«THERMODYNAMIC MODELING OF METAL ADSORPTION AND MINERAL SOLUBILITY IN GEOCHEMICAL SYSTEMS A Dissertation Submitted to the Graduate School of the ...»
Cation substitution, or solid-solution, in a crystal structure can either enhance or decrease the solubility of the compound. For example, the incorporation of Mg2+ into the calcite structure causes an increase in its solubility (Davis et al., 2000), but the incorporation of trace amounts of La3+ results in crystal growth inhibition and a decrease in mineral solubility (Kamiya et al., 2004). Sass and Rai (1987) co-precipitated amorphous Cr3+ and Fe3+ hydroxides, and found these co-precipitates to behave thermodynamically as ideal solid solutions. The closest analog to our study is that of Rai et al. (2004), who determined the effect of Np4+ incorporation into uraninite (UO2) across the entire range of solid solution. The authors observed ideal solid solution behavior, or a decrease in the aqueous U4+ concentration in equilibrium with the solid phase that is equal to the decrease in the mole fraction of U4+ within the solid phase with increasing extents of Np4+ substitution. In the cases of Np(V) and U(VI), the approximately linear dioxo cations Np(V)02+ and U(VI)02+ dominate both solution and crystal chemistries. The Np(V)02+ neptunyl ion is geometrically compatible with U(VI)02 sites in a crystal (Bums, 2005; Forbes and Burns, 2008), but a chargebalancing co-substitution is needed.
5.2 Materials and Methods Soddyite was synthesized using a mild hydrothermal method similar to that reported by Gorman-Lewis et al. (2007) and Klingensmith and Burns (2007). A Np(V) stock solution was prepared from 500 mg of Np02 powder was purchased from Oak Ridge National Laboratory. Approximately 25 mg of the Np02 powder was placed in a •7-mL Teflon cup with a screw top lid. Three mL of concentrated HNO3 was added; the cup was tightly sealed and then placed in a 125-mL Teflon-lined Parr acid digestion vessel. Thirty-five mL of ultrapure water was added to the vessel to provide counterpressure during the heating cycle. The vessel was heated at 150 °C in a Fisher Isotemp oven for 48 hours. After the heating cycle, no NpC2 powder was observed and the solution was a dark brownish green color. A UV spectrum of the solution indicated that the neptunium was in both a pentavalent and hexavalent state by the presence of peaks at 980 cm"1 and 1223 cm"1, respectively. No peaks associated with tetravalent Np were present in the spectrum. A small amount of NaNCh was added to the solution, which resulted in the reduction of Np(VI) into Np(V) as indicated by a color change of the solution from a brownish green to a bright emerald green. The pentavalent Np was precipitated into a relatively insoluble Np hydroxide precipitate using a small amount of a saturated NaOH solution and washed three times with ultrapure water to remove the excess sodium from solution. The precipitate was then re-dissolved in the appropriate amount of 1 M HNO3 to create an approximately 1000 ppm stock solution. A UV spectrum of the final stock solution confirmed that no Np(IV) or Np(VI) was present.
The actual concentration of the stock solution was measured using LA-ICP-MS. Na+ can be incorporated into the Np hydroxide compound and released into solution with the re-dissolution of the precipitate; Therefore, a small amount of Na+ is usually present in the final Np(V) stock solution.
An appropriate aliquot of the neptunyl stock solution (1270 ppm Np(V) in 1.0 M HNO3), was added to each synthesis to achieve the following initial aqueous Np concentrations: 12.7 ppm, 127 ppm, 381 ppm, and 634 ppm. After heating, the resulting precipitate from each synthesis was washed four times with boiling ultrapure 18 MCI water to remove adsorbed Np and/or adhered colloidal material. After the powders were washed with boiling water, a small amount of ultrapure water was added to each product to create a slurry mount on a zero-background quartz plate. The powder was air-dried overnight and then covered with a small piece of Kapton® tape to prevent contamination. A X-ray diffraction pattern was collected on a Scintag powder diffractometer from 10-90° 29 with a step size of 0.02 °/sec and scan speed of 10 seconds. The powder pattern matched those of PDF-00-035-0733 and PDF-01-079-1323 for synthetic soddyite. Fourier transform infrared (FTIR) spectroscopy on the dried powders confirmed that no amorphous phases were present. The concentration of Np(V) incorporated into each synthesized solid phase was calculated by the difference between the known initial Np aqueous concentration in the synthesis solution and the Np concentration that remained in the combined synthesis and wash solutions after synthesis. The soddyite phases reported here have 24, 919, 2730, and 6511 ppm Np incorporated into the crystal structure (Table 1).
We used transmission electron microscopy with electron diffraction spectroscopy (TEM-EDS) to obtain images and the chemical composition of individual soddyite crystals from each of the solid phases synthesized in this study. We observed a constant chemical composition for all crystals examined for each Np-incorporated solid phase, and the crystal cores exhibited the same composition as the rims, suggesting that a single phase was present in each sample. TEM electron diffraction patterns from single crystals of each phase indicated that soddyite was the only phase present in each synthesized powder sample.
