediation of contaminated sites, Ebbs et al. (2001) demonstrated that uranium was more readily extracted when present predominantly as the free uranyl cation. Moreover, while it was clear that some plants preferentially accumulate uranium, none accumulated this latter to more than 3.5 lg/plant except in the presence of soil supplemented with citric acid, an efficient chelator. One of the major contributions of radiological exposure to man from uranium is mining and milling tailings, where enhanced concentrations of uranium and thorium series are observed (Rayno, 1983). Thus, enhanced accumulation of uranium in forage or in drinking water could lead to enhanced uranium content in the milk and beef food chain (Lapham et al., 1989). This paper aims to quantify 90Sr in Emmental type cheese produced in different Western European countries to assess the role of geomorphology in the deposition
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
289
and contamination of the dairy food chain by radionuclides in case of accidental fallout. Another goal is to estimate the role of the geology in the presence of uranium in milk products and to get information about the uranium pathway from feeds to the milk by measuring the 234U/238U ratio in cheese samples. Furthermore, analysis of the 90Sr and uranium isotope content of cheese has been conducted to see if these radioactive components could be used to determine the authenticity and traceability of the origin of dairy products.
2. Experimental 2.1. Sampling Twenty Emmental cheese samples were collected from six European regions (Fig. 1). Three samples were taken from each region (except for Switzerland with six and Finland with two samples). Three regions are adjacent to Switzerland (Savoie (F), Allga¨u (D), Vorarlberg (A)), the two remaining regions (Bretagne (F) and the central region of Finland) are quite distant. Considered together, these six regions account for approximately 77% of European Emmental production. The samples were supplied directly by the cheese factories (Austria, Finland, Switzerland) or were collected with the help of a national dairy research centre (France and Germany). The altitude of production is the altitude of the dairy farms. For alpine regions, the true altitude of pasture might be a little bit different due to the v steepness. All the samples were stored at 20 C until analysis.
Fig. 1. Geographic origins of the collected samples. Al ¼ Allg€au; BR ¼ Bretagne; CH ¼ Switzerland; FI ¼ Finland; SA ¼ Savoie; VO ¼ Vorarlberg.
290
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
2.2. Reagents and equipment All reagents used were of analytical grade (Merck, Darmstadt, Germany or Fluka, Buchs, Switzerland). The cheese samples were oven dried and ashed at 600 v C in a Heraus furnace. Digestion of ash from cheese samples was carried out in an MLS lltraCLAVE microwave apparatus or in an MLS Ethos Plus microwave apparatus (Milestone Inc., Monroe, CT). Chromatographic separations were carried out on a Bio-Rad AG 50W 8 (100–200 mesh) cationic exchanger (90Sr), or on an Eichrom UTEVA resin (U). An anion exchanger (Bio-Rad AG1 2, 100– 200 mesh) was used to isolate plutonium and remove thorium isotopes. The solutions were loaded onto the chromatographic column by a peristaltic pump (Ismatec IPS-8, Zu¨rich, Switzerland). Yttrium and strontium yields as well as calcium content were determined on a Perkin Elmer AS 4100 atomic absorption spectrometer (Y: 410.2 nm; Sr: 460.7 nm, Ca: 422.7 nm). Yttrium oxalate sources were measured on a Tennelec LB 4100 low-level proportional gas counter (Canberra Electronique, Paris, France). Alpha spectra were recorded using a Canberra Alpha Analyst spectrometer and PIPS detector of 450 mm2 in diameter (Canberra Electronique, Paris, France). c-spectra were recorded using a Canberra HPGe well detector GC4523 in a 4 ml geometry (ash) or in a 250 ml geometry (grated cheese) and the analytical software ‘‘Genie 2000’’. 2.3.
