- Email: [email protected]
239Pu isotopic ratios and 239+240Pu total measurements in surface and deep waters around Mururoa and Fangataufa atolls compared with Rangiroa atoll (French Polynesia)
The Science of the Total Environment 237r238 Ž1999. 269]276
240
Pur 239 Pu isotopic ratios and 239q240 Pu total measurements in surface and deep waters around Mururoa and Fangataufa atolls compared with Rangiroa atoll ž French Polynesia/ R. ChiappiniU , F. Pointurier, J.C. Millies-Lacroix, G. Lepetit, P. Hemet Commissariat a Analyse et Sur¨ eillance de ` l’Energie Atomique, Direction DAM-Ile de-France, Departement ´ l’En¨ ironnement. BP 12, 91680, Bruyeres-le-Chatel, ` ˆ France
Abstract The average values of 240 Pur 239 Pu mass isotopic ratios of plutonium deposited in Mururoa and Fangataufa atoll sediments by French atmospheric nuclear tests range from 3.5 to 5%. In order to assess the near field and far field influence of those deposits in the open ocean, two water profiles were measured for 239q240 Pu and 240 Pur 239 Pu using, for the first time, an Inductively Coupled Plasma Mass Spectrometer which was developed to achieve femtogram detection limits. One site was located at the limit of the French territorial waters, which is 22 km distant from Mururoa. The second site was located close to Rangiroa atoll, at a distance of approximately 1200-km from French nuclear test sites. The sample volumes were approximately 500 litres and plutonium was purified prior to mass spectrometry and alpha spectrometry measurements. In Rangiroa, the 239q240 Pu profile is comparable with those already determined in world open oceans but the maximum detected activity, 9 mBqrm3 at 500]600 m is a lot lower than those measured in the northern hemisphere. 240 Pur 239 Pu ratios were measured between 500 and 1000 m and were not statistically different from the typical 0.18" 0.01 ratio which characterises the global fallout. Consequently, any influence of plutonium from the tests in Mururoa and Fangataufa is not apparent at Rangiroa. The vertical distribution of 239q240 Pu near Mururoa shows similar changes with depth but with a slight increase in concentration. 240 Pur 239 Pu mass ratios vary with depth, from 7 to 10% in the upper 500 m and in the deep waters Žbelow 1000 m. to 15]16% between 600 and 1000 m. A contribution from plutonium deposited in the sediments at Mururoa and Fangataufa is observed at the limit of territorial waters, especially in surface and deep waters. Q 1999 Elsevier Science B.V. All rights reserved. Keywords: Oceanic waters; Plutonium; Isotopic ratios; Mururoa
U
Corresponding author. Tel.: q33-1-69-26-53-70; fax: q33-1-69-26-70-65.
0048-9697r99r$ - see front matter Q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 9 9 . 0 0 1 4 1 - 2
270
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
1. Introduction From 1966 to 1974, France conducted atmospheric nuclear tests in the atolls of Mururoa and Fangataufa in French Polynesia. The atolls ŽFig. 1. are located in the central part of the south Pacific ocean at a distance of approximately 1200 km from Rangiroa Ž148479S, 1478059W.. In Mururoa, the tests were carried out above the lagoon, in the western zone ŽDindon. and in the northern zone ŽDenise.. The first three tests were conducted on barges, a few meters above the surface. As the firing point was not high, fissile materials and fission products were deposited in
the lagoon bottom sediments. In the other tests, the devices were suspended with balloons, several hundred meters above the surface and consequently, the effects on the near environment were considerably reduced. During that period, safety experiments were performed on the coral bedrock close to Denise zone which generated plutonium deposition on the ground and the lagoon. After each test, most of the plutonium was removed or included in bitumen but part of it remained. In Fangataufa, the deposit in the sediments was mainly due to a barge test conducted in 1966. The locations, the quantities of plutonium, and the 240r239 Pu ratios are reported on Fig. 2. Plutonium
Fig. 1. Map of French Polynesia: location of Mururoa, Fangataufa and Rangiroa atolls.
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
271
Fig. 2. 239q240 Pu activities, average 240 Pur 239 Pu mass isotope ratios deposited in the sediments of Mururoa and Fangataufa atolls and water column sampling location.
