Leaching Behavior ofa Simulated Bituminized Radioactive Waste Form

Document Sample
Leaching Behavior ofa Simulated Bituminized Radioactive Waste Form Powered By Docstoc
					 Journal of NUCLEAR SCIENCE and TECHNOLOGY, Vol. 40, No. 4, p. 227–237 (April 2003)

          Leaching Behavior of a Simulated Bituminized Radioactive Waste Form
                           under Deep Geological Conditions
                     Shinichi NAKAYAMA* , Yoshihisa IIDA, Tetsushi NAGANO and Toshiyuki AKIMOTO
                             Japan Atomic Energy Research Institute, Tokai-mura, Naka-gun, Ibaraki 319-1195
                                (Received November 6, 2002 and accepted in revised form February 3, 2003)

                The leaching behavior of a simulated bituminized waste form was studied to acquire data for the performance as-
            sessment of the geologic disposal of bituminized radioactive waste. Laboratory-scale leaching tests were performed
            for radioactive and non-radioactive waste specimens simulating bituminized waste of a French reprocessing company,
            COGEMA. The simulated waste was contacted with deionized water, an alkaline solution (0.03-mol/l KOH), and a
            saline solution (0.5-mol/l KCl) under atmospheric and anoxic conditions. The concentrations of Na, Ba, Cs, Sr, Np,
            Pu, NO3 , SO4 and I in the leachates were determined.
                Swelling of the bituminized waste progressed in deionized water and KOH. The release of the soluble compo-
            nents, Na and Cs, was enhanced by the swelling, and considered to be diffusion-controlled in the swelled layers of the
            specimens. The release of sparingly soluble components such as Ba and Np was solubility-limited in addition to the
            progression of leaching. Neptunium, a redox-sensitive element, showed a distinct difference in release between anoxic
            and atmospheric conditions. The elemental release from the bituminized waste specimens leached in the KCl was very
            low, which is likely due to the suppression of swelling of the specimens at high ionic strength.
             KEYWORDS: bituminized radioactive waste, geological disposal, leaching behavior, swelling

I. Introduction                                                         ditions; a 0.03-mol/l potassium hydroxide (KOH) solution
                                                                        representing cement-contacting, highly alkaline groundwater,
   Spent nuclear fuel generated in Japan is being reprocessed
                                                                        and a 0.5-mol/l potassium chloride (KCl) solution simulating
in France and England. Radioactive wastes produced dur-
                                                                        a saline environment such as seawater for possible repository
ing the reprocessing, and extracted uranium and plutonium
                                                                        construction in a coastal area. Deionized water was also used
as well have been returned to Japan. Bituminized radioactive
                                                                        for reference and for a simulant of fresh type groundwater.
waste, which will be returned from the French reprocessing
                                                                        The respective leaching behavior including swelling of the
company COGEMA, is included in such wastes. This waste
                                                                        waste form, degradation of the bitumen matrix, and the re-
contains transuranium (TRU) elements and is to be disposed
                                                                        lease of salt components and waste elements from the waste
of in a deep geologic repository in Japan.1) The performance
                                                                        are described in this paper.
assessment of the disposal requires data on leachability of ra-
dionuclides from the waste form as source term.
   Leaching behavior of bituminized waste generally depends             II. Experimental
on the physical properties and the chemical composition of              1. Preparation of Simulated Bituminized Waste
the waste form, and the chemical environment at the repos-                 Radioactive and non-radioactive bituminized waste speci-
itory. The COGEMA bituminized waste contains a solu-                    mens were synthesized using a straight bitumen 60/80 (Showa
ble salt, sodium nitrate (NaNO3 ), and an insoluble precipi-            Shell Sekiyu K.K.) as the waste matrix. The chemical com-
tation sludge (barium sulfate, BaSO4 ), as two major compo-             positions were simplified compared with the real ones.2) The
nents.2) Waste elements are embedded in either one or both of           simplified non-radioactive waste were mainly used for phys-
these materials dispersed in a matrix of bitumen. Because the           ical and chemical characterization of the bituminized mate-
COGEMA bituminized waste will be disposed of in a deep                  rials, observation of the degradation of the waste matrix (bi-
geological formation in Japan, a study on the behavior of this          tumen), and leaching tests of salt components (Na+ , NO3 − ,
waste form under anoxic conditions is necessary. To date, the           Ba2+ , SO4 2− ) and waste element simulants (stable isotopes
leaching behavior of the COGEMA bituminized waste has not               of strontium (Sr), cesium (Cs) and iodine (I)). Two different
been studied much3, 4) and the release of the embedded waste            types of radioactive bituminized specimens were fabricated
elements under deep geological conditions has been scarcely             to study the leaching behavior of radioactive waste elements.
reported. This is especially true for the release of TRU ele-           One type comprised non-radioactive bituminized materials
ments such as neptunium (Np) and plutonium (Pu).                        doped with radioactive 134 Cs, 237 Np and 239 Pu, the other type
   We have performed laboratory-scale leaching tests un-                doped with only radioactive 90 Sr. The characteristics of these
der atmospheric and anoxic conditions for the COGEMA-                   specimens are described in Table 1.
type bituminized waste, to investigate the leaching behav-                 The salt components and waste elements were added to
ior of this waste form and the release of embedded waste                molten bitumen to prepare bituminized specimens. One of
elements. The leachants employed simulate disposal con-                 the waste elements is Sr, and preconditioning was needed for
                                                                        Sr prior to its introduction into the bitumen. Considering the
     Corresponding author, Tel. +81-29-282-6001, Fax. +81-29-282-
                                                                        much higher quantity of Ba than Sr in the actual waste,2) and
     5934, E-mail: nakayama@sparclt.tokai.jaeri.go.jp

228                                                                                                                    S. NAKAYAMA et al.

