Leaching of actinides and other radionuclides from matrices of Chernobyl “lava” as analogues of vitrified HLW

J. Chem. Thermodynamics 114 (2017) 25–29

Contents lists available at ScienceDirect

J. Chem. Thermodynamics journal homepage: www.elsevier.com/locate/jct

Leaching of actinides and other radionuclides from matrices of Chernobyl ‘‘lava” as analogues of vitrified HLW Bella Yu. Zubekhina, Boris E. Burakov ⇑ V.G. Khlopin Radium Institute (KRI), 2nd Murinsky av. 28, St. Petersburg 194021, Russia

a r t i c l e

i n f o

Article history: Received 11 July 2016 Accepted 25 August 2016 Available online 26 August 2016 Keywords: Chernobyl ‘‘lava” Actinides Leaching Nuclear glass

a b s t r a c t The paper commemorates the 30th anniversary of the severe nuclear accident at the Fourth Unit of the Chernobyl Nuclear Power Plant. Results of the investigation of radioactive glassy lava-like materials formed as a result of the accident are presented. Leaching of Pu, Am and Cs in distilled water and sodium carbonate solution at 25 °C from matrices of black and brown Chernobyl ‘‘lava” is comparable, however, leaching of Cs from black ‘‘lava” in seawater at 25 and 90 °C is almost 2 times higher than that for brown ‘‘lava”. Chemical alteration of black and brown ‘‘lava” in seawater is much more intensive that in distilled water and sodium carbonate solution. Comparison of leaching behavior of Chernobyl ‘‘lava” and aged samples of nuclear glass is not clear so far because of the lack of data. Further study of Chernobyl ‘‘lava” as analogues of vitrified high level waste (HLW) and possibly, Fukushima’s corium is needed. Ó 2016 Elsevier Ltd.

1. Introduction The severe nuclear accident at the Fourth Unit of the Chernobyl Nuclear Power Plant (ChNPP) on 26 April 1986 was accompanied by full destruction of the reactor core. Interaction between hot destroyed fuel cladding and silicate materials of reactor (concrete, sand, serpentinite) caused formation of a lava-like highly radioactive melt, which penetrated into different premises below the reactor basement and solidified, forming the so called Chernobyl ‘‘lava” or fuel-containing masses [1–14]. Some direct measurements of ‘‘lava” volume gave at least 192 m3 [4], although there is still uncertainty related to precise evaluation of total volume of fuelcontaining masses and amount of fuel (calculated on UO2) remaining inside the ‘‘Shelter” (the enclosure built around the ruined reactor) [4,13]. Scientists of the V.G. Khlopin Radium Institute (KRI) took part in the study of the scenario and consequences of the Chernobyl accident as early as May 1986. However most samples of Chernobyl ‘‘lava” from the KRI archive were manually collected and transported to Leningrad (now St. Petersburg) only in 1990 (Fig. 1). It is assumed that the KRI collection of Chernobyl ‘‘lava” is the only one in the world still available for further research. There are two main types of these ‘‘lava” samples, having black and brown color (Fig. 2). Brown color appears to be caused by numerous tiny

⇑ Corresponding author. E-mail addresses: [email protected] (B.Yu. Zubekhina), burakov@peterlink. ru (B.E. Burakov). http://dx.doi.org/10.1016/j.jct.2016.08.029 0021-9614/Ó 2016 Elsevier Ltd.

inclusions of uranium oxide phases containing up to several wt% zirconium [10]. Black ‘‘lava” contains fewer inclusions and its color can be related to uranium dissolved in the glass matrix and to radiation defects. In some samples the glass matrix of black ‘‘lava” contains iron up to 6–7 wt% but usually the content of Fe in black glass matrix does not exceed 1 wt% [6,10]. Thus the color is not an obvious indicator of oxidation state or other physicochemical conditions of formation. Bulk chemical composition of Chernobyl ‘‘lava”, results of electron-probe microanalyses of glass-like ‘‘lava” matrices (avoiding inclusions of different crystalline phases) and content of main radionuclides in ‘‘lava” are summarized in Tables 1–3. In 1990 some newly formed material of yellow color was observed on the surface of Chernobyl ‘‘lava” in different places [4]. Study of this material at KRI [11] confirmed that it consists of different uranyl phases such as UO4
4H2O (analogue of natural studtite); UO3
2H2O (analogue of epiianthinite); UO2  CO3 (analogue of rutherfordine) and Na4(UO2)(CO3)3. Also the sodium carbonate phases Na3H(CO3)2
2H2O and Na2CO3
H2O were identified among the secondary uranium minerals [11]. This observation proved that active chemical alteration of Chernobyl ‘‘lava” is going on at the present time inside the ‘‘Shelter”. Analysis of chemical composition of water existing inside the ‘‘Shelter” [12] confirmed that such solutions are alkaline (pH = 8.5–10) with high content of carbonate and bicarbonate ions (up to 2 g/l and 8 g/l, respectively). Chemical alteration supports radionuclide migration from ‘‘lava” matrices into groundwater which should be kept under permanent monitoring. Moreover, possible formation of secondary