total activities have already been reported ŽBourlat et al., 1995; Musa et al., 1996. as well as isotopic ratios ŽChiappini et al., 1996, 1997.. The total quantity of plutonium deposited is approximately 2 = 10 13 Bq in Mururoa sediments and 7.4= 10 12 Bq in Fangataufa sediments. As the nature and the configuration of the tests were different one from another, the plutonium signatures varied accordingly. In Mururoa, safety tests yielded 240 Pur 239 Pu ratios lower than 0.03 and barge tests ratios close to 0.05. Consequently, in
the centre of the atoll, the ratios vary and the average value is close to 0.035. 240 Pur 239 Pu ratios in Fangataufa are quite constant at approximately 0.05 because the origin is unique. The plutonium deposited in the sediments is slowly remobilized into the water, leading to a higher concentration in the lagoon water than in the ocean. Those concentrations have been regularly measured and already published ŽBourlat et al., 1995.. Most of time, they are lower than 1 Bqrm3. As Mururoa and Fangataufa lagoons are
272
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
in contact with the ocean, open oceanic waters are supplied with a fraction of lagoon waters resulting in a slight increase in plutonium concentration close to the passes in the immediate vicinity of the atoll. Previous studies ŽBourlat et al., 1996; Hamilton et al., 1996. indicated that oceanic profiles at the Mururoa and Fangataufa territorial water limit and in French Polynesia were very similar. The general patterns were the same: plutonium concentration increased quickly with depth to a maximum value at 600 m. Then, it decreased quickly down to 1000 m and more slowly after. Those typical profiles, similar to those generally observed in world oceanic waters ŽBowen et al., 1980; Cochran et al., 1987; Sakanoue, 1987; Miyake and Saruhashi, 1988., led to the conclusion that the influence in the open ocean of plutonium deposited by France through atmospheric nuclear tests was negligible. Those conclusions were based on total plutonium measurements. On the other hand, total plutonium results are not sufficiently accurate to determine slight differences. Plutonium isotopic ratios, and more particularly 240 Pur 239 Pu, are more indicative of any difference but they have rarely been published because high quality of mass spectrometry and quantitative radiochemistry are needed to measure environmental levels. A complete method for such determinations has already been published ŽBuesseler and Halverson, 1987., involving Thermo Ionisation Mass Spectrometry. Only one paper ŽBertine et al., 1986. presented results in two sea water profiles from the north Pacific, one close to Bikini atoll and the second in the open ocean and discussed the origin of the signatures. The discrimination of various sources is made possible if one assume that global stratospheric fallout is well known. Global stratospheric fallout is characterised by a typical 0.18" 0.01 in marine sediments throughout the Northern Hemisphere ŽBuesseler and Sholkowitz, 1987.. This value is indistinguishable from the global mean of 0.176" 0.014 established following a worldwide survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.. In order to assess the near field effect of Mururoa and Fangataufa plutonium deposits, two water columns were measured in our laboratory
for total plutonium and 240 Pur 239 Pu ratios using a high-sensitivity Inductively Coupled Plasma Mass Spectrometer. That kind of instrument has never been used to measure plutonium in oceanic waters because, in the past, it suffered from an insufficient analytical performance. Recently, we modified and developed an instrument in order to increase sensitivity and to reach femtogram detection limits ŽChiappini et al., 1996.. With that configuration, we obtained sufficient precision on 240 Pur 239 Pu mass ratios in ocean waters to discriminate between various sources.
2. Experimental 2.1. Locations and conditions of sampling. The first water column was sampled in 1993 at the limit of territorial waters ŽFig. 2.. The second was collected in French Polynesia, close to Rangiroa atoll ŽFig. 1. in 1996. For the two columns, the sample volume was close to 500 l Žfrom 500 to 575 l.. Three hundred litre bottles were used to collect the samples. 2.2. Sample preparation procedure For Rangiroa samples, the waters were acidified to a pH of 1.7, on board immediately after sampling. For Mururoa samples, the waters were acidified a few hours after sampling. In both cases, the waters were completely filtered using a 0.45-mm pore size filter. Prior to the plutonium purification steps, known amounts of 236 Pu and 242 Pu were added to the filtrated samples. The sample and spike were then equilibrated for at least one night. Na 2 SO 3 solution was added to perform a redox cycle that lasted a minimum of 4 h as well as a pure Fe solution. Then, concentrated NH 4 OH was added to reach a pH of ; 9. The iron hydroxide that formed, carried the plutonium and was allowed to settle overnight. The precipitate was then dissolved in an 8 N HNO3 media and NaNO2 was added prior to anion exchange chromatographic separation. Dowex AG1-X8 resins ŽBioRad, Ivrysur-Seine, France. were used to separate most of
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
273
Fig. 3. ICP-MS instrumental detection limits for actinides as compared with that of alpha spectrometry on standard solutions. The line is relative to a 1.2 fg detection limit for ICP-MS. The relatively high detection limits for 232 Th, 235 U and 238 U are attributable to the high blank values of the nuclides present in the reagents and water used.
the matrix and most of the uranium from the plutonium. In parallel, the filters were ashed, dissolved in 8 N HNO3 and added to the solutions prior to the separations. Dowex AG1-X4 resins ŽBioRad. were then used to optimise the purification of plutonium from uranium. The plutonium was then eluted and electroplated on a stainless steel disc. 230q240 Pu was measured using alpha spectrometry. After this measurement, the plutonium plated on the disk was leached with hot HClrHNO3 and HF. The solution was then dried, purified through anionic chromatography and dissolved in 2% mrm ultra-pure nitric acid for ICP-MS measurements. High-sensitivity Inductively Coupled Plasma Mass Spectrometry was used in this study to perform the measurements. The performance of that system has already been described ŽChiappini et al., 1996.. The instrument was modified and developed to achieve femtogram detection limits for the actinide elements. It can be used to advantage for measurements of radionuclides with a radioactive half-life in excess of 650 years if a 4000-min measurement by alpha spectrometry provides only the detection limit ŽFig. 3.. On that figure, a detection limit for ICP-MS of 1.2 fg is used to compare the performance of mass and alpha spectrometry. With the methodology described above, sam-
ples from the Mururoa water column were analysed for 239q240 Pu and 240 Pur 239 Pu using alpha spectrometry and ICP-MS. Samples from the Rangiroa water column were analysed for 239q240 Pu using alpha spectrometry and partly for 239q240 Pu and 240 Pur 239 Pu using ICP-MS.