Table 1 Sample description of synthetic, simulated bituminized                  Sr-doped (Ba,Sr)SO4 .
  waste used for leaching tests                                                  Care was taken for all types of bituminized waste to avoid
                                                                              sedimentation of heavy components, (Ba,Sr)SO4 , during the
     Dimension                    50 mmφ ×15 mmH
     Weight                       34.7±1.8 g (46 specimens)
                                                                              mixing of molten bitumen and the chemicals in a heated
     Penetrability                4.3±0.9 mm (25◦ C)                          beaker. Sedimentation results in heterogeneity of the waste
     Softening point              52±0.3◦ C                                   form and makes a proper interpretation of leaching data im-
     Ignition point               >300◦ C                                     possible. Stirring of the mixture in a heated beaker had to be
     Dynamic viscosity            (8.31±0.26)×10−4 m2 /s (120◦ C)             rigorous enough to avoid sedimentation, but had to be gentle
                                  (6.79±0.13)×10−5 m2 /s (180◦ C)             enough on the other hand to avoid splashing.
                            Chemical contents
                                                                              2. Characterization of Synthetic Bituminized Waste
                     Non-                     Radioactive                         Samples
                  radioactive    134                                             Physical and chemical homogeneity is essential to prop-
                                     Cs-237 Np-239 Pu-   90
                                containing specimen            specimen       erly estimate the leachability of embedded elements through
                                                                              the leaching tests. Homogeneity is reflected by the uniform
     Bitumen        60 wt%             60 wt%                   60 wt%
                                                                              dispersion of waste element-bearing salt particles in the bi-
     NaNO3         17.5 wt%          17.5 wt%                 17.5 wt%
     (Ba,Sr)SO4     20 wt%             20 wt%                   20 wt%
                                                                              tuminized specimens. As mentioned above, particular care
     CsI            2.5 wt%           2.5 wt%                  2.5 wt%        was required to avoid sedimentation of the heavy component,
        Sr            —                 —                     200 kBqa)       (Ba,Sr)SO4 .
         Cs           —              140 kBqb)                   —               The uniform dispersion was examined by two different
         Np           —              3.0 MBq                     —            methods; differential-thermo analysis (DTA) and chemical
         Pu           —               60 kBq                     —            analysis followed by chemical separation of organic and in-
a)   Doped to (Ba,Sr)SO4
                                                                              organic components. The DTA was used for determination of
b)   Doped to an aqueous mixture solution of CsI and NaNO3                    NaNO3 and the water content. Determination of BaSO4 was
                                                                              made by chemical analysis.
                                                                                 The solidified, 15mm-thick bituminized waste specimens
the chemical analogy of these two elements, we expect that                    were sliced into three disks of about 5mm thickness; upper,
Sr is present as a Ba–Sr–SO4 ternary solid, (Ba,Sr)SO4 , in the               middle and bottom parts. A few grams of each of the thin-
COGEMA bituminized waste form. We, therefore, prepared                        ner disks were weighed and subjected to DTA and chemi-
the (Ba,Sr)SO4 solid beforehand. This solid was produced                      cal analysis. DTA detects the melting temperature of NaNO3
by precipitation in a mixture of aqueous solutions of barium                  and quantifies its amount based on the heat of fusion needed.
nitrate, strontium nitrate and sodium sulfate. The precipi-                   In the chemical analysis, the thinner disks were contacted
tate was washed by deionized water several times followed                     with trichloroethylene to dissolve the organic component, bi-
by freeze-drying. X-ray powder diffractometry indicated for-                  tumen. The solid residue and the dark liquid were separated
mation of a crystalline solid of (Ba,Sr)SO4 . The amount of Sr                by centrifugation and decantation of the liquid. This sepa-
is much smaller than that of Ba in the actual waste form, but a               ration was repeated until the liquid phase became colorless.
higher Sr–Ba ratio of 1:9 was employed in this study to allow                 The residue contained the inorganic salts, and was weighed
reliable analyses for Sr.                                                     after drying in air. The weight difference between this dried
   The non-radioactive bituminized waste specimens were                       solid and the initial specimen is the weight of the bitumen
prepared as follows; bitumen was melted by heating in a                       content in the specimens. The dried solid contains soluble
1-liter stainless steel beaker at 120 to 140◦ C, the aqueous                  NaNO3 , and insoluble precipitation sludge particles (BaSO4 ).
CsI-NaNO3 mixture was added to the molten bitumen by a                        Deionized water was added to dissolve NaNO3 . The aque-
tube pump under stirring to evaporate excessive water, the                    ous phase containing NaNO3 was separated from the insol-
pulverized (Ba,Sr)SO4 was added, portions of the stirred                      uble solid (BaSO4 ) by centrifugation followed by decanta-
molten mixture of bitumen and chemicals were transferred to                   tion. The final residue was dried and weighed. The DTA
50 mmφ ×15 mmH stainless steel containers, and finally, the                    yielded an average NaNO3 content of 19.1±0.7 wt% for the
mixture was allowed to cool at room temperature.                              upper, middle and bottom parts of a 50 mmφ ×15 mmH spec-
   In preparing the 134 Cs-237 Np-239 Pu-containing specimens,                imen, which indicates homogeneous dispersion of NaNO3
    Pu in a 3-mol/l nitric acid solution was added to molten bi-              salt particles in the specimen. The chemical analysis cor-
tumen, followed by additions of 237 Np and 134 Cs dissolved in                roborated homogeneous dispersion: The contents of bitumen,
CsI-NaNO3 mixture solutions, and by pulverized (Ba,Sr)SO4 .                   NaNO3 , and BaSO4 were 58.8±1.1 wt%, 19.8±0.2 wt% and
Before the additions of these solutions, sodium bicarbonate                   21.4±1.3 wt%, respectively. Similar analytical results were
(NaHCO3 ) solution as a pH buffer and sodium hydroxide                        obtained from several other specimens, which shows uniform
(NaOH) solution as a neutralization agent were added, in or-                  dispersion of embedded chemical elements in fabricated bitu-
der to protect the bitumen from strongly acidic and alkaline                  minized waste specimens.
conditions.                                                                      Microscopic observations were also carried out. Environ-
   The other type of radioactive waste, 90 Sr-containing spec-                mental Scanning Electron Microscopy (E-SEM) was applied
imens, were prepared by mixing molten bitumen with                            to observe surfaces of bituminized waste specimens. Surfaces
the aqueous CsI-NaNO3 solution followed by addition of                        were also scanned with Energy-Dispersive X-ray diffractom-

                                                                                   JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY
Leaching Behavior of a Simulated Bituminized Radioactive Waste Form                                                                    229

etry (EDX) to analyze elemental distributions of Na, Ba and               runs for differing leaching environments (leachant, tempera-
sulfur (S). Unlike conventional SEM, E-SEM allows pres-                   ture, atmospheric/anoxic condition). Each run typically con-
sures around the saturated water vapor pressures in the sam-              sisted of 10 to 15 samples with leaching periods ranging
ple cell, and can provide images of wet solid surfaces and                from 1 day to longer than 200 days. Each of the samples
water-containing samples. We obtained E-SEM images of                     was dedicated for sampling at a fixed leaching time. At
the surfaces of bituminized waste specimens before and af-                the fixed sampling times, pH and redox potential (E H ) of
ter leaching.                                                             the leachates were measured, and chemical analyses were
   The physical characteristics of the bituminized specimens              made. Three milliliters of the leachates were removed, and
such as penetrability,5) softening point,5) ignition point6) and          filtered through a 0.45 µm membrane filter to analyze the dis-
viscosity5) were measured in accordance with the Japanese                 solved fractions. Stable isotopes (Na, Sr, I, Cs, and Ba), as
Industrial Standards. A penetrability of 4.3±0.9 mm (25◦ C),              well as 237 Np and 239 Pu in the leachates were determined
a softening point of 52±0.3◦ C, and an ignition point of                  by inductively coupled plasma mass spectroscopy, ICP-MS
>300◦ C were obtained for 18 specimens. The average dy-                   (Yokogawa Analytical Systems Inc., HP-4500), and anions
namic viscosity for five samples was (8.31±0.26)×10−4 m2 /s                (NO2 − , NO3 − and SO4 2− ) were determined by ion chro-
at 120◦ C and (6.79±0.13)×10−5 m2 /s at 180◦ C. These values              matography. Interfering ions such as Cl− were removed by
are also listed in Table 1.                                               an ion exchange resin prior to the anion determination. Ra-
                                                                          dioisotopes of 90 Sr, 134 Cs and 237 Np were determined radio-
3. Leaching Tests                                                         metrically: 90 Sr through β-counting with a liquid scintilla-
   Bituminized specimens cast in 50 mmφ ×15 mmH stain-                    tion counter (LSC), and 134 Cs and 237 Np through γ -counting
less steel containers were contacted with 190 ml of aque-                 with an HPGe detector. Non-radioactive Cs and radioactive
ous leachants in sealed polypropylene containers under atmo-                  Cs released from the 134 Cs-237 Np-239 Pu-containing speci-
spheric and anoxic conditions. Only flat end-surfaces of the               mens were determined using ICP-MS and an HPGe detector,
specimens were exposed to the leachants. The initial surface-             respectively, and the results of the two determinations showed
to-volume ratio was 0.002 m2 /190 ml=10 m−1 .                             good agreement. Agreements were also obtained between
   Employed leachants were mainly deionized water, a 0.03-                ICP-MS determination of non-radioactive Sr and radiomet-
mol/l potassium hydroxide (KOH) solution and a 0.5-mol/l                  rically determined 90 Sr, both of which were released from the
potassium chloride (KCl) solution. A 0.01-mol/l calcium hy-                  Sr-containing specimens, even though fairly large measure-
droxide [Ca(OH)2 ] solution was also used in a few runs. The              ment errors were associated with the ICP-MS determination
KOH and Ca(OH)2 solutions simulate cement-reacted, highly                 due to the very low concentrations. Neptunium-237 released
alkaline groundwater, and the KCl solution simulates a saline             from the 134 Cs-237 Np-239 Pu-containing specimens was deter-
environment such as seawater. The concentration of the KCl                mined by both ICP-MS and HPGe, and the results were in
solution, 0.5-mol/l, was chosen as a typical value of the ionic           good agreement. The observed good reproducibility between
strength of seawater. Deionized water represents fresh type               different measurements demonstrates the appropriateness of
groundwater and is also used for the reference leachant. We               the applied analytical procedures.
did not employ NaOH and NaCl as leachant solutions to al-                    Degradation of waste matrices was studied using non-
low analyses of Na ions leached from bituminized samples.                 radioactive bituminized waste specimens. The specimens
Under atmospheric conditions, Ca precipitates out from solu-              were leached with deionized water or 0.01-mol/l Ca(OH)2 at
tions as CaCO3 (s). The concentration of Ca(OH)2 as leachant              25◦ C and 45◦ C, and the total organic carbon (TOC) in the
was hence limited to values below the solubility of CaCO3 (s).            leachates was determined. Because no organic substances
   The leaching tests were performed at the temperatures of               were added to our synthetic bituminized waste specimens, the
25±1◦ C and 45±1◦ C. The temperature of 45◦ C was se-                     TOC represents the release of organic materials from bitumen
lected in consideration of the elevated temperature at reposi-            itself, and therefore, is indicative of the degree of degradation
tory depth and of the upper limitation (softening point: 52◦ C)           of bitumen. Microscopic observation on the surfaces was per-
of the solid waste form. Radioactive runs were carried out at             formed before and after leaching.
only 45±1◦ C. Leaching tests under anoxic conditions were                    As described above, leaching tests under anoxic conditions
performed in a controlled atmosphere glove box under argon                were performed in a controlled atmosphere glove box un-
(O2 <1 ppm). Reducing conditions were maintained by using                 der argon. Bituminized waste specimens used for the tests
0.01-mol/l hydrazine monohydrate solutions.a                              were introduced into the glove box through a gas-exchanging
   Leaching tests were conducted in duplicate and triplicate              port, placed in polypropylene containers and contacted with
                                                                          leachants in the glove box to initiate the leaching runs. Air
                                                                          enclosed in bituminized waste specimens during the fabrica-
     A different agent, sodium hydrosulfite (Na2 S2 O4 ), may be effec-
                                                                          tion was degassed in the gas-exchanging port. This generated
     tive but was not used in this study. The choice was made in terms
     of minimizing effects on the determinations of anions leached
                                                                          fine pores on some of the specimens. This physical defor-
     from bituminized waste specimens. Hydrazine produces NO3 −           mation probably affected the leaching behavior. As will be
     as a result of its oxidation. The amount is anticipated to be much   mentioned later, releases of embedded materials from 90 Sr-
     smaller than that of NO3 − leached from the waste specimens, and     containing bituminized waste specimens under anoxic condi-
     will not yield large errors in the NO3 − determination. Sodium hy-   tions were appreciably higher than those from other two types
     drosulfite produces SO4 2− through its oxidation. The amount of       of specimens (non-radioactive specimens, 134 Cs-237 Np-239 Pu-
     produced SO4 2− may be much larger than that of released SO4 2−      containing specimens).
     during leaching, and makes the determination impossible.