90
Sr measurements
Details of the separation technique used for determining the 90Sr in cheese samples have been described in an earlier paper (Geering et al., 1990). Briefly, to determine 90Sr, 90Y activity was measured. The higher b- energy of 90Y (2.3 MeV) led to a better counting efficiency and the shorter half-life (64.5 h) of 90Y made a radionuclidic purity control possible. The separation of strontium was achieved successively on two ion exchange columns (Dowex 50w, 50 ml, then 8 ml) using first a (trans-1,2-cyclohexylene-dinitrilo) tetraacetic acid (CyDTA) solution and then a sodium citrate solution to elute strontium. After adding an yttrium carrier, the latter solution was allowed to stand for about 10 days to let 90Y grow partially. Yttrium was then separated from strontium on another ion exchange column (Dowex 50w, 8 ml) by elution with a sodium malonate solution. Finally, yttrium oxalate was precipitated and a thin source was prepared by filtration on a 0.24 lm v Millipore cellulose ester filter (n GSWOP2400). To ensure a low background, the source activity was measured in anticoincidence in a Tennelec LB 4100 w proportional counter. Under these conditions, the background was found to be 0.004 dps. The source was measured typically for 100 h and intermediate results were automatically taken every 4 h to check the purity of 90Y. A detection limit of 0.05 Bq/kg can be achieved. 2.4. Measurements of actinides Five gram of cheese ash, 20 ml of 8 M HNO3 and 232U tracer (52:1 0:6 mBq) were introduced into a flask suitable for microwave digestion. In the case of plu-
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
291
tonium determination, 242Pu (44 0:4 mBq) was added. The digestion of ash was v v carried out using the following programme: 5 min to 100 C, 5 min to 170 C, 5 v v v min at 170 C, 5 min to 200 C and 5 min at 200 C. After cooling, the solution was filtered and pumped onto an anion exchange column (Bio-Rad AG1 2, 6 ml, 100–200 mesh, NO 3 form) and the effluent was transferred onto an Eichrom1 U/TEVA extraction cartridge (U/TEVA, Eichrom Europe, Paris, France, 2 ml, 50–100 mesh, NO 3 form). The stacked columns were washed with 10 ml 8 M HNO3, then with 40 ml 3 M HNO3. The two columns were separated and the U/TEVA column washed with 10 ml of 9 M HCl to remove thorium. Uranium was eluted from the U/TEVA column by 20 ml of 0.01 M HCl. In the case of plutonium determination, the anion exchange column was washed with 20 ml of concentrated HCl to remove thorium, then the plutonium was eluted by reduction of Pu4+ to Pu3+ with 50 ml of an hydroxylamine solution. The uranium fraction was evaporated and uranium electrodeposited on a stainless 2 steel disk in a HSO 4 =SO4 buffer at pH 1.9, over 75 min at 1.2 A. Plutonium was electrodeposited in the same way.
3. Results and discussion 3.1. Measurements of
90
Sr