3. Results 3.1. Rangiroa water column The 239q240 Pu activities in samples collected from the surface to 2300 m and 240 Pur 239 Pu mass ratios in samples collected from 500 to 1000 m are presented in Fig. 4. There are small differences in the mass spectrometric and alpha plutonium measurements on three samples but the uncertainties, corresponding to a 95% confidence interval, indicate that all the results are statistically indistinguishable. Between the surface and 200 m, the plutonium activity is close to 2 mBqrm3. It increases quickly down to 500]600 m to reach a maximum of 9 mBqrm3 then it decreases rapidly to 1000 m and more slowly thereafter to reach a value slightly lower than 2 mBqrm3 at 2300 m. Those general patterns are comparable with those observed in world open oceans. However, as was expected,
274
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
Fig. 4. 239q240 Pu activity and 240 Pur 239 Pu mass isotope ratios measured in the water column sampled in Rangiroa: Ža. global stratospheric fallout, measured in Northern Hemisphere marine sediments ŽBuesseler and Sholkowitz, 1987.; Žb. global world mean ratio determined through a survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.; and Žc. 240 Pur 239 Pu average mass isotope ratios measured in Mururoa and Fangataufa sediments.
the maximum activity is a lot lower than those already measured in the Northern Hemisphere. In the north Pacific, Miyake and Saruhashi Ž1988. reported a maximum activity of approximately 20 mBqrm3, in 1978, at a depth of 500 m and Sakanoue Ž1987. indicated maximum concentrations ranging between 60 and 80 mBqrm3, in 1982, at depths between 500 and 800 m. In the northern Atlantic, maximum values of more than 30 mBqrm3 were ascertained during the years
1983 to 1985 at depths between 800 and 1000 m. Finally, a recent paper ŽMerino et al., 1997. indicates that, in the Catalan sea, the maximum concentration of plutonium is measured at 400 m and is of the order of 25 mBqrm3. 240 Pur 239 Pu mass ratios measured from 500 to 1000 m in the Rangiroa water column are similar and there is no significant difference at the 95% confidence interval. The mean value is 0.166 and the standard deviation is 0.011. This can be compared with the typical 0.18" 0.01 ratio which characterises the global stratospheric fallout measured in Northern Hemisphere marine sediments ŽBuesseler and Sholkowitz, 1987. and also the global mean of 0.176" 0.014 determined through a world-wide survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.. Even though 240 Pur 239 Pu mass ratios measured in Rangiroa are not statistically different from the global fallout ratios, they are a little lower. That difference could be explained by the stratospheric contribution of French atmospheric nuclear tests. Indeed, UNS considered that global deposition from tests occurred approximately 75% in the hemisphere of input. As an example, based on fission products measurements, the UN indicated ŽUN, 1993. that the estimated 137 Cs deposition from French tests in the Southern Hemisphere was 13%. The contribution of plutonium from French tests is not well quantified but as experiments were of low yield and produced lower 240 Pur 239 Pu mass ratios than 18%, that contribution could account for the slight decrease in plutonium ratios that have been determined in Rangiroa. 3.2. Mururoa water column The total 239q240 Pu activities and 240 Pur 239 Pu mass ratios in samples collected from the surface to 2300 and the associated 95% confidence intervals are presented in Fig. 5. Differences in the mass spectrometric and alpha measurements of total plutonium can be observed. This is probably caused by the use of different isotope dilution spikes, i.e. 236 Pu for alpha spectrometry vs. 242 Pu for mass spectrometry. Despite the differences, paired, two-tailed t-tests indicate no significant difference at the 95% confidence interval between
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
Fig. 5. 239q240 Pu activity and 240 Pur 239 Pu mass isotope ratios measured in the water column sampled at the Mururoa territorial water limit: Ža. global stratospheric fallout, measured in Northern Hemisphere marine sediments ŽBuesseler and Sholkowitz, 1987.; Žb. global world mean ratio determined through a survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.; and Žc. 240 Pur 239 Pu average mass isotope ratios measured in Mururoa and Fangataufa sediments.
ICP-MS and alpha spectrometry determinations. The uncertainties relative to the mass spectrometric measurements are lower than those obtained with alpha spectrometry which is consistent with the ICP-MS lowest detection limits. With respect to Rangiroa, the vertical distribution of 239q240 Pu shows similar changes with depth but the surface concentration is higher, 4]5 mBqrm3. The maximum activity, approximately 12 mBqrm3, was measured at 600 m. At 2300 m,
275
the concentration is slightly higher than 2 mBqrm3. 240 Pur 239 Pu mass ratios vary with depth. Between the surface and 100 m, ratios range between 7 and 10%. Then they increase down to 600 m. Between 600 and 1000 m, they range between 15]16% then they decrease down to 2300 m to reach values of 7]8% ŽFig. 5.. Those variations reflect the relative influence of Mururoa and Fangataufa lagoon waters. In Fig. 6 239q240 Pu activities from Rangiroa and Mururoa are given. Plutonium activities are reported from ICP-MS measurements except when those results are not available Žfrom the surface to 400 m and from 1250 to 2300 m in Rangiroa.. Plutonium concentrations are higher in Mururoa but the differences do not exceed 2 mBqrm3 and do not vary significantly with depth. Consequently, the relative contribution from Mururoa is higher in the surface region and in deep waters Žfrom 1250 m. when plutonium concentrations are low. This is consistent with the 240 Pur 239 Pu mass ratios profile presented in Fig. 5. The uncertainties on 240 Pur 239 Pu are of the order of 10]30% at the 95% confidence interval. This is due to the very low mass of 240 Pu in the samples, frequently of the order of 10]20 fg. Even if it is lower than what can be obtained using Thermo Ionisation Mass Spectrometry, it is sufficient to distinguish among various sources. In order to increase that precision, further developments are currently being investigated in our laboratory to decrease ICP-MS detection limit.