VOL. 40, NO. 4, APRIL 2003
230                                                                                                                                                                                                                          S. NAKAYAMA et al.

III. Results and Discussion                                                                                                                                                       from bituminized waste is controlled by the same mechanism.
                                                                                                                                                                                  The released fractions of Na and Cs were in the order of 1%.
1. Leaching of Bituminized Waste Elements
                                                                                                                                                                                  This means that only a few percent of the initial volume of the
(1) Release of Soluble Salt Components (NaNO3 , Cs, I)
                                                                                                                                                                                  bituminized specimens were leached, and that corresponding
   The release of soluble salt components from non-                                                                                                                               amounts of embedded elements were contacted with the per-
radioactive, synthetic bituminized waste is plotted in Fig. 1.                                                                                                                    colated leachants.
The fraction of Na released at 45◦ C under atmospheric con-                                                                                                                          The release of nitrate (NO3 − ) and iodine (I− ) is plotted in
ditions is plotted in Fig. 1(a). This fraction increased with                                                                                                                     correlation with the release of Na+ and Cs+ , respectively, in
time up to about 3% during 224 days in deionized water and                                                                                                                        Figs. 1(c) and (d). Nitrite (NO2 − ) was detected in the leach-
0.03-mol/l KOH solutions. The rapid increase in release cor-                                                                                                                      ing tests under anoxic conditions. The molar concentration
responded to initiation of swelling of the waste samples. The                                                                                                                     of NO2 − was about two orders of magnitude lower than that
release of Na in 0.5-mol/l KCl solutions was lower than in the                                                                                                                    of NO3 − . There are two possible pathways for generation of
other two leachants. Such suppression of release from bitu-                                                                                                                       NO2 − in the systems investigated: Reduction of NO3 − and
minized waste was observed earlier,7–9) and this observation                                                                                                                      oxidation of the employed reductant, hydrazine. In the first
will be discussed later. Because Na is not redox-sensitive, the                                                                                                                   case, the NO2 − would have to be taken into account as part
chemical speciation is not affected by the surrounding redox                                                                                                                      of the released components, but for this possibility there is no
environment and therefore similar results were obtained also                                                                                                                      clear experimental evidence and it is not considered further.
under the anoxic conditions. The release at 45◦ C was five to                                                                                                                      The concentration of nitrate is plotted for anoxic conditions in
ten times higher than that at 25◦ C. This temperature effect of                                                                                                                   Fig. 1(c). No distinguishable difference was revealed between
enhancing release was common to all leaching experiments.                                                                                                                         the release of Na+ and NO3 − and its time dependence. This
For this reason, only results obtained at 45◦ C are presented for                                                                                                                 implies that Na+ and NO3 − are released from bituminized
discussion in the following parts of this paper; no quantitative                                                                                                                  waste congruently.
analysis on the temperature effect is presented.                                                                                                                                     The bitumen surface is negatively charged through de-
   The released fraction of Cs was similar to that of Na as
                                                                                                                                                                                  protonation of carboxyl groups and provides adsorption sites
shown in Fig. 1(b); Cs is also not redox-sensitive and only                                                                                                                       for cations. It could be anticipated that the release of Na+
the results for atmospheric conditions are shown. The results                                                                                                                     from bituminized waste specimens is slower than that of
of Figs. 1(a) and (b) indicate that the release of these cations                                                                                                                  NO3 − , and that the concentrations of Na+ in leachates are
                                                                                                                                                                                  lower than those of NO3 − due to the adsorption of Na+ on
                           5                                                                                       5
                                                                                                                                                                                  the bitumen surface. However, the density of the adsorp-
                                   (a) Na, 45 C, atmospheric
                                                                                                                           (b) Cs, 45 C, atmospheric        0.8                   tion sites on the bitumen surface is in the order of 1018 /m2
                           4                                                                                       4
                                                                                                                                                                                  (∼1 µmol/m2 ),10) which is much smaller than the number of
                                                                                          Released fraction ( )
  Released fraction ( )

                                                                     12                                                                                     0.6
                                                                                                                                                                                  Na+ ions released in the leachants (in the order of mmol).
                                                                                                                                                                  [Cs] (mmol/L)
                                                                          [Na] (mmol/L)

                           3                                                                                       3

                                                                     8                                                                                      0.4                   Neither a slower release (retardation) nor lower concentration
                                                                                                                                                                                  of Na+ in comparison to NO3 − was observable.

                           1                                         4                                             1                                        0.2
                                                                                                                                                                                     The concentrations of I− in the leachates are plotted in Fig.
                           0                                         0                                             0                                       0
                                                                                                                                                                                  1(d) in correlation to the concentrations of Cs+ . The gen-
                               0          100       200
                                            Time (day)
                                                                   300                                                 0           100       200
                                                                                                                                      Time (day)
                                                                                                                                                                                  eral tendency of the I− release is the same as that of Cs+ :
                                                                                                                                                                                  Higher release at higher temperature, no difference between
                          30                                                                                       10
                                                                                                                                                                                  atmospheric and anoxic conditions, and lower release in KCl
                                   (c) Nitrate, 45 oC
                                                                                                                             (d) CsI, 45 C                                        than in deionized water and KOH. However, as seen in Fig.
                          10                                                                                           1                                                          1(d), higher concentrations of I− than of Cs+ were detected
 [NO3] (mmol/L)

                                                                                                                                                                                  in many runs. The reason is unknown, and possible explana-
                                                                                          [I] (mmol/L)

                            3                                                                                      0.1                                                            tions include a possible contamination of the employed chem-
                                                                                                                                                                                  icals (KOH, KCl) by trace amounts of I− , and a possibly un-
                            1                                                                                     0.01
                                                                                                                                                                                  successful exclusion of natural I background in the ICP-MS
                                              atmospheric & anoxic
                                                                                                                                         atmospheric & anoxic                     measurements.
                            0.3           1        3          10     30                               0.001                      0.01        0.1    1      10                        Adsorption of Cs on the bitumen surface is a less likely
                                                                                                                                                                                  cause of the lower concentration in comparison to I− , be-
                                              [Na] (mmol/L)                                                                             [Cs] (mmol/L)

Fig. 1 The release of soluble salt components from non-radioactive,                                                                                                               cause the amount of Cs+ that can be adsorbed is limited by
  synthetic bituminized waste; (a) Na at 45◦ C under atmospheric                                                                                                                  competition with Na+ whose concentration is four orders of
  conditions, (b) Cs at 45◦ C under atmospheric conditions, (c)                                                                                                                   magnitude higher, and because Cs+ is not strongly adsorbed
  NO3 − under atmospheric and anoxic conditions at 25◦ C and 45◦ C,                                                                                                               by bitumen.11–13)
  correlated with Na, (d) I under atmospheric and anoxic conditions                                                                                                               (2) Release of Insoluble Components
  at 25◦ C and 45◦ C, correlated with Cs                                                                                                                                             Barium and strontium are components of the precipitation
     Plotted data in (a) and (b) are the average of duplicate or tripli-
                                                                                                                                                                                  sludge, and the leached amounts were much lower than those
  cate runs, and plot symbols are; q deionized water; s 0.03-mol/l
  KOH; v 0.5-mol/l KCl. The error bars in (c) and (d) include sta-
                                                                                                                                                                                  of Na and Cs. The concentrations of Sr in the leachates, for
  tistical errors associated with duplicate or triplicate runs as well                                                                                                            example, were about one order of magnitude lower than those
  as systematic errors of the analytical determination of elemental                                                                                                               of Cs, although almost the same quantities of these elements
  concentrations.                                                                                                                                                                 were contained in the bituminized waste specimens. The de-