The results of 90Sr determination are displayed in Table 1. Calcium content is used as an internal standard and the activities are reported as Bq g1 Ca. Table 1 shows that the calcium content of fresh cheese is almost constant and the average value is 9:64 0:5 g kg1 of cheese. 90Sr activities range from 0.034 to 0.121 Bq g1 Ca. The limited number of samples selected for this study makes it difficult to interpret the 90Sr activity in terms of the geographic origin of a given cheese sample but individual results clearly show a significant correlation (r ¼ 0:708, Student t-test ¼ 6:02) between the altitude of production of the cheese and the 90Sr content (Fig. 2), which makes it possible to distinguish Emmental type cheese produced in Bretagne and in Finland from those produced in Alpine regions (Switzerland, Savoie, Allga¨u and Vorarlberg). A study conducted by Schuller et al. (2002) on the fallout from global weapons of 137Cs in soils of south-central Chile demonstrated that the 137Cs deposition found in different locations correlates with altitude (r ¼ 0:714) and annual rainfall rate (r ¼ 0:791). Therefore, rainfall and snowfall can be considered as the driving forces for soil contamination by fallout. The 90Sr deposition following the Chernobyl accident has been very low in Western Europe. In particular, Friedli et al. (1991) observed that virtually all the Chernobyl 90Sr seems to have vanished from soil, grass and milk a few months after the deposition. Therefore, all the 90Sr measured in this study can be considered as radiostrontium from weapons. The results show that the geomorphology plays a critical role in the distribution of 90Sr from nuclear bomb test fallouts in soils and, consequently, in milk and cheese. In Bretagne, virtually no relief is present to generate strong precipitations leading to a high contamination of soil by 90Sr. A similar
96.72 96.60 96.47 95.88 96.97 96.93 96.36 92.28 96.70 96.79 96.62 96.21 96.73 96.61 96.87 96.87 96.33 96.85 96.72 96.53 96.54 96.50
AL1 AL2 AL3 BE1 BE2 BE3 BR1 BR2 BR3 SA1 SA2 SA3 SG1 SG2 SG3 VO1 VO2 VO3 FI1 FI2 FI1 FI2
760 800 900 480 470 640 100 100 100 540 350 540 750 610 450 950 710 690 200 300 200 300
Altitude of production 9:37 0:3 9:67 0:3 9:45 0:3 9:32 0:3 9:36 0:3 9:49 0:3 10:26 0:3 10:60 0:3 10:63 0:3 9:69 0:3 9:96 0:3 10:4 0:3 9:63 0:3 9:76 0:3 9:27 0:3 9:48 0:3 9:35 0:3 9:21 0:3 8:83 0:3 9:03 0:3 8:83 0:3 9:03 0:3
g Ca (kg fresh weight)1
Sr (Bq g1 Ca)
0:121 0:008 0:066 0:005 0:088 0:006 0:078 0:007 0:066 0:006 0:034 0:003 0:053 0:004 0:028 0:002 0:039 0:003 0:058 0:004 0:036 0:004 0:068 0:005 0:072 0:006 0:071 0:005 0:042 0:007 0:105 0:006 0:120 0:008 0:102 0:007 0:058 0:005 0:036 0:003
90
1:07 0:17 1:03 0:12 1:51 0:11 3:54 0:17 2:56 0:11 1:47 0:10 2:05 0:15 2:17 0:12 0:79 0:07 1:34 0:2 1:10 0:10 1:46 0:15 1:90 0:10 2:35 0:30 2:91 0:30 1:37 0:10 2:03 0:20 2:82 0:11 0:32 0:09 0:40 0:07 0:56 0:10 0:18 0:04
U (Bq g1 Ca)
234
0:79 0:17 0:82 0:12 1:09 0:15 2:90 0:20 2:14 0:11 1:05 0:10 1:27 0:10 1:51 0:09 0:52 0:09 1:24 0:20 0:83 0:14 1:23 0:13 1:35 0:10 1:33 0:20 2:59 0:20 1:05 0:10 1:82 0:21 2:34 0:10 0:27 0:06 0:28 0:05 0:38 0:10 0:17 0:04
U (Bq g1 Ca)
238
Al: Allga¨u (D), BE: Bern (CH), BR: Bretagne (F), SA: Savoie (F), SG: St-Gall (CH), VO: Vorarlberg (D), FI: Finland.
Percentage lost on ignition
Sampling area 1:35 0:25 1:26 0:15 1:38 0:14 1:22 0:12 1:20 0:12 1:40 0:15 1:61 0:18 1:44 0:10 1:53 0:15 1:08 0:20 1:31 0:15 1:19 0:14 1:41 0:15 1:77 0:20 1:12 0:15 1:30 0:15 1:12 0:15 1:18 0:10 1:18 0:4 1:42 0:4 1:47 0:3 1:1 0:2
U/238U
234
1:49 0:2 1:68 0:2 1:59 0:2 1:68 0:2 1:68 0:2 1:96 0:3 1:49 0:2 1:59 0:2 1:87 0:3 1:96 0:3 1:87 0:3 1:87 0:3 1:77 0:3 1:59 0:2 1:68 0:2 2:05 0:3 2:15 0:3 1:31 0:2 n.m. n.m. n.m. n.m.