4. Conclusion A contribution of the plutonium deposited by the French atmospheric nuclear tests in the sediments of Mururoa and Fangataufa can only be observed in the oceanic waters at the limit of territorial waters, which is 22 km from the atolls. The plutonium concentrations in seawater at the 22-km limit exceed by only 2 mBqrm3 those observed at Rangiroa. That contribution is slight but plutonium ratio signatures are different and were measured for the first time using a high sensitivity Inductively Coupled Plasma Mass
276
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276 137
Fig. 6. Comparison between 239q240 Pu activities measured in the water column sampled in Rangiroa and in the water column sampled at the Mururoa territorial water limit.
Spectrometer. The differences are more clearly observed in surface and deep waters, below 1000 m, when plutonium concentrations are low. Between 600 and 1000 m, 240 Pur 239 Pu mass ratios do not differ considerably from global fallout. Any influence of plutonium from the tests at Mururoa and Fangataufa is not apparent at Rangiroa. References Bertine KK, Chow TJ, Koide M, Goldberg ED. J Environ Radioact 1986;3:189]201. Bourlat Y, Millies-Lacroix JC, Nazard R. Determination of plutonium radioactivity in Mururoa lagoon waters. J Radioanal Nucl Chem Articles 1995;197Ž2.:393]414. Bourlat Y, Millies-Lacroix JC, Lepetit G, Bourguignon J.
Cs, 90 Sr, and 239q240 Pu in world ocean water samples collected from 1992 to 1994. In: Radionuclides in the oceans. Inputs and Inventories. les Editions de Physique 1996:75]93. Bowen V, Noshkin VE, Livingston HD, Volchok HL. Fallout radionuclides in the Pacific Ocean: Vertical and horizontal distributions, largely from GEOSECS stations. Earth Planet Sci Lett 1980;49:411]434. Buesseler KO, Halverson JE. The mass spectrometric determination of fallout 239 Pu and 240 Pu in marine samples. J Environ Radioact 1987;5:425]444. Buesseler KO, Sholkowitz ER. The geochemistry of fallout plutonium in the north Atlantic: II. 240 Pur 239 Pu ratios and their significance. Geochim Cosmochim Acta 1987;51: 2623]2637. Chiappini R, Taillade JM, Brebion S. Development of a ´ high-sensitivity inductively coupled plasma mass spectrometer for actinide measurement in the femtogram range. J Anal At Spectrom 1996;11:497]503. Chiappini R, Taillade JM, Pointurier F. Low level plutonium and uranium ratio measurements using a high-sensitivity Inductively Coupled Plasma Mass Spectrometer. In: ESARDA Proceedings. Geel, Belgium, 25]27 February, 1997:129]137. Cochran JK, Livingston HD, Hirschberg DJ, Surprenant LD. Natural and anthropogenic radionuclide distributions in the Northwest Atlantic Ocean. Earth Planet Sci Lett 1987; 84:135]152. Hamilton TF, Millies-Lacroix. JC, Hong GH. 137 Cs, 90 Sr, and Pu isotopes in the Pacific Ocean: Sources and trends. In: Radionuclides in the oceans. Inputs and Inventories. Les Editions de Physique 1996:29]58. Krey PW, Hardy EP, Pachucky C, Rourke F, Coluzza J, Benson WK. Mass isotopic composition of global fallout plutonium in soils. Transuranium nuclides in the environment, IAEA-SM-199r39. Vienna: IAEA, 1976:671]678. Merino J, Sanchez-Cabeza JA, Bruach JM, Masque ´ P, Pujol LI. Artificial radionuclides in a High Resolution Water Column Profile from the Catalan Sea ŽThe Northwestern Mediterranean .. Radioprotection-Colloques 1997;32ŽC2.: 85]90. Miyake Y, Saruhashi K. Contents of 137 Cs, plutonium and americium isotopes in the southern ocean waters. Papers Meteorol Geophys 1988;39Ž3.:95]113. Musa C, Bourlat Y, Millies-Lacroix JC. Lagoon sediment radioactivity in Mururoa and Fangataufa. Report n801rSMSRBrDIR, 22 pages. Service Mixte de Surveillance Radiologique et Biologique, 4 March 1996. Sakanoue M. Transuranium nuclides in the environment. Radiochim Acta 1987;42:103]112. UN. Sources and effects of ionizing radiations. UN Scientific Committee on the Effects of Atomic Radiation, 1993 Report to the General Assembly, with scientific annexes. UN sales publication E.94.IX.2. UN, New York, 1993.