                                                                                                                                                                                       JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY
Leaching Behavior of a Simulated Bituminized Radioactive Waste Form                                                                                                                                                                      231

pendence of the Ba–Sr release on leaching conditions was                                                                                                        The concentrations of SO4 2− must be nearly equal to
consistent with that of the Na–Cs release: Higher release at                                                                                                 the sum of the concentrations of Ba2+ and Sr2+ , i.e.,
higher temperature, suppression of release in KCl leachant,                                                                                                  [SO4 2− ]∼[Ba2+ ]+ [Sr2+ ], neglecting adsorption of these ions
and no effect of redox conditions. Figures 2(a) and (b) show                                                                                                 on the bitumen surface. This relation is almost satisfied, but
typical results for the release of Ba, Sr and SO4 . Figure                                                                                                   the measured [SO4 2− ] was a little higher than [Ba2+ ]+[Sr2+ ]
2(a) shows the concentrations in deionized water under atmo-                                                                                                 as seen in Fig. 2. Trace amounts of sulfur (ca. 4 wt%) con-
spheric conditions at 45◦ C, and Fig. 2(b) the concentrations                                                                                                tained in bitumen may be a cause of the higher concentrations
in 0.5-mol/l KOH solution under anoxic conditions at 45◦ C.                                                                                                  of SO4 2− .
   The concentrations of Sr and SO4 increased with time as                                                                                                   (3) Release of Neptunium and Plutonium
the leaching proceeded, and their transport in the specimens                                                                                                    Neptunium showed a distinct difference in concentration
is considered to be controlled by diffusion like that of Na                                                                                                  depending on the leaching conditions. Figure 3 shows the
and Cs. Unlike the release of Na and Ca, however, the re-                                                                                                    concentrations of Np in the leachates released from the 134 Cs-
lease of Ba, Sr and SO4 does not seem to be affected by                                                                                                      237
                                                                                                                                                                 Np-239 Pu-containing specimens at 45◦ C under atmospheric
swelling of the bituminized waste specimens. The increased                                                                                                   and anoxic conditions. The concentrations of Np in all
concentrations of Ba during the first few tens of days of leach-                                                                                              of the three leachates were below the detection limit (ca.
ing were followed by constant concentrations around a few                                                                                                    2×10−10 mol/l) under anoxic conditions. Higher concentra-
tens of µmol/l, with fluctuations depending on the leaching                                                                                                   tions were detected, on the other hand, in the three leachates
conditions. If about 1% of Ba of the initial weight in the                                                                                                   under atmospheric conditions. This difference obviously re-
bituminized specimens were dissolved in the same way as                                                                                                      flects the difference in the solubility between Np(IV) and
Na and Cs, the concentrations of Ba would be as high as                                                                                                      Np(V), which are the dominant oxidation states under the em-
1 mmol/l. The constant lower concentrations are indicative                                                                                                   ployed atmospheric and anoxic conditions, respectively.
of the solubility-controlled release of Ba.                                                                                                                     Neptunium was added in the form of Np(V) ions to heated
   The precipitation sludge was simulated by (Ba,Sr)SO4 in                                                                                                   bitumen in preparing the bituminized waste specimens. The
this study. This solid dissolves to release Ba2+ , Sr2+ and                                                                                                  chemical form of Np in the organic-rich environment at el-
SO4 2− in the leachate-filled pores of the specimens. The con-                                                                                                evated temperatures is hardly predictable, but is surmised
centrations of Ba2+ in the leachates increased and do not ex-                                                                                                to be Np(IV) oxide [NpO2 (s)], rather than Np(IV) hydrox-
ceed values limited by the solubility of BaSO4 (s). The con-
centrations of Sr2+ also increased, but could eventually reach
constant values limited by the solubility of SrSO4 (s).                                                                                                                                             10

   The equilibrium constant for the reaction (BaSO4 (s)=
Ba2+ +SO4 2− ) is 10−10 at 25◦ C,14) and both of the measured                                                                                                                                         -5                 KOH, DW

concentrations of Ba2+ and SO4 2− were in the same order, as
                                                                                                                                                                                     [Np] (mol/L)

seen in Fig. 2. The values of the measured concentrations                                                                                                                                           10

of Ba2+ of about 10 µmol/l are reasonable in the light of
this thermodynamic interpretation.b The maximum concen-                                                                                                                                             10

tration of Sr2+ is predicted to be in the order of 1 mmol/l,                                                                                                                                                   anoxic

considering the equilibrium constant of 10−6.5 for the reac-                                                                                                                                    10

                                                                                                                                                                                                         0   100        200        300
tion SrSO4 (s)=Sr2+ +SO4 2− . In the leaching tests the con-
                                                                                                                                                                                                             Time (day)
centrations of Sr2+ increased with time as leaching of the
bituminized waste proceeded, but did not reach equilibrium                                                                                                   Fig. 3 The concentrations of Np released from the 134 Cs-237 Np-
values within the leaching periods of about 200 days. The                                                                                                      239
                                                                                                                                                                   Pu-containing specimens at 45◦ C under atmospheric and
concentrations are limited not by the solubility of SrSO4 (s)                                                                                                  anoxic conditions
but by the dissolved amount of (Ba,Sr)SO4 .                                                                                                                        Atmospheric; q deionized water,             0.03-mol/l KOH, s
                                                                                                                                                               0.5 mol/l KCl. The short bars represent the presumed maximum
                                                                                                                                                               concentrations of Np calculated based on the released fractions of
                              100                                                                        100                                                   Na for each of the leachants; “DW”: Deionized water. Anoxic; the
                                 (a) 45 oC, deionized water                                                             o
                                                                                                                   (b) 45 C, KOH             [SO4]             concentrations were below detection limit “DL” (2×10−10 mol/l
 [Ba], [Sr], [SO4] ( mol/L)

                                                                            [Ba], [Sr], [SO4] ( mol/L)

                              80                                                                         80            anoxic
                                                                                                                                                               expressed by the broken line, - - -) in all of the leachants. Plotted
                              60                                                                         60
                                                                                                                                                               data are the average of duplicate or triplicate runs.

                              40                                                                         40
                                                                                                                                                                  Leaching samples are not in thermodynamic equilibrium in
                              20                                                                         20                                                       the leaching tests, because the chemical composition and ionic
                                                               [Ba]                                                                          [Sr]
                                                                                                                                                                  strength of the leachates change with time as the leaching pro-
                               0                                                                          0
                                    0     100         200             300                                      0            100        200             300        gresses. Thermochemical interpretations made here are semi-
                                             Time (day)                                                                       Time (day)                          quantitative using equilibrium constants for the relevant reactions
                                                                                                                                                                  of interest at 25◦ C for infinite dilution (zero ionic strength), rather
Fig. 2 The release of Ba2+ , Sr2+ and SO4 2− from non-radioactive,                                                                                                than a precise analysis including chemical reactions in which all
  synthetic bituminized waste; (a) deionized water at 45◦ C under                                                                                                 of the ions present in the leachates participate, the effect of ionic
  atmospheric conditions, (b) 0.03-mol/l KOH at 45◦ C under anoxic                                                                                                strength on the activity of the ions, and sorption equilibrium. The
  conditions                                                                                                                                                      errors caused by this approximation are possibly as high as of one
     Plotted data are the average of duplicate or triplicate runs.                                                                                                order of magnitude.