K (Bq g1 Ca)
40
Table 1 Results of the radiochemical analyses carried out on different samples of Emmental type cheese of different origins (uncertainties of the measurements given in 2r)
292 P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
293
situation is encountered in Finland. The Emmental type cheese produced in these two countries displayed the lowest 90Sr activities. In Alpine regions, 90Sr cheese activity is clearly a matter of the altitude of production (Fig. 2). Gatsberger et al. (2001) demonstrated that not only 137Cs but also the 90Sr uptake from soil to plant is enhanced in Alpine pastures compared to lowland pastures. The results of 90Sr determination in grass in Switzerland during the years 2000–2002 (Froidevaux et al., 2003) display the same trends, and values of 2.8–6.4 Bq kg1 dry weight (dw) are obtained in lowland Switzerland (mainly Swiss lowland), 7.5–11.8 Bq kg1 in the Jura mountains and 8.7–15 Bq kg1 in Alpine and sub-Alpine pastures (Table 2). Furthermore, the results of 90Sr in grass normalised to the Ca content show that the mineral matrix (essentially Ca and K) of grass from Jura and the Alps is enriched in 90Sr compared to lowland pastures. In addition, 90Sr (Bq g1 Ca) content of milk is enhanced in Alpine regions compared to the Swiss lowland (average 0.05 Bq g1 Ca in the Swiss lowland vs. 0.20 Bq g1 Ca in the Davos region). This fact is supported by the larger 90Sr content of the mineral matrix of cheese produced in Alpine locations. 3.2. Measurements of
137
Cs and
40
K
137
Cs and 40K were determined by c-spectrometry on cheese ash (4 ml) or grated cheese (250 ml). 137Cs activity in cheese stays under the detection limit of 0.1 Bq kg1 for all the samples. Natural 40K activity is low compared to the activity measured in milk (average 40K in Swiss milk was 44 5 Bq g1 Ca, n ¼ 15, data from Geering et al., 2001, personal communication) and ranges in cheese between
Fig. 2. 90Sr activity (Bq g1 Ca) as a function of the altitude of cheese sampling location. Error bars are given in 2r.
294
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
Table 2 Results of the radiochemical analyses carried out on different samples of grass for 90Sr determination in Switzerland in 2002 (data from Froidevaux et al., 2003). Uncertainties of measurements are reported as 2r Sr Bq (kg dry weight)1
Sr Bq g1 Ca
Location of sampling
Swiss region
90
90
Givrine St-Cergue Intragna Sessa Rodi-Fiesso Coire Davos Gene`ve Go¨sgen Mu¨hleberg Villigen Beznau Grangeneuve
Swiss Jura Swiss Jura Tessin (south Alps) Tessin (south Alps) Tessin (south Alps) Est sub-Alps Est Alps Swiss lowland (west) Swiss lowland (centre) Swiss lowland (centre) Swiss lowland (north) Swiss lowland (north) Swiss lowland (west)
11:8 0:5 3:3 0:1 16:8 0:4 11:5 0:3 6:7 0:4 6:7 0:3 8:7 0:4 2:8 0:1 6:4 0:3 2:7 0:1 2:6 0:4 4:8 0:2 4:9 0:5
1:1 0:03 0:20 0:009 1:12 0:02 1:12 0:03 0:38 0:02 0:21 0:02 0:52 0:02 0:77 0:04 0:24 0:01 0:19 0:01 0:46 0:07 0:20 0:01 0:40 0:04
1.31 and 2.15 Bq g1 Ca. Based on the 40K in milk and in cheese samples, the decontamination factor for alkaline cations is about 20, which means that most of the alkaline cations are in a soluble form in which they are transferred to the whey during the cheese making process. Cheese making has been widely used as a milk decontamination process for 137Cs in areas contaminated by the Chernobyl accident (Patel and Prasad, 1993). 3.3. Measurements of uranium Uranium activity is low in all the cheese samples. The lowest values are found in Finland cheese (0.17–0.38 mBq 238U g1 Ca) and the highest ones in Switzerland cheese (1.07–2.90 mBq 238U g1 Ca). Measurements of uranium by a-spectrometry makes it possible to determine both 234 and 238 uranium isotopes. Surprisingly, depletion in 238U is observed in all the cheese samples, the highest depletion level being found in Bretagne cheese (234 U=238 U ¼ 1:53) and the lowest in cheese produced in the Savoie area (234 U=238 U ¼ 1:19). Depletion of 238U in water is a well known phenomenon. This reflects the process of a-recoil during the decay of 238U which leads to high 234U/238U ratios in river waters, and a flux of 234U into the oceans from marine sediment pore-water (Henderson, 2002). On the other hand, the soils present an equilibrium between the parent (238U) and the daughter (234U) or a slight depletion in 234U, due to the preferential weathering of 234U. In closed systems older than 1 106 y, such as the earth’s mantle, uranium decay chains are at equilibrium. Uranium analysis of grass samples collected in different regions of Switzerland shows a large range of activities (0.25–14.3 Bq kg1 dw) but no depletion in 238U. Instead, an equilibrium of uranium isotopes is observed in three samples and a slight depletion in 234U is observed in one case (Table 3 and Fig. 3), reflecting the situation taking place in the soil, e.g. no
Grass Grass Grass Grass
Rodi-Fiesso (CH)a Leibstadt (CH)a Davos (CH)a Mu¨hleberg (CH)a
Froidevaux et al. 2003.