26

B.Yu. Zubekhina, B.E. Burakov / J. Chem. Thermodynamics 114 (2017) 25–29

Fig. 1. Large fragment of Chernobyl ‘‘lava” prepared for packing and delivery to Leningrad. Photo had been taken by B. Burakov inside ‘‘Shelter” in 1990.

uranium minerals on the surface of some ‘‘lava” samples kept at KRI under laboratory conditions has been observed in 2011, when part of KRI Chernobyl collection was repacked (Fig. 3). Although Chernobyl ‘‘lava” contain inclusions of some crystalline phases including different solid solutions ‘‘UO2–ZrO200 , high-uranium zircon, (Zr1xUx)SiO4, etc. [5,8], their matrices are characterized by relatively homogeneous distribution of radionuclides [14]. Therefore the glass-like matrix of the ‘‘lava” might be considered as an analogue of aged (30 years old) vitrified high level radioactive waste (HLW). A study of radionuclide leaching from Chernobyl ‘‘lava” provides very important information to be used for both thermodynamic and kinetic modeling of long-term behavior of vitrified HLW. Some results obtained from the study of Chernobyl ‘‘lava” can be used to predict physico-chemical properties of the ‘‘corium” formed at the Fukushima nuclear power plant. This paper summarizes current and earlier data on chemical durability of Chernobyl ‘‘lava” in distilled water, simulated sea water and aqueous solution of sodium-carbonate.

Fig. 2. Some samples of Chernobyl ‘‘lava” from KRI collection: (1) brown ‘‘lava” from steam-discharge corridor and (2) black ‘‘lava” from Elephant Foot” (room 217). Photos had been taken by B. Burakov and V. Zirlin in 2011.

Table 1 Simplified bulk chemical composition of Chernobyl ‘‘lava” [6,7]. Type of ‘‘lava”

Chemical composition, wt%

Black Brown

U

Zr

Na

Fe

Mg

Ca

Si

Al

4–5 8–7

2–6 5–6

2–10 4

0.3–6 1–2

1–5 4

3–13 5

19–36 31–33

3–8 4

Table 2 Results of electron-probe microanalyses of glass-like silicate matrix of Chernobyl ‘‘lava” avoiding inclusions of crystalline phases [6,10]. Type of ‘‘lava”

Black Brown

Chemical composition, wt% U

Zr

K

Na

Fe

Mg

Ca

Si

Al

2.7–4.0 2.0–2.4

3.1–3.7 2.4–2.9

1.4–2.7 1.2–2.3

0.4 0.6

0.3–6.7 0.2–0.4

1.2–3.2 3.5–4.4

5.1–7.2 4.5–8.2

28–37 37

2.7–4.4 2.8–4.0

Table 3 Content of main radionuclides in Chernobyl ‘‘lava” on 06.2013 [15] and recalculated for 26.04.1986 [1,7]. Type of ‘‘lava”

Radionuclides, Bq/g 137

Cs

Black Brown

7

2
10 (2.3
107) 4.1
107

144

Ce 9

(2
10 ) (2.1
109)

154

Eu 5

5
10 (1.3
106) 1.2
106

244

Cm 4

5
10 (1.2
107) 1.1
105

241

Am 6

1.2
10 (3.5
107) 2.8
106

239,240

Pu

238

5

4.3
107 (3.8
107) 9.2
105

8.2
10 (7.3
107) 1.8
106

Pu

B.Yu. Zubekhina, B.E. Burakov / J. Chem. Thermodynamics 114 (2017) 25–29

27

Fig. 3. Secondary uranium minerals (marked by arrows) presumably formed under laboratory conditions at KRI during period since 1990 (after arrival to KRI) till 2011 (when collection of ‘‘lava” samples was repacked and visually restudied). Photos by B. Burakov and V. Zirlin in 2011.