VOL. 40, NO. 4, APRIL 2003
232                                                                                                        S. NAKAYAMA et al.

ide [Np(OH)4 (s)] or Np(V) oxide/hydroxide considering the       obtained concentrations of Np (<2×10−10 mol/l) under the
more stable nature of Np(IV) oxide.                              anoxic conditions in the leaching tests strongly suggest that
   The aqueous speciation of Np in the leachates is estimated    the release of Np from bituminized waste is controlled by the
by using thermodynamic data and the measured pH and E H          solubility of NpIV O2 (s) under the anoxic conditions relevant
values. The pH and E H values of the leachates in contact        for deep repositories.
with the tested 134 Cs-237 Np-239 Pu-containing specimens un-       The concentrations observed under atmospheric conditions
der anoxic conditions were: (pH, E H )=(9.1 to 9.9, −100 to      are higher and are controlled by different reactions. The mea-
−300 mV) in deionized water, (pH, E H )=(8.6 to 9.7, −100        sured E H values in the leachates were high enough for Np to
to −350 mV) in 0.5-mol/l KCl, and (pH, E H )=(11.9 to 12.3,      be stable as pentavalent, Np(V). (E H : 400–480 mV in deion-
−100 to −500 mV) in 0.03-mol/l KOH (after 50 days of             ized water; 420–525 mV in 0.5-mol/l KCl; 260–300 mV in
leaching). NpIV (OH)4 (aq) is expected to be the dominant        0.03-mol/l KOH). The dominant aqueous species of Np(V) in
aqueous species of Np under these conditions.15, 16) The max-    these leachates are expected to be NpV O2 + , NpV O2 OH0 (aq)
imum concentration of NpIV (OH)4 (aq) in equilibrium with        and NpV O2 (OH)2 − , depending on the pH of the leachates
NpO2 (s) is given by the solubility of the solid. Yamaguchi15)   (8.5–9.0 in deionized water; 8.5–9.1 in 0.5-mol/l KCl; 12.1–
selected a value of 10−7.7 mol/l, based on Nakayama et al.17)    12.4 in 0.03-mol/l KOH). These species are assumed to be
However, Nakayama et al.17) also reported a general decrease     in equilibrium with NpIV O2 (s), and its overall dissolution re-
in the solubility from 10−7.5 to <10−9 mol/l for prolonged       actions are described by the following equilibrium constants
equilibration periods, as mentioned by OECD NEA.16) The          (log K ) at zero ionic strength:

      pH < 10.7           NpO2 (s) = NpO2 + + e−                             log K = −9.4
                                                                 +   −
      10.7 < pH < 11.7    NpO2 (s) + H2 O = NpO2 (OH) + H + e
                                                                             log K = −20.1
      pH > 11.7           NpO2 (s) + 2H2 O = NpO2 (OH)2 − + 2H+ + e−         log K = −31.8.

   The concentrations calculated with these thermodynamic        spheric and anoxic conditions.15) The dissolution reaction
data and the measured E H and pH values are a few orders of      can be described as PuO2 (am)+H2 O=Pu(OH)4 0 (aq) with
magnitude higher than the experimentally obtained concen-        log K = − 10.1 at zero ionic strength,15) giving concentra-
trations of Np, and the thermochemical interpretations alone     tions of Pu below 10−10 mol/l as determined in the leaching
do not successfully explain the results shown in Fig. 3.         tests. Available data on release of TRU elements from bitu-
   The released fractions of Na from the same bituminized        minized waste is limited. Sneyers and Van Iseghem3) reported
specimens were as high as about 3%, meaning that nearly          Pu and Am concentrations of 10−10 to 10−13 mol/l.
the same volume of the specimens was leached. The release           The elemental releases from the 90 Sr-containing speci-
of Np from bituminized waste under atmospheric conditions        mens leached under anoxic conditions were consistent among
is limited by the progression of leaching, rather than by its    embedded elements, but clearly different from those from
solubility. If 3% of NpO2 (s) embedded in the bituminized        the other two types (non-radioactive and 134 Cs-237 Np-239 Pu-
waste specimens are assumed to be exposed to the perco-          containing) of specimens leached under the same conditions.
lated leachants and dissolved, the concentrations of Np would    The 90 Sr-containing specimens showed much higher release.
be about 1×10−5 mol/l in the deionized water (DW) and            As mentioned above, generation of fine pores occurred in
KOH leachates, and about 1×10−7 mol/l in the KCl leachates.      some of the bituminized waste specimens used for leach-
These values are attainable maximum concentrations of Np,        ing tests under anoxic conditions as a result of air evacua-
indicated in Fig. 3 by short bars.                               tion. Such pores were observed on the air-evacuated 90 Sr-
   The measured concentrations of Np were close to the max-      containing specimens. These specimens showed higher re-
imum values, but did not correspond exactly. The concen-         lease of embedded elements than other specimens, and as a
trations in the deionized water and KOH leachants increased      possible cause for the higher release, the specimens swelled
during the first 100 days of leaching, then decreased by 1/2 to   more strongly than others. The measurement results on el-
1/5 and seemingly became steady. The reason for this varia-      emental release from the 90 Sr-containing specimens leached
tion in concentration is not understood. Removal of colloidal    under anoxic conditions are being excluded from the present
particles, which probably formed from degraded organic sub-      discussion because these specimens were not leached nor-
stances and contained adsorbed Np could decrease Np con-         mally, and it is impossible to interpret the respective results
centration during filtration of the leachates prior to analyti-   for comparison with other experiments.
cal determinations. However, the formation of colloids was
detected only in TOC determinations and no further analysis      2. Leaching Behavior of COGEMA-type Bituminized
was made.                                                           Waste
   The measured concentrations of Pu were below the de-          (1) Surface Observation of Leached Bituminized Waste Sam-
tection limit of the ICP-MS measurement (2×10−10 mol/l)              ples
under all leaching conditions employed. The stable solid            A non-radioactive bituminized waste specimen containing
phase of Pu is estimated to be PuIV O2 (s) and the dom-          NaNO3 (20 wt%) and BaSO4 (20 wt%) was subjected to mi-
inant aqueous species is Pu(IV) (OH)4 0 under both atmo-         croscopic observation of the surfaces. The SEM image of the

                                                                      JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY
Leaching Behavior of a Simulated Bituminized Radioactive Waste Form                                                                                                  233

surface of a fresh (non-leached) specimen is shown in Fig.             elements.
4(a), and results of elemental mapping are given in Figs. 4(b),           A physical degradation typical for bituminized waste is
(c) and (d) for Na, Ba and S, respectively. The majority of            swelling. Hygroscopic materials such as soluble salts in the
Na is present in NaNO3 particles. Scatters of Na around the            waste absorb contacting water and dissolve. The dissolution
NaNO3 particles probably represent Na-containing smaller               of the salt forms pores filled with saline water. These pores
fragments separated from the particles by mixing during sam-           serve as paths for mass transport in the waste form. Since
ple preparation. The density of S shown in Fig. 4(d) looks             pores are larger than the original salt particle in volume, this
higher than that of Ba in Fig. 4(c) because it represents S            leads to an observable swelling of the waste. The swelling ex-
from SO4 ions of BaSO4 plus S contained in the actual bi-              pands the area exposed to surrounding water and accordingly
tumen matrix.                                                          enhances the release of embedded salt components and waste
   This bituminized waste specimen was leached at 60◦ C for            elements.18) The degree and the time of initiation of swelling
30 days in deionized water. The surface exposed to the wa-             vary greatly depending on the physical and chemical proper-
ter swelled and was analyzed by using E-SEM/EDX. Pores                 ties of bituminized waste forms.
of 10 to 100 µm in diameter and finer dispersed white par-                 Dissociation of surface carboxyl groups as a result of inter-
ticles were observed, as seen in Figs. 4(e), (f) and (g). The          actions with water forms the main chemical degradation of a
EDX spectra for the swollen surface showed intense peaks of            bitumen matrix. The degree of the chemical degradation of
carbon, oxygen, S and Ba, and did not show a peak of Na.               bitumen was quantified by determining the amount of total
Soluble NaNO3 was dissolved during leaching and left pores,            organic carbon (TOC) released with the leachates. A series
while fine particles of insoluble BaSO4 remained in the bitu-           of leaching tests was carried out, where non-radioactive bi-
men matrix.                                                            tuminized waste specimens were leached in deionized water
(2) Degradation of Waste Matrix                                        and 0.01-mol/l Ca(OH)2 solutions at 25 and 45◦ C under aer-
   Physical and chemical degradation of the waste matrix,              obic conditions. The amounts of TOC and salt components in
bitumen, is a cause of release of embedded salt and waste              the leachates were determined.
                                                                          Specimens leached in deionized water began swelling at
                                                                       about the 100th day of leaching, and the specimen at 45◦ C
                                                                       swelled remarkably after the 150th day. As shown in Fig.
                                                                       5(a), the release of Na increased accordingly, in particular at
                                                                       45◦ C, and the release of other elements such as Ba, Sr, Cs
                                                                       and I also increased (data not shown). No significant swell-
                                                                       ing was observed on specimens leached with Ca(OH)2 solu-
                                                                       tions, and the release of Na and other elements did not in-
                                                                       crease much. The difference in magnitude of swelling re-
                                                                       sulted from the difference in water uptake between the two
                                                                       leachants.19) The results shown in Fig. 5(a) indicate the in-
                                                                       volvement of swelling in the release of embedded elements.
                                                                       On the other hand, the release of TOC, plotted as a function
                                                                       of time in Fig. 5(b), increased gradually with time with no ap-
                                                                       preciable effect of the swelling. The release of TOC was en-
                                                                       hanced in Ca(OH)2 solutions rather than in deionized water,