240
Pur 239 Pu isotopic ratios and 239q240 Pu total measurements in surface and deep waters around Mururoa and Fangataufa atolls compared with Rangiroa atoll ž French Polynesia/ R. ChiappiniU , F. Pointurier, J.C. Millies-Lacroix, G. Lepetit, P. Hemet Commissariat a Analyse et Sur¨ eillance de ` l’Energie Atomique, Direction DAM-Ile de-France, Departement ´ l’En¨ ironnement. BP 12, 91680, Bruyeres-le-Chatel, ` ˆ France
Abstract The average values of 240 Pur 239 Pu mass isotopic ratios of plutonium deposited in Mururoa and Fangataufa atoll sediments by French atmospheric nuclear tests range from 3.5 to 5%. In order to assess the near field and far field influence of those deposits in the open ocean, two water profiles were measured for 239q240 Pu and 240 Pur 239 Pu using, for the first time, an Inductively Coupled Plasma Mass Spectrometer which was developed to achieve femtogram detection limits. One site was located at the limit of the French territorial waters, which is 22 km distant from Mururoa. The second site was located close to Rangiroa atoll, at a distance of approximately 1200-km from French nuclear test sites. The sample volumes were approximately 500 litres and plutonium was purified prior to mass spectrometry and alpha spectrometry measurements. In Rangiroa, the 239q240 Pu profile is comparable with those already determined in world open oceans but the maximum detected activity, 9 mBqrm3 at 500]600 m is a lot lower than those measured in the northern hemisphere. 240 Pur 239 Pu ratios were measured between 500 and 1000 m and were not statistically different from the typical 0.18" 0.01 ratio which characterises the global fallout. Consequently, any influence of plutonium from the tests in Mururoa and Fangataufa is not apparent at Rangiroa. The vertical distribution of 239q240 Pu near Mururoa shows similar changes with depth but with a slight increase in concentration. 240 Pur 239 Pu mass ratios vary with depth, from 7 to 10% in the upper 500 m and in the deep waters Žbelow 1000 m. to 15]16% between 600 and 1000 m. A contribution from plutonium deposited in the sediments at Mururoa and Fangataufa is observed at the limit of territorial waters, especially in surface and deep waters. Q 1999 Elsevier Science B.V. All rights reserved. Keywords: Oceanic waters; Plutonium; Isotopic ratios; Mururoa
U
Corresponding author. Tel.: q33-1-69-26-53-70; fax: q33-1-69-26-70-65.
0048-9697r99r$ - see front matter Q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 9 9 . 0 0 1 4 1 - 2
270
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
1. Introduction From 1966 to 1974, France conducted atmospheric nuclear tests in the atolls of Mururoa and Fangataufa in French Polynesia. The atolls ŽFig. 1. are located in the central part of the south Pacific ocean at a distance of approximately 1200 km from Rangiroa Ž148479S, 1478059W.. In Mururoa, the tests were carried out above the lagoon, in the western zone ŽDindon. and in the northern zone ŽDenise.. The first three tests were conducted on barges, a few meters above the surface. As the firing point was not high, fissile materials and fission products were deposited in
the lagoon bottom sediments. In the other tests, the devices were suspended with balloons, several hundred meters above the surface and consequently, the effects on the near environment were considerably reduced. During that period, safety experiments were performed on the coral bedrock close to Denise zone which generated plutonium deposition on the ground and the lagoon. After each test, most of the plutonium was removed or included in bitumen but part of it remained. In Fangataufa, the deposit in the sediments was mainly due to a barge test conducted in 1966. The locations, the quantities of plutonium, and the 240r239 Pu ratios are reported on Fig. 2. Plutonium
Fig. 1. Map of French Polynesia: location of Mururoa, Fangataufa and Rangiroa atolls.
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
271
Fig. 2. 239q240 Pu activities, average 240 Pur 239 Pu mass isotope ratios deposited in the sediments of Mururoa and Fangataufa atolls and water column sampling location.
total activities have already been reported ŽBourlat et al., 1995; Musa et al., 1996. as well as isotopic ratios ŽChiappini et al., 1996, 1997.. The total quantity of plutonium deposited is approximately 2 = 10 13 Bq in Mururoa sediments and 7.4= 10 12 Bq in Fangataufa sediments. As the nature and the configuration of the tests were different one from another, the plutonium signatures varied accordingly. In Mururoa, safety tests yielded 240 Pur 239 Pu ratios lower than 0.03 and barge tests ratios close to 0.05. Consequently, in
the centre of the atoll, the ratios vary and the average value is close to 0.035. 240 Pur 239 Pu ratios in Fangataufa are quite constant at approximately 0.05 because the origin is unique. The plutonium deposited in the sediments is slowly remobilized into the water, leading to a higher concentration in the lagoon water than in the ocean. Those concentrations have been regularly measured and already published ŽBourlat et al., 1995.. Most of time, they are lower than 1 Bqrm3. As Mururoa and Fangataufa lagoons are
272
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
in contact with the ocean, open oceanic waters are supplied with a fraction of lagoon waters resulting in a slight increase in plutonium concentration close to the passes in the immediate vicinity of the atoll. Previous studies ŽBourlat et al., 1996; Hamilton et al., 1996. indicated that oceanic profiles at the Mururoa and Fangataufa territorial water limit and in French Polynesia were very similar. The general patterns were the same: plutonium concentration increased quickly with depth to a maximum value at 600 m. Then, it decreased quickly down to 1000 m and more slowly after. Those typical profiles, similar to those generally observed in world oceanic waters ŽBowen et al., 1980; Cochran et al., 1987; Sakanoue, 1987; Miyake and Saruhashi, 1988., led to the conclusion that the influence in the open ocean of plutonium deposited by France through atmospheric nuclear tests was negligible. Those conclusions were based on total plutonium measurements. On the other hand, total plutonium results are not sufficiently accurate to determine slight differences. Plutonium isotopic ratios, and more particularly 240 Pur 239 Pu, are more indicative of any difference but they have rarely been published because high quality of mass spectrometry and quantitative radiochemistry are needed to measure environmental levels. A complete method for such determinations has already been published ŽBuesseler and Halverson, 1987., involving Thermo Ionisation Mass Spectrometry. Only one paper ŽBertine et al., 1986. presented results in two sea water profiles from the north Pacific, one close to Bikini atoll and the second in the open ocean and discussed the origin of the signatures. The discrimination of various sources is made possible if one assume that global stratospheric fallout is well known. Global stratospheric fallout is characterised by a typical 0.18" 0.01 in marine sediments throughout the Northern Hemisphere ŽBuesseler and Sholkowitz, 1987.. This value is indistinguishable from the global mean of 0.176" 0.014 established following a worldwide survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.. In order to assess the near field effect of Mururoa and Fangataufa plutonium deposits, two water columns were measured in our laboratory
for total plutonium and 240 Pur 239 Pu ratios using a high-sensitivity Inductively Coupled Plasma Mass Spectrometer. That kind of instrument has never been used to measure plutonium in oceanic waters because, in the past, it suffered from an insufficient analytical performance. Recently, we modified and developed an instrument in order to increase sensitivity and to reach femtogram detection limits ŽChiappini et al., 1996.. With that configuration, we obtained sufficient precision on 240 Pur 239 Pu mass ratios in ocean waters to discriminate between various sources.
2. Experimental 2.1. Locations and conditions of sampling. The first water column was sampled in 1993 at the limit of territorial waters ŽFig. 2.. The second was collected in French Polynesia, close to Rangiroa atoll ŽFig. 1. in 1996. For the two columns, the sample volume was close to 500 l Žfrom 500 to 575 l.. Three hundred litre bottles were used to collect the samples. 2.2. Sample preparation procedure For Rangiroa samples, the waters were acidified to a pH of 1.7, on board immediately after sampling. For Mururoa samples, the waters were acidified a few hours after sampling. In both cases, the waters were completely filtered using a 0.45-mm pore size filter. Prior to the plutonium purification steps, known amounts of 236 Pu and 242 Pu were added to the filtrated samples. The sample and spike were then equilibrated for at least one night. Na 2 SO 3 solution was added to perform a redox cycle that lasted a minimum of 4 h as well as a pure Fe solution. Then, concentrated NH 4 OH was added to reach a pH of ; 9. The iron hydroxide that formed, carried the plutonium and was allowed to settle overnight. The precipitate was then dissolved in an 8 N HNO3 media and NaNO2 was added prior to anion exchange chromatographic separation. Dowex AG1-X8 resins ŽBioRad, Ivrysur-Seine, France. were used to separate most of
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
273
Fig. 3. ICP-MS instrumental detection limits for actinides as compared with that of alpha spectrometry on standard solutions. The line is relative to a 1.2 fg detection limit for ICP-MS. The relatively high detection limits for 232 Th, 235 U and 238 U are attributable to the high blank values of the nuclides present in the reagents and water used.
the matrix and most of the uranium from the plutonium. In parallel, the filters were ashed, dissolved in 8 N HNO3 and added to the solutions prior to the separations. Dowex AG1-X4 resins ŽBioRad. were then used to optimise the purification of plutonium from uranium. The plutonium was then eluted and electroplated on a stainless steel disc. 230q240 Pu was measured using alpha spectrometry. After this measurement, the plutonium plated on the disk was leached with hot HClrHNO3 and HF. The solution was then dried, purified through anionic chromatography and dissolved in 2% mrm ultra-pure nitric acid for ICP-MS measurements. High-sensitivity Inductively Coupled Plasma Mass Spectrometry was used in this study to perform the measurements. The performance of that system has already been described ŽChiappini et al., 1996.. The instrument was modified and developed to achieve femtogram detection limits for the actinide elements. It can be used to advantage for measurements of radionuclides with a radioactive half-life in excess of 650 years if a 4000-min measurement by alpha spectrometry provides only the detection limit ŽFig. 3.. On that figure, a detection limit for ICP-MS of 1.2 fg is used to compare the performance of mass and alpha spectrometry. With the methodology described above, sam-
ples from the Mururoa water column were analysed for 239q240 Pu and 240 Pur 239 Pu using alpha spectrometry and ICP-MS. Samples from the Rangiroa water column were analysed for 239q240 Pu using alpha spectrometry and partly for 239q240 Pu and 240 Pur 239 Pu using ICP-MS.