Tap water Tap water Drinking water Drinking water Milk Milk Milk
Ullava (Fi) Lapinlahi (Fi) Swiss lowland Swiss Alps Swiss lowland 1 Swiss lowland 2 Swiss lowland 3
a
Type of sample
Site of sampling
0:25 0:01 1:29 0:04 14:3 0:3 2:71 0:08
234
U (Bq kg1 dw)
0:63 0:06 0:45 0:06 3:3 0:3
1:30 0:16 1:7 0:3
234
U (mBq l1)
0:25 0:01 1:33 0:04 14:3 0:3 3:08 0:08
U (Bq kg1 dw)
238
1:3 0:16 1:4 0:2 4–33 4–196 0:58 0:06 0:38 0:06 2:9 0:3
U (mBq l1)
238
0:85 0:02
0.0140.0005
U (Bq g1 Ca) 238
1:10 0:10 1:20 0:10 1:12 0:10
1:0 0:2 1:20 0:2
U/238U
234
1:00 0:06 0:97 0:04 1:00 0:03 0:88 0:05
U/238U
234
This work This work Bosshard (1992) Bosshard (1992) This work This work This work
Reference
Table 3 Results of the radiochemical analyses carried out on different samples of water, milk and grass. Uncertainties of measurements are reported as 2r
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298 295
296
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
Fig. 3. 234U/238U ratio in cheese and grass samples. Error bars are given in 2r.
fractionation of uranium isotopes during the transfer of uranium from the soil to the plant. One hypothesis which could explain the depletion in 238U in cheese is that the contamination of milk by uranium arises through drinking water instead of forage. It would mean that the metabolisms of drink and food are different. Taking into account an ingestion of 40 kg d1 of dry forage containing the lowest activity, e.g. 0.25 Bq 238U by the cow, the daily intake of 238U would be 10 Bq, with a 234 U/238U ratio of 1:00 0:06. With a milk yield of 30 l d1, the uranium excretion (based on cheese and milk content) through milk would however be at most 0:060 0:009 Bq d1 with a 234U/238U ratio close to 1:35 0:10 (Swiss cheese and milk average). This is only a very slight fraction of the intake that cannot be attributed to forage. On the other hand, if we take into account an ingestion of 60 l of water at 10 mBq l1 of 238U (Swiss average in drinking water, data from Bosshard et al., 1992), the daily intake would be only 0.6 Bq through drinking, with a 234U/238U ratio being close to 1.20. The uranium content of Finnish cheese was found to be significantly lower than the uranium content of the cheese produced in the other locations. As Finnish soils contain large deposits of uranium, this was an unexpected result. Therefore, samples of the drinking water of cows were collected in the two farms from which the two cheese samples originated and analysed for uranium. The results show that the two water samples have a very low uranium content (Table 2). These results seem to indicate that the uranium contamination of milk and cheese is due to the drinking water rather than to the forage. 3.4. Measurements of plutonium 238
Pu and 239/240Pu activities were determined on one sample from each location, except Finland (AL1, BE1, SA1, VO1, SG1, BR1), and were found to be less than
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
297
the detection limit of 0.3 mBq kg1 of cheese. The results confirm that weapon grade plutonium is strongly bound to the soil particles and is not transferred to a significant extent to the plant. 4. Conclusions The study of Emmental type cheese produced in six different Western European countries suggests that the 90Sr content of dairy products could be used to distinguish between the production zones. Clearly, Bretagne and Finland cheeses can be distinguished from Alpine cheeses on the base of 90Sr content. Geomorphology played a strong role in the distribution of weapons grade 90Sr, which is reflected in the significant correlation measured between 90Sr activity and the altitude of grazing. Furthermore, the results show that uranium isotope measurements in cheese can be used to get information on the mechanism of uranium transfer from feed to milk. In particular, this investigation throws some light on the role of water on the milk contamination by uranium. In the next step towards the modelling of the contamination of the dairy food chain by radionuclides, we intend to improve our knowledge of 90Sr and uranium pathways from the cow’s feed and drink by measuring 90Sr and uranium isotope activity in cow’s excreta such as urine and faeces. Acknowledgements F. Barraud and T. Schmittler are acknowledged for their technical help. This v work was supported by the Federal Office of Public Health under contract n 3189.001.4. References Bosshard, E., Zimmerli, B., Schlatter, C., 1992. Uranium in the diet: risk assessment of its nephro- and radiotoxicity. Chemosphere 24, 309–321. Bosset, J.-O., Albrecht, B., Badertscher, R., Dalla Torre, M., Gauch, R., Imhof, R., Isolini, D., Lavanchy, P., Meyer, J., Spahr, U., Wismer, M., Burger, M., Bontognali, R., Rezzonico, E., Quadroni, G., 2001. Caracte´risation microbiologique, chimique et sensorielle de laits, de caille´s et de fromages de che`vre tessinois de types Formaggini (Bu¨scion, Robiola) et Formaggella. Travaux de chimie alimentaire et d’hygie`ne 92, 546–580. Bundt, M., Albrecht, A., Froidevaux, P., Blaser, P., Flu¨hler, H., 2000. Impact of preferential flow on radionuclide distribution in soil. Journal of Environmental Science and Technology 34, 3895–3899. Ebbs, S., Brady, D., Norvell, W., Kochian, L., 2001. Uranium speciation, plant uptake, and phytoremediation. Practice Periodical of Hazardous, Toxic and Radioactive Waste Management 5, 130– 135. Friedli, C., Geering, J.-J., Lerch, P., 1991. Some aspects of behaviour of 90Sr in the environment. Radiochimica Acta 52, 237–240. Froidevaux, P., Geering, J.-J, Schmittler, T., Barraud, F., Valley, J.-F, 2001. Mesures de plutonium et d’ame´ricium dans l’environnement. Environmental radioactivity and radiation exposure in Switzer-
298
P. Froidevaux et al. / J. Environ. Radioactivity 72 (2004) 287–298
land. Feral Office of Public Health, Division of Radiation Protection, Chemin du Muse´e 3, 1700 Fribourg, Switzerland ((Chap. 7.1) (ISBN 3-905235-40-4)). Froidevaux, P., Geering, J.-J., Friedrich-Be´net, K. , Schmittler, T., Barraud, F., Valley, J.-F., Volkk, H., Jungok, M., Burger, M., Schmidt, E., Byrde, F., Jakob, A., Astner, M., 2003. Radioactivite´ du sol et de l’herbe. Environmental radioactivity and radiation exposure in Switzerland, chap. 4.3, Federal Office of Public Health, Division of Radiation Protection, Chemin du Muse´e 3, 1700 Fribourg, Switzerland. Gastberger, M., Steinhau¨sler, F., Gerzabeck, M.H., Hubmer, A., 2001. Fallout strontium and caesium transfer from vegetation to cow milk at two lowland and two Alpine pastures. Journal of Environmental Radioactivity 54, 267–273. Geering, J.-J., Friedli, C., Lerch, P., 1990. Method of determination of 90Sr in the environment. Journal of Trace Microprobe Techniques 8, 211–230. Geering, J.-J., Carlone, F., Valley, J.-F, 1995. Mesures de 90Sr dans le sol, l’herbe, le lait, le froment, les verte`bres et les dents. Environmental radioactivity and radiation exposure in Switzerland. Feral Office of Public Health, Division of Radiation Protection, Chemin du Muse´e 3, 1700 Fribourg, Switzerland ((Chap. B.3.11 et B3.12) (ISBN 3-905235-21-8)). Henderson, G., 2002. Seawater (234U/238U) during the last 800 thousand years. Earth Planetary Science Letters 199, 97–110. Lapham, S.C., Millard, J.B., Samet, J.M., 1989. Health implications of radionuclide levels in cattle raised near U mining and milling facilities in Ambrosia Lake, New Mexico. Health Physics 56, 327–340. Patel, A.A., Prasad, S.R., 1993. Decontamination of radioactive milk: a review. International Journal of Radiation Biology 63, 405–412. Rayno, D.R., 1983. Estimated dose to man from uranium milling via the beef/milk food chain pathway. Science of the Total Environment 31, 219–241. Schuller, P., Voigt, G., Handl, J., Ellies, A., Olivia, L., 2002. Global weapon’s fallout 137Cs in soils and transfer to vegetation in south-central Chile. Journal of Environmental Radioactivity 62, 181–193.