                                                                                                  1.2                               0.2
                                                                                                         (a) Na                               (b) TOC
                                                                         Released amount (mmol)




                                                                                                  0.0                               0.0
Fig. 4 E-SEM images of a synthetic bituminized waste specimen;                                       0     50     100 150 200 250         0      50     100   150   200
  (a) NaNO3 salt particles and fine BaSO4 precipitation sludge par-
  ticles, (b) Na mapping, (c) Ba mapping and (d) S mapping on          Fig. 5 Releases of Na and TOC from bituminized waste
  the surface of a fresh (non-leached) specimen. E-SEM images of            The TOC is a dissociation product of the waste matrix (bi-
  the surface of a leached specimen (30 days, 60◦ C, deionized wa-       tumen), and evidences the chemical degradation of bitumen;
  ter); (e) ×500, 560 Pa, (f) ×1,000, 805 Pa and (g) ×2,500, 530 Pa.     q 45◦ C in deionized water,   25◦ C in deionized water, s 45◦ C
  Dispersed fine particles in photographs (e), (f) and (g) are BaSO4      in 0.01 mol/l Ca(OH)2 , 25 C in 0.01 mol/l Ca(OH)2 .

VOL. 40, NO. 4, APRIL 2003
234                                                                                                                                                   S. NAKAYAMA et al.

and at 45◦ C rather than in 25◦ C. The higher release of TOC in
Ca(OH)2 solutions is a result of alkaline attack on the surface

                                                                                     water uptake (g)
carboxyl group. Figures 5(a) and (b) show that the chemical
degradation of bitumen was not affected by swelling, and that                                                   0.4
embedded soluble elements were released incongruently with                                                      0.2
the chemical degradation of the matrix.                                                                         0.0
(3) Effects of Insoluble Precipitation Sludge Particles                                                            0          50      100     150      200
                                                                                                                                   Time (day)
     (BaSO4 )
   Smaller amounts of soluble, hygroscopic salts suppress
swelling of bituminized waste forms and the resulting releases

                                                                                   leached fraction of Cs ( )
of elements.7) Swelling is suppressed also by the presence of                                                   0.4

insoluble salts (CaCO3 ).7)                                                                                     0.3

   The COGEMA-type bituminized waste form of our interest                                                       0.2
contains hygroscopic soluble NaNO3 and insoluble BaSO4 as                                                       0.1
major components. This is one of the differences in chemi-                                                      0.0
cal composition from other bituminized waste forms such as                                                         0          50      100       150    200
                                                                                                                                   Time (day)
Belgian’s “Eurobitum” and Japan’s JNC waste form: “Euro-
bitum” contains NaNO3 (30 wt%) and CaSO4 (8 wt%), and             Fig. 6 Effects of precipitation sludge particles, BaSO4 , on (a)
JNC’ waste form contains only NaNO3 (45–55 wt%).                    swelling quantified by water uptake and (b) release of a repre-
   To investigate the effects of the sparingly soluble compo-       sentative embedded element, Cs
nent, BaSO4 , on the leaching behavior of the COGEMA-type              The compositions of tested specimens are;
bituminized waste form, we examined the swelling and the
                                                                                                                                                  Weight %
elemental releases through a series of leaching tests. Non-           Sample     Symbol
radioactive bituminized waste specimens were leached in                                                                       Bitumen CsNO3 NaNO3 BaSO4
deionized water at 45◦ C for up to 160 days. Five series of           “Blank”                                                      100            0           0     0
samples containing different proportions of NaNO3 , BaSO4             “Cs”                  s                                       95            5           0     0
and CsNO3 were used. The degree of swelling during leach-             “Na(0)”                                                       65            5           0    30
ing was quantified through the amount of water absorbed by             “Na(10)”                                                      65            5          10    20
the specimens. The specimens were taken up from leach-                “Na(20)”              q                                       65            5          20    10
ing containers and weighed after removal of adhering solu-
tion droplets. The difference in weight before and after the
leaching represents the water uptake.                             degree of swelling and Cs release in the BaSO4 -containing
   As seen in Fig. 6(a), specimens containing higher con-         “Na(0)” series.
tents of soluble salt, NaNO3 , showed larger water up-            (4) Effects of Leachants
take, but the difference in the BaSO4 content did not lead        (a) KCl
to a significant difference in water uptake. This is sug-             The 0.5-mol/l KCl leachant was employed to simulate a
gested by a comparison between the results of the BaSO4 -         saline water in the leaching tests. As seen in Figs. 1, 2
containing “Na(0)” series (NaNO3 0 wt%+BaSO4 30 wt%               and 3, the releases of elements from bituminized waste in
+CsNO3 5 wt%) and those of the BaSO4 -free “Cs” series            this leachate were apparently lower than those in the other
(NaNO3 0 wt%+BaSO4 0 wt%+CsNO3 5 wt%). No signif-                 two leachants, deionized water and 0.03-mol/l KOH solu-
icant difference in the degree of swelling is indicative of the   tions. One more difference was that swelling of leached spec-
suppressing effect of BaSO4 on swelling of the COGEMA-            imens did not occur with the 0.5-mol/l KCl. The effects of
type bituminized waste form.                                      this leachant on bituminized materials were observed through
   Specimens containing higher proportions of NaNO3 re-           leaching tests.
leased larger amounts of Cs, as seen in Fig. 6(b). This de-          Non-radioactive bituminized waste specimens were leached
pendence on the salt composition is consistent with the de-       for two weeks at the room temperature in KCl solutions of dif-
pendence on swelling. Swelling is likely to dominate the re-      fering concentrations of 0.05 to 1.0 mol/l. Figure 7 shows the
lease of soluble elements such as Cs from COGEMA-type             surfaces of the leached specimens. The surfaces of the spec-
bituminized waste. The BaSO4 -containing “Na(0)” series re-       imens leached in solutions of lower ionic strength (deionized
leased a slightly larger amount of Cs than the BaSO4 -free        water, 0.05-mol/l KCl and 0.1-mol/l KCl) wrinkled as a re-
“Cs” series. These results imply that BaSO4 is not expected       sult of swelling, but those leached in solutions of higher ionic
to affect the leaching behavior and release of soluble elements   strength (0.5- and 1-mol/l KCl) did not. Lower ionic strengths
from COGEMA-type bituminized waste forms.                         resulted in more remarkable swelling. Brodersen et al.8) re-
   The BaSO4 -containing “Na(0)” series specimens contain a       ported a similar observation, in which water uptake of bitu-
total of 35 wt% of salt, and the BaSO4 -free “Cs” series speci-   men was suppressed in NaNO3 and NaCl solutions.
mens contain 5 wt%. The hygroscopic soluble CsNO3 is cov-            A few reasons are inferred regarding the effects of ionic
ered by a larger amount of bitumen in the “Cs” series than in     strength. Firstly, the high ionic-strength leachants are so vis-
the “Na(0)” series . This difference in coverage of salts by      cous that they penetrate into specimens very slowly, which
bitumen is thought to have caused the small increase in the       leads to retarded dissolution of soluble salts and retarded

                                                                       JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY
Leaching Behavior of a Simulated Bituminized Radioactive Waste Form                                                                                                                                                                                     235

        Fig. 7 Bituminized waste leached in deionized water and different KCl solutions for two weeks at room temperature
              Leachant (from left to right); deionized water, 0.05-mol/l KCl, 0.1-mol/l KCl, 0.5-mol/l KCl, 1.0-mol/l KCl.