2. Samples and methods All samples described in the paper were collected inside the ‘‘Shelter” in 1990, brought to KRI and kept under laboratory conditions since 1990. Self-destruction (disintegration) of some pieces of black ‘‘lava” was observed in 2011 but only solid pieces (without visible cracks) of both ‘‘lava” types were used for leaching. For experiments in distilled water and sodium carbonate solution the following samples have been used: (1) Two samples of black ‘‘lava”, of approximate weight 0.6 and 0.8 g. Gamma-radiation dose at distance 0.1 m was 0.06 and 0.10 mSv/h, respectively; (2) Two samples of brown ‘‘lava”, of approximate weight 0.3 and 0.4 g. Gamma-radiation dose at distance 0.1 m was 0.10– 0.14 mSv/h. In order to prevent possible contamination of solutions by tiny ‘‘lava” particles each sample has been placed on a glass filter and then put into the Teflon test vessel (Fig. 4). During experiments all vessels were closed with caps to prevent evaporation. All tests were carried out at approximately normal atmospheric pressure 101–102 kPa. Tests with distilled water and sodium carbonate solution were carried out for 7, 14, 28, 56 and 270 days, with replacement of solution after each test. For each test the time counting restarts

after replacement of the solution. After experiments the volume of leachate has been measured to make sure that it is equal to initial volume. In order to provide constant stable temperature 90 °C the vessels were placed into the oven with thermostatic regulation system (±1 °C). All experiments at room temperature were carried out without precise temperature stabilizing and highest temperature range proposed is within 20–25 °C. Sampling of leachate was carried out only between the walls of the glass filter and the Teflon vessel. Possible sorption of radionuclides on the surface of glass and Teflon was ignored. Sodium carbonate solution simulating real alkaline solutions inside ‘‘Shelter” was prepared using calcined Na2CO3 and distilled water in order to achieve concentration 2 g
L1 (pH = 11). In order to study leaching behavior of Chernobyl ‘‘lava” in seawater, it was decided to carry out long-term static leach tests at 25 and 90 °C for several months using the following simulant of seawater of the following chemical composition (in g/l): NaCl – 24; MgCl2 – 5; Na2SO4 – 4; CaCl2 – 1; KCl – 0.7; NaHCO3 – 0.2; KBr – 0.1. Two pieces of black ‘‘lava” and two samples of brown ‘‘lava” of approximate weight 0.2–0.3 g were used for this experiment. Each ‘‘lava” fragment has been placed into Teflon test vessel with 20 ml of model seawater. Two pieces (black and brown ‘‘lava”) were set at 90 °C, another ones were set at 25 °C. Leachates have been analyzed by alpha and gamma spectrometry. A Canberra 7401 alpha spectrometer has been used for determination of actinides (239Pu, 238Pu, 241Am, 243Am, 244Cm). For

Fig. 4. Samples of brown (1) and black (2) Chernobyl ‘‘lava” prepared for static leach tests in distilled water and sodium-carbonate solution. Pieces of ‘‘lava” lie on a glass filter placed into solution inside the Teflon test vessel. Sampling of solution for analysis of radionuclides was carried out between walls of glass filter and Teflon vessel.

28

B.Yu. Zubekhina, B.E. Burakov / J. Chem. Thermodynamics 114 (2017) 25–29

Table 4 Normalized mass loss of actinides and fission products from static leach test in distilled water at 25 °C. ‘‘lava” type

NL, g/m2

Days

Pu + 240Pu

241

Am + 238Pu

239

137

154

Cs

Eu

4

Black

7 14 28 56 270

0.3 3.1 0.2 1.4 0.7

0.4 9.4 0.7 2.0 1.2

<10 5.9
103 8.0
104 <104 5.0
102

<103 1.5
102 <103

Brown

7 14 28 56 270

0.1 1.9 0.04 1.3 1.4

0.2 4.4 0.14 1.2 1.8

2.8
103 1.2
102 5.6
103 <104 0.2

<103

137

154

1.2
102 <103

Table 5 Normalized mass loss of actinides and fission products from static leach test in sodium carbonate solution at 25 °C. ‘‘lava” type

NL, g/m2

Days

Pu + 240Pu

241

Am + 238Pu

239

Black

Brown

Table 6 Normalized mass loss of ‘‘lava” type

Black Brown

137

Cs

Eu

7 14 28 56 270

0.2 1.5 0.1 1.0 1.2

0.3 4.0 0.2 2.1 2.0

4.8
10 4.8
103 5.0
103 <104 7.0
102

<103

7 14 28 56 270

0.2 1.3 0.1 0.9 1.2

0.3 3.6 0.3 1.3 0.9

1.6
104 1.0
103 3.0
103 <104 4.0
102

4.7
102 <104 5.3
103 <103 0.5

Cs from static leach test in seawater for 4 months. NL, g/m2 25 °C

90 °C

0.3 0.1

7.8 4.2

determination of fission products (137Cs, 134Cs, 154Eu, 144Ce) a Canberra gamma-spectrometer with multi-channel analyzer DSA1000 and Ge-detector has been used. Normalized mass loss (NL) was calculated as follows: NL = A W/A0S, where A – total activity of radionuclide in solution and absorbed on the walls of test vessels after leaching, Bq; A0 – initial activity of radionuclide in the specimen, Bq; W – initial mass of the specimen, g; S – specimen geometric surface area (recalculated from sample volume), m2.