3. Results 3.1. Rangiroa water column The 239q240 Pu activities in samples collected from the surface to 2300 m and 240 Pur 239 Pu mass ratios in samples collected from 500 to 1000 m are presented in Fig. 4. There are small differences in the mass spectrometric and alpha plutonium measurements on three samples but the uncertainties, corresponding to a 95% confidence interval, indicate that all the results are statistically indistinguishable. Between the surface and 200 m, the plutonium activity is close to 2 mBqrm3. It increases quickly down to 500]600 m to reach a maximum of 9 mBqrm3 then it decreases rapidly to 1000 m and more slowly thereafter to reach a value slightly lower than 2 mBqrm3 at 2300 m. Those general patterns are comparable with those observed in world open oceans. However, as was expected,
274
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
Fig. 4. 239q240 Pu activity and 240 Pur 239 Pu mass isotope ratios measured in the water column sampled in Rangiroa: Ža. global stratospheric fallout, measured in Northern Hemisphere marine sediments ŽBuesseler and Sholkowitz, 1987.; Žb. global world mean ratio determined through a survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.; and Žc. 240 Pur 239 Pu average mass isotope ratios measured in Mururoa and Fangataufa sediments.
the maximum activity is a lot lower than those already measured in the Northern Hemisphere. In the north Pacific, Miyake and Saruhashi Ž1988. reported a maximum activity of approximately 20 mBqrm3, in 1978, at a depth of 500 m and Sakanoue Ž1987. indicated maximum concentrations ranging between 60 and 80 mBqrm3, in 1982, at depths between 500 and 800 m. In the northern Atlantic, maximum values of more than 30 mBqrm3 were ascertained during the years
1983 to 1985 at depths between 800 and 1000 m. Finally, a recent paper ŽMerino et al., 1997. indicates that, in the Catalan sea, the maximum concentration of plutonium is measured at 400 m and is of the order of 25 mBqrm3. 240 Pur 239 Pu mass ratios measured from 500 to 1000 m in the Rangiroa water column are similar and there is no significant difference at the 95% confidence interval. The mean value is 0.166 and the standard deviation is 0.011. This can be compared with the typical 0.18" 0.01 ratio which characterises the global stratospheric fallout measured in Northern Hemisphere marine sediments ŽBuesseler and Sholkowitz, 1987. and also the global mean of 0.176" 0.014 determined through a world-wide survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.. Even though 240 Pur 239 Pu mass ratios measured in Rangiroa are not statistically different from the global fallout ratios, they are a little lower. That difference could be explained by the stratospheric contribution of French atmospheric nuclear tests. Indeed, UNS considered that global deposition from tests occurred approximately 75% in the hemisphere of input. As an example, based on fission products measurements, the UN indicated ŽUN, 1993. that the estimated 137 Cs deposition from French tests in the Southern Hemisphere was 13%. The contribution of plutonium from French tests is not well quantified but as experiments were of low yield and produced lower 240 Pur 239 Pu mass ratios than 18%, that contribution could account for the slight decrease in plutonium ratios that have been determined in Rangiroa. 3.2. Mururoa water column The total 239q240 Pu activities and 240 Pur 239 Pu mass ratios in samples collected from the surface to 2300 and the associated 95% confidence intervals are presented in Fig. 5. Differences in the mass spectrometric and alpha measurements of total plutonium can be observed. This is probably caused by the use of different isotope dilution spikes, i.e. 236 Pu for alpha spectrometry vs. 242 Pu for mass spectrometry. Despite the differences, paired, two-tailed t-tests indicate no significant difference at the 95% confidence interval between
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276
Fig. 5. 239q240 Pu activity and 240 Pur 239 Pu mass isotope ratios measured in the water column sampled at the Mururoa territorial water limit: Ža. global stratospheric fallout, measured in Northern Hemisphere marine sediments ŽBuesseler and Sholkowitz, 1987.; Žb. global world mean ratio determined through a survey in terrestrial soils in 1970]1971 ŽKrey et al., 1976.; and Žc. 240 Pur 239 Pu average mass isotope ratios measured in Mururoa and Fangataufa sediments.
ICP-MS and alpha spectrometry determinations. The uncertainties relative to the mass spectrometric measurements are lower than those obtained with alpha spectrometry which is consistent with the ICP-MS lowest detection limits. With respect to Rangiroa, the vertical distribution of 239q240 Pu shows similar changes with depth but the surface concentration is higher, 4]5 mBqrm3. The maximum activity, approximately 12 mBqrm3, was measured at 600 m. At 2300 m,
275
the concentration is slightly higher than 2 mBqrm3. 240 Pur 239 Pu mass ratios vary with depth. Between the surface and 100 m, ratios range between 7 and 10%. Then they increase down to 600 m. Between 600 and 1000 m, they range between 15]16% then they decrease down to 2300 m to reach values of 7]8% ŽFig. 5.. Those variations reflect the relative influence of Mururoa and Fangataufa lagoon waters. In Fig. 6 239q240 Pu activities from Rangiroa and Mururoa are given. Plutonium activities are reported from ICP-MS measurements except when those results are not available Žfrom the surface to 400 m and from 1250 to 2300 m in Rangiroa.. Plutonium concentrations are higher in Mururoa but the differences do not exceed 2 mBqrm3 and do not vary significantly with depth. Consequently, the relative contribution from Mururoa is higher in the surface region and in deep waters Žfrom 1250 m. when plutonium concentrations are low. This is consistent with the 240 Pur 239 Pu mass ratios profile presented in Fig. 5. The uncertainties on 240 Pur 239 Pu are of the order of 10]30% at the 95% confidence interval. This is due to the very low mass of 240 Pu in the samples, frequently of the order of 10]20 fg. Even if it is lower than what can be obtained using Thermo Ionisation Mass Spectrometry, it is sufficient to distinguish among various sources. In order to increase that precision, further developments are currently being investigated in our laboratory to decrease ICP-MS detection limit.