swelling of the specimens. Secondly, because dissolution of         of the leachant, D the diffusivity of the component in the bi-
soluble salts is driven by the difference in chemical poten-        tuminized waste form and t the time period of leaching. C 0 ,
tial at the interface of leachants and the surface of the salt      S, V and t are given by the experimental conditions, and C
particles, and because the concentrated leachant brings about       is measured. The D value calculated from Eq. (1) enables to
a smaller difference in chemical potential, the dissolution of      estimate the time duration during which a certain fraction of
the salts is delayed. Thirdly, because diffusion of ions in the     the component of interest is confined in the waste form. This
aqueous phase occurs according to the difference in chemical        value has often been regarded as a kind of “lifetime” of the
potential between adjacent media, and because the concen-           waste form.
trated leachant has a high chemical potential, diffusion of ions       It should be noted that the application of Eq. (1) to predict
in bituminized waste specimens out of the surfaces is slow.         the lifetime of waste forms assumes no physical transforma-
   High ionic-strength media like saline water are expected         tion of the diffusion medium, i.e., of the waste forms. This
to suppress swelling of bituminized waste forms, and accord-        is not always the case for bituminized waste due to its plastic
ingly to suppress the release of embedded components.               nature. A bituminized waste form will swell due to the inter-
(b) Ca(OH)2                                                         nal pressure, and shrink due to external pressures in the deep
   As shown in Fig. 5(b), the amount of TOC released was            underground repositories. The surface-to-volume ratio, S/V ,
larger in Ca(OH)2 solutions than in deionized water. This           gradually varies with time as a result of deformation. A bi-
result differs from that for the release of Na shown in Fig.        tuminized waste form cannot be expected to keep its original
5(a), where the release of salt components under atmospheric        configuration for a very long time. The D value is, however,
conditions was suppressed in Ca(OH)2 solutions. Broder-             available as a numerical index for one of the physical prop-
sen et al.8) and Sazarashi et al.19) have also reported this        erties of waste forms, and allows to make a semi-quantitative
suppression. It probably resulted from physical coverage of         comparison of performance for differing types of bituminized
the surface of the bituminized waste specimen by a precipi-         waste forms.
tate of CaCO3 (s) that was formed in the presence of CO3 2− -          We applied Fick’s law to the present results on the release
containing leachants. Such a suppression was not observed           of Na, NO3 and Sr for approximate estimates of the diffusiv-
for the specimens leached under the anoxic Ar-gas atmo-             ity, D. Figure 8 shows the released fractions of Na, NO3 and
sphere where CO2 gas was also removed from leachants. This          Sr plotted as a function of the square root of leaching time.
effect of CaCO3 (s) coverage is expected in waste repositories      The release curves consist of two segments, and their linear
where CO2 gas and Ca are supplied.

                                                                                              4                                                                                           0.15
3. Diffusivity of Ions in the Waste Form as Measure for                                               (a) Na, NO3                                                                                    (b) Sr, 45 oC
   Performance                                                                                                  25ºC 45ºC                                                                               Deionized water
                                                                                                                                                                  Released fraction ( )
                                                                      Released fraction ( )

                                                                                              3                                                                                                         KOH
   The mass transport in bituminized waste forms has been in-                                         Na

terpreted through diffusion-controlled process.20) On the as-                                 2
                                                                                                                                                                                                                     -15    2
                                                                                                                                                                                                                           m /s

sumption that the mass transport can be approximately de-                                                   D =2x10
                                                                                                                               -12    2
                                                                                                                                     m /s
                                                                                                                                                                                          0.05            -16    2                           -16    2
scribed by Fick’s law, the cumulative released fraction of an                                 1
                                                                                                                                                  -13    2
                                                                                                                                                        m /s                                          1x10      m /s               8x10            m /s

embedded component is linearly related to the square root of                                          8x10
                                                                                                                      m /s
                                                                                                                                                                                                                            D = 2x10
                                                                                                                                                                                                                                       -16    2
                                                                                                                                                                                                                                             m /s
time as                                                                                       0                                                                                           0.00
                                                                                                  0                   5            10        15              20                                  0           5               10          15               20
       C     2S Dt                                                                                                            Time (day)

          =          ,                                (1)
      C0     V    π                                                 Fig. 8 Diffusivity of (a) Na+ and NO3 − and (b) Sr2+ released from
where C is the released amount of the component, C 0 the              bituminized waste into deionized water
initial content of the component in the bituminized waste                The tested specimens swelled after the 100-th day at 25◦ C and
form, S the surface area of the waste form, V the volume              56-th day at 45◦ C.

VOL. 40, NO. 4, APRIL 2003
236                                                                                                                             S. NAKAYAMA et al.

increase against the square root of time suggests diffusion-                 ter 100 days of leaching and enhanced diffusivity of Sr, as
controlled transport of ions. The tested bituminized speci-                  shown in Fig. 8(b). The diffusivity of Sr was smaller than that
mens swelled after the 100th day at 25◦ C and after the 56th                 of Na and NO3 . The difference in diffusivity between Na and
day at 45◦ C. The first segments in Fig. 8 represent diffu-                   Sr is too large to be attributed to the difference in diffusivity
sional transport of the ions in bituminized specimens before                 of these ions in free water. The limited number of adsorp-
swelling, and the second segments represent diffusional trans-               tion sites on the bitumen surface might retard diffusion of Sr,
port in the swelled layer of the specimens. Swelling is accom-               whose concentrations were a few orders of magnitude lower
panied by an increase in porosity and a decease in tortuosity                than those of Na. The D values obtained in this study are
of the diffusion paths. These changes in physical configura-                  listed in Table 2 together with published values.
tion of diffusion paths enhance diffusivity after swelling.                     Diffusivity values obtained after swelling of the bitu-
   The released fractions of Na+ and NO3 − were very similar                 minized waste forms employed in this study are larger than
over the leaching periods at both temperatures. D values cal-                literature values. The “Eurobitum R85/40,” which is based
culated from Eq. (1) are not constant with varying S/V ratios                on a soft bitumen like our specimens, is reported to show
during progression of the leaching, and are not anticipated to               remarkable swelling,3) but the diffusivity of Na+ and NO3 −
be distinguishably precise for Na+ and NO3 − . Almost identi-                is lower than our values. The comparison indicates that the
cal D values were hence obtained for both of these ions. D at                “COGEMA”-type bituminized waste form has a somewhat
45◦ C before swelling was given by the slope of the first seg-                lower resistance with regard to the release of embedded com-
ments to be about 8×10−15 m2 /s. The fractions released be-                  ponents than other types of bituminized waste forms.
fore swelling were so small at 25◦ C that a meaningful D value                  The long-term release of embedded components from bi-
was not obtained, but D is certainly lower than 8×10−15 m2 /s.               tuminized waste forms is the result of a combined effect of
After swelling, D values increased to about 3×10−13 m2 /s at                 hygroscopicity of embedded salts, dissolution of the salts,
25◦ C and about 2×10−12 m2 /s at 45◦ C. The higher D at 45◦ C                formation of solution-filled pores and pore-connecting paths,
resulted from more remarkable swelling at the temperature                    and diffusion of dissolved ions, under the swelling pressure
than at 25◦ C, rather than the increase in diffusivity of the ions           and external pressures caused by disposal conditions. A sim-
in water at the temperature range.                                           ple diffusional transport in intact media as described above
   Swelling of the bituminized waste specimens occurred af-                  is not able to describe the release. Brodersen has success-

                      Table 2 Diffusivity D of some ions obtained from leaching tests of bituminized waste specimens

                                                      Leaching test
  Bituminized waste     Bitumen                                                                       D (m2 /s)                       Reference
                                       Temperature    Period          Leachant

                                       Room tem-      15         Deionized water,     Na+ : (2–3)×10−14      (swelling not         Yamaguchi et al.,
  “Kansai Electric”   (not reported)    perature,    months       mortar water                                 reported)              199922)
                                        60, 90◦ C

                                                                                      Na+ : 1.4×10−13
                                                                                      NO3 − : 8.4×10−14
                                                                Clay water                                   (after swelling)
                                          23◦ C                                       Ca2+ : 6.9×10−15
                                                                                      SO4 2− : 9.8×10−16

                                                                                      Na+ : 1.5×10−13                                Sneyers and
                                                                Clay-cement water                          (after swelling)
                                                                                      NO3 − : 8.9×10−14                              Van Iseghem,
  Eurobitum R85/40        Soft                       720 days
                                                                                      Na+ : 3.4×10−13                                   19983)
                                                                                      NO3 − : 1.7×10−13
                                                                Clay water                                   (after swelling)
                                                                                      Ca2+ : 2.2×10−14
                                          40◦ C
                                                                                      SO4 2− : 1.2×10−14

                                                                                      Na+ : 1.1×10−13
                                                                Clay-cement water                          (after swelling)
                                                                                      NO3 − : 6.7×10−14

                                                                                      Na+ , NO3 − : <8×10−15 (before swelling)
                                          25◦ C
                                                                                      Na+ , NO3 − : 3×10−13 (after swelling)

                                                                                      Na+ , NO3 − : 8×10−15 (before swelling)
                                                                Deionized water
                                                                                      Na+ , NO3 − : 2×10−12 (after swelling)
      “COGEMA”            Soft                       224 days                                                                         This study
                                                                                      Sr2+ : 2×10−16 (before swelling)
                                          45◦ C
                                                                                      Sr2+ : 2×10−15 (after swelling)