To begin a solubility experiment, approximately 125 mg of a soddyite powder, 100 mg of silica gel (to buffer aqueous Si concentrations), and 7 ml of 18 MQ ultrapure water were placed in a teflon-coated centrifuge tube. In order for the systems to reach equilibrium more quickly, the starting solutions were spiked with 10"45 M U022+ and 10"3 5 M Si, each of which is below the expected equilibrium concentrations for these species. Experiments were conducted at pH 3.4, and the pH was adjusted using small volumes of concentrated HNO3 until the pH did not vary. Between sampling intervals, experimental tubes were slowly agitated on a rotary shaker. Aliquots of 400 \xL of the experimental solutions were extracted periodically over a period of 24 days. To extract a
sample, the Teflon tubes were centrifuged at 20,000 g for 2.5 minutes. An aliquot of the resulting supernatant was removed, filtered through a 0.20 \im nylon filter, and stored for dilution. Following dilution, these aliquots were analyzed for total aqueous Np, U, and Si concentrations using inductively coupled plasma optical emission spectroscopy (ICP-OES) for U and Si, and inductively coupled plasma mass spectroscopy (ICP-MS) for Np. Samples analyzed for Np were internally standardized with 1 ppb Tl and Bi.
Repeat analyses of element internal standards indicated that instrumental uncertainty was ±3.6%. X-ray diffraction analyses of the solid phase after each experiment did not reveal a change in crystallinity, and confirmed that soddyite was the only crystalline phase present in each solubility experiment.
5.3 Results and Discussion The concentrations of U, Si, and Np increased from undersaturation to reach steady-state within 10 days. As expected, the steady-state aqueous Np concentration increases as a function of the Np concentration in the solid phase (Figure 5.1 A). The steady-state aqueous Si concentrations for each experiment are, within analytical uncertainty, independent of the Np concentration in the solid phase used in each experiment (Figure 5.IB). The average steady-state Si concentration value for all experiments is 10"266 molal, suggesting that Si was successfully buffered in solution by the amorphous silica gel, which has reported solubility values ranging from 10"2 38 to 10"271 molal (Morey et al., 1964; Walther and Helgeson, 1977). However, the steadystate aqueous U concentrations decrease appreciably as the concentration of Np incorporated in the solid phase increases (Figure 5.IB). We average the measured pH and U, Si, and Np concentrations from the final five sampling periods (between days 14
Figure 5.1: Measured aqueous concentrations of Np (A) and U and Si (B) during solubility experiments at pH 3.
4. Solid phase Np concentrations are: 24 ppm ( • ), 919 ppm (A), 2730 ppm ( • ), and 6511 ppm ( • ).
and 22 of the solubility experiments) for use in the thermodynamic calculations. The measured pH and element concentrations for each experiment are given in Table 2.
Using the steady-state measured aqueous Np concentrations and the known Np content of each solid phase, distribution coefficients (Kd) were calculated for each phase
where m NP, solution represents the aqueous phase Np molality, and m NP, solid represents the solid phase concentration of Np in units of mol kg"1. Log Kd values from experiments involving the 24, 919, 2730, and 6511 ppm Np solid phases are -3.70, -4.46, -4.16, and
-4.22, respectively. The high value for the 24 ppm Np solid phase experiment likely has the highest experimental uncertainty; the aqueous Np concentrations in this experiment were near the detection limit for the ICP-MS approach. The other Log Kd values are relatively close to one another, and do not vary systematically as a function of Np content in the solid phase. The apparent constancy of the Kd values suggests that aqueous Np concentrations in equilibrium with solid phases with higher Np contents can be estimated using these same Kd values.
The effect of Np incorporation on the solubility of the solid phase in terms of the release of U into solution is much larger than would be predicted assuming ideal solidsolution behavior for the neptunyl substitution for uranyl. Although the exact structural mechanisms of incorporation have not been determined, because of charge balance considerations, substitution of Np(V)02+ for U(VI)022+ in the soddyite structure is most likely accompanied by concomitant substitution of Na+ into a vacant cavity ') •
where x is the number of uranyl sites in the solid phase occupied by the uranyl cation, and for pure soddyite, x = 2. The equilibrium constant for reaction (2) is termed the
solubility product, and can be expressed as:
where a represents the thermodynamic activity of the subscripted aqueous species or solid phase. The thermodynamic standard state for H2O is the pure phase at the pressure and temperature of interest. The standard state for aqueous species is a hypothetical one molal solution at the pressure and temperature of interest that behaves as if it were
infinitely dilute, and activities are related to molalities by:
represents the charge of the ion of interest, and / is the solution ionic strength.
In each experiment with a different Np-incorporated soddyite, we measured the total concentrations of U, Np, and Si, and under the experimental conditions, these concentrations are equal to the concentrations of UO2, NpC2, and Si02(aq). We measured pH for each sample, and from the known concentration of base addition to each system, we can calculate the Na+ molality. According to equation (3), concentration values for each aqueous species, in conjunction with a value for Ksp, yield a calculated value for the activity of the particular Np-incorporated soddyite used in each experiment.
In this study, in order to isolate the effect of Np substitution, we use the 24 ppm Np soddyite solubility measurement to calculate an internally consistent baseline value for Ksp of 10 6 56. This value is slightly higher than, but within experimental uncertainty of, the value calculated by Gorman-Lewis et al. (2007).
Because we know the mole fraction of Np in each solid that was studied, we can
directly calculate the activity coefficient of each solid phase:
where ^soddyite is the mole fraction of uranyl soddyite, and Ysoddyue is the activity coefficient of the solid. Ideal solid-solution behavior is characterized by a Tsoddytte value of 1. Equation 3 is insensitive to the activities of NpC2+ and Na+ when values of x are close to 2, as in these experiments. Therefore, at fixed pH and buffered Si activities, and with increasing Np incorporation in the soddyite, the solid phase would demonstrate ideal solid-solution behavior if the activity of U022+ decreased by the same amount as Soddyite- This behavior was observed for Np+4 incorporation into UO2 by Rai et al.