4. Conclusion A contribution of the plutonium deposited by the French atmospheric nuclear tests in the sediments of Mururoa and Fangataufa can only be observed in the oceanic waters at the limit of territorial waters, which is 22 km from the atolls. The plutonium concentrations in seawater at the 22-km limit exceed by only 2 mBqrm3 those observed at Rangiroa. That contribution is slight but plutonium ratio signatures are different and were measured for the first time using a high sensitivity Inductively Coupled Plasma Mass
276
R. Chiappini et al. r The Science of the Total En¨ ironment 237 r 238 (1999) 269]276 137
Fig. 6. Comparison between 239q240 Pu activities measured in the water column sampled in Rangiroa and in the water column sampled at the Mururoa territorial water limit.
Spectrometer. The differences are more clearly observed in surface and deep waters, below 1000 m, when plutonium concentrations are low. Between 600 and 1000 m, 240 Pur 239 Pu mass ratios do not differ considerably from global fallout. Any influence of plutonium from the tests at Mururoa and Fangataufa is not apparent at Rangiroa. References Bertine KK, Chow TJ, Koide M, Goldberg ED. J Environ Radioact 1986;3:189]201. Bourlat Y, Millies-Lacroix JC, Nazard R. Determination of plutonium radioactivity in Mururoa lagoon waters. J Radioanal Nucl Chem Articles 1995;197Ž2.:393]414. Bourlat Y, Millies-Lacroix JC, Lepetit G, Bourguignon J.
Cs, 90 Sr, and 239q240 Pu in world ocean water samples collected from 1992 to 1994. In: Radionuclides in the oceans. Inputs and Inventories. les Editions de Physique 1996:75]93. Bowen V, Noshkin VE, Livingston HD, Volchok HL. Fallout radionuclides in the Pacific Ocean: Vertical and horizontal distributions, largely from GEOSECS stations. Earth Planet Sci Lett 1980;49:411]434. Buesseler KO, Halverson JE. The mass spectrometric determination of fallout 239 Pu and 240 Pu in marine samples. J Environ Radioact 1987;5:425]444. Buesseler KO, Sholkowitz ER. The geochemistry of fallout plutonium in the north Atlantic: II. 240 Pur 239 Pu ratios and their significance. Geochim Cosmochim Acta 1987;51: 2623]2637. Chiappini R, Taillade JM, Brebion S. Development of a ´ high-sensitivity inductively coupled plasma mass spectrometer for actinide measurement in the femtogram range. J Anal At Spectrom 1996;11:497]503. Chiappini R, Taillade JM, Pointurier F. Low level plutonium and uranium ratio measurements using a high-sensitivity Inductively Coupled Plasma Mass Spectrometer. In: ESARDA Proceedings. Geel, Belgium, 25]27 February, 1997:129]137. Cochran JK, Livingston HD, Hirschberg DJ, Surprenant LD. Natural and anthropogenic radionuclide distributions in the Northwest Atlantic Ocean. Earth Planet Sci Lett 1987; 84:135]152. Hamilton TF, Millies-Lacroix. JC, Hong GH. 137 Cs, 90 Sr, and Pu isotopes in the Pacific Ocean: Sources and trends. In: Radionuclides in the oceans. Inputs and Inventories. Les Editions de Physique 1996:29]58. Krey PW, Hardy EP, Pachucky C, Rourke F, Coluzza J, Benson WK. Mass isotopic composition of global fallout plutonium in soils. Transuranium nuclides in the environment, IAEA-SM-199r39. Vienna: IAEA, 1976:671]678. Merino J, Sanchez-Cabeza JA, Bruach JM, Masque ´ P, Pujol LI. Artificial radionuclides in a High Resolution Water Column Profile from the Catalan Sea ŽThe Northwestern Mediterranean .. Radioprotection-Colloques 1997;32ŽC2.: 85]90. Miyake Y, Saruhashi K. Contents of 137 Cs, plutonium and americium isotopes in the southern ocean waters. Papers Meteorol Geophys 1988;39Ž3.:95]113. Musa C, Bourlat Y, Millies-Lacroix JC. Lagoon sediment radioactivity in Mururoa and Fangataufa. Report n801rSMSRBrDIR, 22 pages. Service Mixte de Surveillance Radiologique et Biologique, 4 March 1996. Sakanoue M. Transuranium nuclides in the environment. Radiochim Acta 1987;42:103]112. UN. Sources and effects of ionizing radiations. UN Scientific Committee on the Effects of Atomic Radiation, 1993 Report to the General Assembly, with scientific annexes. UN sales publication E.94.IX.2. UN, New York, 1993.