                                                                                      Sr2+ : 1×10−16 (before swelling)
                                                                0.03-mol/l KOH
                                                                                      Sr2+ : 8×10−16 (after swelling)

                                                                                    JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY
Leaching Behavior of a Simulated Bituminized Radioactive Waste Form                                                                     237

fully modeled hygroscopicity and swelling properties of bi-                 Management XXI (Mat. Res. Soc. Proc. 506), I. G. McKinley,
                                                                            C. McCombie, ed., Material Research Society, Pennsylvania,
tuminized waste forms, and gave qualitative explanations on                 565–572 (1998).
the release rate of embedded salts.21)                                 4)   S. Camaro, B. Simondi-Teisseire, P. P. Vistoli, “Long-term be-
                                                                            havior of bituminized waste,” Proc. Int. Workshop on the Safety
IV. Conclusions                                                             and Performance Evaluation of Bituminization Processes for
                                                                            Radioactive Waste (Radwaste Bituminization ’99), Nuclear Re-
   Laboratory-scale leaching tests were performed for bitu-                 search Institute Rez, Czech Republic, June 29–July 2, 1999,
                                                                            p. 157–160 (1999).
minized waste simulating that of a French reprocessing com-            5)   JIS, Japanese Industrial Standards K 2207:1996 Petroleum As-
pany, COGEMA. The specimens were leached under the at-                      phalts, Japanese Standards Association, (1996).
mospheric and anoxic conditions in deionized water, which              6)   JIS, Japanese Industrial Standards K 2265:1996 Crude Oil and
                                                                            Petroleum Products — Determination of flash point, Japanese
represents fresh type groundwater, an alkaline solution (0.03-              Standards Association, (1996).
mol/l KOH) representing cement-contacting, highly alkaline             7)   IAEA, Bituminization of Radioactive Wastes, Technical Report
groundwater, or a saline solution (0.5-mol/l KCl) simulating                Series No. 116, International Atomic Energy Agency, (1970).
                                                                       8)   K. Brodersen, B. M. Pedersen, A. Vinther, Comparative Study
a saline environment.                                                       of Test Methods for Bituminized and Other Low- and Medium-
   Chemical compositions of leachants influenced swelling                    level Solidified Waste Materials, RISO-M-2415, Riso National
of the bituminized waste. Swelling progressed in deionized                  Laboratory, (1983).
water and the alkaline solution, and enhanced the release              9)   M. Sazarashi, Y. Ikeda, Y. Takashima, T. Mine, “Leaching
                                                                            behavior of soluble elements from bituminized wastes (II),”
of the soluble components, Na and Cs, from the specimens.                   Preprints 1998 Fall Mtg., At. Energy Soc. Jpn., M42, 859
Their release was considered to be diffusion-controlled in the              (1998), [in Japanese].
swelled layers of the specimens. The release of sparingly sol-        10)   L. R. Van Loon, Z. Kopajtic, “The adsorption of radionuclides
                                                                            on bitumen — Part I: Strontium,” Radiochim. Acta, 55, 83–89
uble components such as Ba and Np was solubility-limited                    (1991).
in addition to the progression of leaching, and was not en-           11)   S. G. Burney, Sorption Effects in the Leaching of 137 Cs from Or-
hanced by swelling. The elemental release from the bitu-                    ganic Matrix Forms, AERE R 12421 DOE/RW/87.083, (1987).
                                                                      12)   R. Hietanen, M. Alaluusua, T. Jaakkola, Sorption of Cesium,
minized waste leached in the 0.5-mol/l KCl was very low,                    Strontium, Iodine, Nickel and Carbon in Mixtures of Concrete,
which is likely due to the suppression of swelling of the sam-              Crushed Rock and Bitumen, Report YJT-85-38, (1985).
ples at the high ionic strength.                                      13)   M. Alaluusua, M. Hakanen, Sorption of Cesium, Strontium,
                                                                            Cobalt and Technetium in Mixtures of Concrete, Crushed Rock
   The surrounding redox conditions influenced leaching be-                  and Bitumen, Report YJT-87-10, (1987).
havior of redox-sensitive elements. Neptunium, a redox-               14)   A. Bard, R. Parsons, J. Jordan, Standard Potentials in Aqueous
sensitive element, showed a distinct difference in release be-              Solution, Mercel Dekker, New York, (1983).
                                                                      15)   T. Yamaguchi, Consideration on Thermodynamic Data for
tween anoxic and atmospheric conditions.                                    Predicting Solubility and Chemical Species of Elements in
                                                                            Groundwater, 2. Pu, Np, JAERI-Data/Code 2000-031, Japan
Acknowledgment                                                              Atomic Energy Research Institute, (2000), [in Japanese].
                                                                      16)   J. Lemire, J. Fuger, H. Nitsche, P. Potter, M. H. Rand, J.
   The authors acknowledge N. Yanase, N. Ito and J.                         Rydberg, K. Spahiu, J. C. Sullivan, W. J. Ullman, P. Vitorge,
                                                                            H. Wanner, Chemical Thermodynamics 4: Chemical Thermo-
Kuwabara (Japan Atomic Energy Research Institute: JAERI)                    dynamics of Neptunium and Plutonium, (Ed.) OECD Nuclear
for their instructions and assistance in experimental proce-                Energy Agency, Elsevier, (2001).
dure, T. Yamaguchi (JAERI), D. Ochem and B. Simondi-                  17)   S. Nakayama, T. Yamaguchi, K. Sekine, “Solubility of nep-
                                                                            tunium(IV) hydrous oxide in aqueous solutions,” Radiochim.
Teisseire (CEA, France) for valuable discussions, and M.                    Acta, 74, 15–19 (1996).
Ochs (BMG Engineering Ltd., Switzerland) for his stylis-              18)   IAEA, Bituminization Processes to Condition Radioactive
tic revisions. The experiments were carried out at WASTEF                   Wastes, Technical Reports Series No. 352, International
                                                                            Atomic Energy Agency, (1993).
“Cold” laboratories and NUCEF of JAERI. This study was                19)   M. Sazarashi, Y. Ikeda, M. Kumagai, Y. Takeshita, “Leach-
conducted under contract with the former Science and Tech-                  ing behavior of soluble elements from bituminized wastes,”
nology Agency (now Ministry of Education, Culture, Sports,                  Preprints 1997 Fall Mtg., At. Energy Soc. Jpn., I21, 663 (1997),
                                                                            [in Japanese].
Science and Technology) of Japan based on the Special Ac-             20)   J. L. Means, L. A. Smith, K. W. Nehring, S. E. Brauning, A.
count Law for Electric Power Development Promotion.                         R. Gavaskar, B. M. Sass, C. C. Wiles, C. I. Mashni, The Appli-
                                                                            cation of Solidification/Stabilization to Waste Materials, Lewis
                                                                            Publ., Boca Raton, p. 195 (1993).
References                                                            21)   K. Brodersen, “Hygroscopic materials in bituminized waste.
                                                                            experiments and modelling,” Proc. Int. Workshop on Safety and
 1) TRU Coordination Office, Progress Report on Disposal Con-                Performance Evaluation of Bituminization Processes for Ra-
    cept for TRU Waste in Japan, JNC TY1400 2000-002, TRU                   dioactive Waste (Radwaste Bituminization ’99), Nuclear Re-
    TR-2000-02, Japan Nuclear Cycle Development Institute and               search Institute Rez, Czech Republic, June 29–July 2, 1999,
    The Federation of Electric Power Companies, (2000).                     p. 149–152 (1999).
 2) J. C. Alder, D. F. McGinnes, Model Radioactive Waste Inven-       22)   T. Yamaguchi, T. Amaya, A. Mukunoki, M. Fujiso, “Leaching
    tory for Swiss Waste Disposal Projects, Volume 2: Database,             test of bituminized wastes under geological conditions,” Proc.
    NAGRA Technical Report 93-21, National Cooperative for the              Int. Workshop on the Safety and Performance Evaluation of Bi-
    Disposal Radioactive Waste, (1994).                                     tuminization Processes for Radioactive Waste (Radwaste Bitu-
 3) A. Sneyers, P. Van Iseghem, “The leaching behavior of bitu-             minization ’99), Nuclear Research Institute Rez, Czech Repub-
    minized radioactive waste in the geologic disposal conditions           lic, June 29–July 2, 1999, p. 161–164 (1999).
    of the boom clay formation,” Scientific Basis for Nuclear Waste

VOL. 40, NO. 4, APRIL 2003

Shared By: