Contamination of Japan 20 – Evaluation of a transuranic component in reactor derived releases from Fukushima Dai-ichi to the marine environment

Evaluation of a transuranic component in reactor derived releases from Fukushima Dai-ichi to the marine environment

Candise L. Henry 1, Timothy C. Kenna2, Pere Masqué3, Núria Casacuberta3, Ken O. Buesseler4
1Department of Earth & Environmental Sciences of Columbia University, New York, NY, USA, 2Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY, USA, 3Universitat Autònoma de Barcelona, Bellaterra, Spain, 4Woods Hole Oceanographic Institution, Woods Hole, USA,


The incident at Fukushima Dai-ichi on March 11, 2011 was the consequence of several events (earthquake, tsunami, equipment failures, etc.) and led to the release of radioactive contaminants into the atmosphere and ocean.

Although the specific details regarding the mechanism and/or the extent of radioactivity released to the environment are lacking, significant levels of fission/activation products such as 131I, 134Cs, and 137Cs have been measured in the vicinity of Fukushima.

As part of a research cruise in June, 2011 to study the problem, we received water samples collected between 30 and 600 km offshore. The main objective of our work is to determine the presence of/characterize a transuranic component in the reactor releases. Here we report results from Pu isotopic analysis of selected samples from the cruise. The Pu isotopic composition is diagnostic in resolving reactor-derived contamination from background levels associated with global fallout. Based on the relationship between the Fukushima derived Pu, 90Sr and 137Cs we estimate the total amount input to the study site from direct release.

1. Determine whether or not there is a transuranic component in the reactor releases.
2. Characterize the isotopic composition (239Pu, 240Pu, and 237Np)
3. Quantify the amount of Pu relative to the other radionuclides released from Fukushima Dai-ichi.

• Samples collected for UAB Sr/Cs analysis (see map right) • Modified method of Waples, J.T., Orlandini, K.A. (2010) to
separate Pu and Np • Column fractions sent to LDEO for additional purification and
analysis of Pu-239, Pu-240, Np-237 by ICP-MS (modified after Kenna (2002)

Background – Pre Fukushima (2009 NE Pacific data)

• Primary source of Pu and 237Np to the environment is fallout from nuclear weapons tests 1950s 1960s
240 239 • Pu/ Pu atom ratio typically ~0.18
• Ratio is elevated in the Pacific (~0.22) as a result of input from Tests conducted in the Marshall Islands (early 1950s)
• Cs and Np typically conservative in seawater • Pu more particle reactive (subject to removal on particles)

Fig. 1 Comparison of water column distributions between the SAFe site (30°N/ 140°W) samples collected in 2009 (solid symbols) and GEOSECS Station GX-202 (33.6°N/ 139.34°W) measured in 1973 (open symbols). (A) dissolved 239,240Pu. The subsurface maximum observed in the upper water column in 1973 has decreased and moved to deeper waters and bottom water Pu concentrations have increased. This is related to vertical fluxes of Pu-bearing particles and subsequent re-mineralisation by biological activity as well as physical circulation processes. Enhanced 239+240Pu concentrations in deep bottom waters in the Pacific Ocean have been observed and this may be explained by rapid transport of Pu isotopes attached to large particles through the water column to deep ocean sediments and re-suspension by near-bottom currents. (B) 137Cs – A decrease in the 137Cs maximum is also evident, but no subsidence is observed. (C) 237Np distribution is similar to 137Cs but longer half-life and better detection limits provide more detail. There is some evidence of 237Np at depth.

Fig. 2 Map showing the sampling stations of KOK-1108 in relation to Fukushima (red star) and the Kuroshio Current.

Fig. 3 239,240Pu Activities and 240Pu/239Pu atom ratios for selected samples. Left panel shows surface data; lower panel shows depth profiles. The dashed line represents the 240Pu/239Pu average value for nuclear weapons tests. The measured 240Pu/239Pu data are mostly consistent with fallout from weapons testing (0.22 in the N. Pacific). A few values are significantly higher than the fallout value. 237Np data are unavailable for these samples.

Fig. 4 (A) 240Pu/239Pu vs. 90Sr activity. 4 (B) 240Pu/239Pu vs. 137Cs activity. Linear regression indicates a strong correlation between increasing levels of 90Sr and 137Cs and increases in the 240Pu/239Pu atom ratio.

Fig. 5. Estimates of Fukushima derived Pu (yellow bars) and global fallout (green bars) in (left panel) surface samples and (lower panel) depth profiles. The Fukushima derived Pu accounts for between 0 and 50% of the total Pu measured in our samples.

Fig. 6 (A) 137Cs and Fukushima derived 239,240Pu in seawater samples; suggestion of two relationships (Discrete releases?, fractionation of Pu from Cs?) (B) Water column 137Cs and Fukushima derived Pu inventories. The linear relationship (R2 = 0.96) allows us to use estimates of 137Cs in the direct releases to the ocean from Fukushima (~3.5 x 1015 Bq) to predict the amount of Pu released (~6 x 109 Bq 239,240Pu). This is approximately 0.00004% of the total 239,240Pu released to the environment from anthropogenic nuclear activities (e.g., weapons testing, intentional and accidental releases), which is estimated to be ~15 x 1015 Bq.

Pu end-member mixing calculations
Based on the measured 134Cs/137Cs activity ratio (~1) the average burn-up of the fuel rods can is estimated to be ~25 gigawatt-days/metric ton of uranium (GWD/MTU), which in turn corresponds to a 240Pu/239Pu atom ratio ~0.46 (R. Web, personal communication). Using this value as the Fukushima 240Pu/239Pu end-member and 0.22 as the global fallout 240Pu/239Pu end-member, we can estimate the relative contributions for each source using a simple two end-member mixing model (eq. 1). Subscripts m, GF, and F indicate the measured 240Pu/239Pu value and the global fallout and Fukushima end-member values, respectively.

1. Pu isotopic analysis provides evidence of input from a source that is enriched in 240Pu relative to fallout from nuclear weapons tests and consistent with being derived from a nuclear reactor.
2. Application of a two end-member mixing model indicates that Fukushima derived Pu ranges between 0-50% of the total Pu measured.
3. The relationship between 137Cs and Fukushima derived 239,240Pu water column inventories is linear, allowing the total Pu contained in the direct releases from Fukushima to be estimated at ~6GBq.
4. This is likely a lower limit as it does not account for Pu loss due to particle scavenging in the water column

end quote.