3

1.2
102 <103

3. Results and discussion The results of leach tests are summarized in Tables 4–6. Expanded uncertainties Ur (for level of confidence 0.95) range from 30 to 50% of measured normalized mass loss. Although the amount of fuel (uranium oxide) dissolved in matrix of brown Chernobyl ‘‘lava” is two times higher than that in black ‘‘lava” there are no essential differences in chemical resistance of both ‘‘lava” types in distilled water and sodium carbonate solution at 25 °C. It is interesting to notice an essential NL decrease after 28 days in all experiments (Tables 4 and 5). It might be explained by formation of thin layer of silica gel preventing leaching from the ‘‘lava” surface and its further mechanical destruction. In seawater leaching of Cs at 25 and 90 °C increased dramatically (Table 6). Chemical alteration was clearly visually observed for black ‘‘lava” after 4 months at 90 °C in seawater (Fig. 5), but

Fig. 5. Fragment of black Chernobyl ‘‘lava” before (1) and after (2) leach-alteration test in simulated seawater at 90 °C for 4 months.

29

B.Yu. Zubekhina, B.E. Burakov / J. Chem. Thermodynamics 114 (2017) 25–29 Table 7 Results of long-term leach tests for different types of nuclear glass. Glass

Radionuclide content

T, °C

Time, years

NL, g/m2

Refs.

K-26

137

Uncontrolled

1 16 1 6 No data

2.6 13 6
105 (all b-c-emitters) 18
105 (all b-c-emitters) (2.5–3.5)
day1

[16]

SRS Tank 15 SON68

Cs 3.7
103 Bk/g 137 Cs 2.3
105 Bk/g 244 Cm 1.2 wt% 244 Cm 3.25 wt%

No data

the sample of brown ‘‘lava” remained visually similar to the initial material. The attempt to compare our results obtained from leach tests of Chernobyl ‘‘lava” with previous published data on study on aged nuclear glasses [16–18] was not completed in detail but some data are given in Table 7. This reflects the lack of information related to the study of glass samples doped with high amounts of real radionuclides. 4. Conclusions The results obtained allow us to make the following conclusions: leaching of Pu, Am and Cs in distilled water and sodium carbonate solution at 25 °C from matrices of black and brown Chernobyl ‘‘lava” is comparable despite bulk content of uranium (originated from fuel) in brown ‘‘lava” is higher than that in black ‘‘lava”. Leaching of Cs from black ‘‘lava” in seawater at 25 and 90 °C is almost 2 times higher than that for brown ‘‘lava”. It can be caused by partial Cs incorporation into some numerous and perhaps less reactive crystalline phases (inclusions) in brown ‘‘lava” when main part of Cs in black ‘‘lava” is related to glass matrix. Chemical alteration of black and brown ‘‘lava” in seawater is much more intensive than in distilled water and sodium carbonate solution. Such effects should be taken into consideration for modeling behavior of Fukushima’s corium, which can be surrounded by seawater. Comparison of leaching behavior of Chernobyl ‘‘lava” and aged samples of nuclear waste glass is hampered by general lack of data and, therefore, further study of Chernobyl ‘‘lava” as analogues of vitrified HLW is needed. The data obtained in this study can help constrain both thermodynamic and kinetic modeling of leaching. Acknowledgements The work was partly supported by V.G. Khlopin Radium Institute. Authors are very grateful to Mrs. Larisa Nikolaeva and Mr. Vladimir Zirlin of the V.G. Khlopin Radium Institute for sampling of Chernobyl ‘‘lava” under extreme conditions and their help in all experiments. Authors thank Prof. Alexandra Navrotsky of the University of California, Davis, for valuable discussion and important technical comments. References [1] A.A. Borovoy, B.Ya. Galkin, A.P. Krinitsyn, V.M. Markushev, E.M. Pazukhin, A.N. Kheruvimov, K.P. Checherov, New products formed by reaction of fuel with construction materials in the 4th block of the Chernobyl NPP, Sov. Radiochem. 32 (6) (1990) 659–667.

[17] [18]

1.7 day1

[2] V.G. Savonenkov, A.S. Krivokhatskii, Yu.V. Dubasov, E.M. Pazukhin, Radioactive man-made formations from the destroyed 4th Unit of Chernobyl NPP, Radiochemistry 33 (4) (1991) 140–149. [3] Yu.M. Rogozin, E.A. Smirnova, V.G. Savonenkov, A.S. Krivokhatsky, V.A. Avdeev, E.Yu. Sagaidachenko, Leaching of radionuclides from some newly formed products, extracted from reactor zone of 4th unit of Chernobyl NPP, Radiochemistry 33 (4) (1991) 160–167. [4] A.N. Kiselev, A.Yu. Nenaglyadov, A.I. Surin, K.P. Checherov, Experimental study of lava-like fuel containing masses (FCM) at 4th Unit of ChNPP (based on results obtained in 1986–1991). Issue of IAE, Moscow 1992 (in Russian). [5] E.B. Anderson, B.E. Burakov, E.M. Pazukhin, High-uranium zircon from Chernobyl lavas”, Radiochim. Acta 60 (1993) 149–151. [6] M. Trotabas, J-Y. Blanc, B. Burakov, E. Anderson, J. Duco, Examination of Chernobyl samples. Impact on the accident scenario understanding. Report DMT/92/309, SETIC/LECR-92/36, Report IPSN/93/02, Report RI-1-63/92, March 1993. [7] E.M. Pazukhin, Fuel-containing lavas of the Chernobyl NPP 4th block – topography, physicalchemical properties, formation scenario, Radiochemistry 36 (2) (1994) 109–154. [8] B.E. Burakov, E.B. Anderson, B.Ya. Galkin, E.M. Pazukhin, S.I. Shabalev, Study of Chernobyl ‘‘hot” particles and fuel containing masses: implications for reconstruction the initial phase of the accident, Radiochim. Acta 65 (1994) 199–202. [9] Yu.V. Dubasov, V.G. Savonenkov, E.A. Smirnova, Classification of radioactive products formed during the Chernobyl accident, Radiochemistry 38 (2) (1996) 95–109. [10] B.E. Burakov, E.B. Anderson, S.I. Shabalev, E.E. Strykanova, S.V. Ushakov, M. Trotabas, J.-Y. Blanc, P. Winter, J. Duco, The behaviour of nuclear fuel in first days of the Chernobyl accident, in: Materials Research Society Symposium Proceedings Scientific Basis for Nuclear Waste Management XX, vol. 465, 1997, pp. 1297–1308. [11] B.E. Burakov, E.B. Anderson, E.E. Strykanova, Secondary uranium minerals on the surface of chernobyl ‘‘Lava”. In: Materials Research Society Symposium Proceedings Scientific Basis for Nuclear Waste Management XX, vol. 465, 1997, pp. 1309–1311. [12] A.P. Krinitsyn, I.Ya. Simanovskaya, O.L. Strikhar, Study of interaction of water with constructional and fuel-containing materials inside ‘‘Shelter”, Radiochemistry 40 (3) (1998) 279–288. [13] A.A. Borovoy, E.P. Velihov, Experience of Chernobyl, Part 1, Moscow, 2012 (in Russian). [14] I. Vlasova, A. Shiryaev, B. Ogorodnikov, B. Burakov, E. Dolgopolova, R. Senin, A. Averin, Y. Zubavichus, S. Kalmykov, Radioactivity distribution in fuelcontaining materials (Chernobyl ‘‘lava”) and aerosols from the Chernobyl ‘‘Shelter”, Radiat. Meas. 83 (2015) 20–25. [15] R. Nasirow, P. Poeml, Gamma-Ray Spectrometry of Chernobyl Ceramic Samples, Internal Report of JRC Institute of Transuranium Elements, Karlsruhe, 2013. [16] M.I. Ojovan, J.H. Russel, N.V. Ojovan, W.E. Lee, Corrosion of alkali-borosilicate waste glass K-26 in non-saturated conditions, J. Nucl. Mater. 340 (2005) 12– 24. [17] C.M. Jantzen, D.I. Kaplan, N.E. Bibler, D.K. Peeler, M.J. Plodinec, Performance of a buried radioactive high level waste (HLW) glass after 24 years, J. Nucl. Mater. 378 (2008) 244–256. [18] S. Peuget, J.-M. Delaye, C. Jegou, Specific outcomes of the research on the radiation stability of the French nuclear glass towards alpha decay accumulation, J. Nucl. Mater. 444 (2014) 77–91.

JCT 16-559