A series of solutions containing 0.1-10 mM Pu(IV) in the Simple Feed matrix was subjected to vis-NIR measurements. The spectral overlay obtained is shown in Fig. 4.4 (left). It was observed that Pu(IV) spectral features at 500-1000 nm were not obstructed by spectroscopic features of

4.5 Vis-NIR spectra of variable Pu(IV) in TBP/dodecane extraction phase (left) and corresponding calibration plots for Pu(IV) (right). Detection limit is 0.014 mM for Pu(IV) using the 640 nm band.

UO2(NO3)2 or HNO3, and linear calibration plots were obtained using four characteristic Pu(IV) bands in this region (Fig. 4.4 right). The detection limit for Pu(IV) was determined to be 0.08 mM using the 659 nm band.

The distribution of Pu(IV) from the aqueous phase into the organic phase can be followed using vis-NIR spectroscopy of both the aqueous and organic phases. The characteristic Pu(IV) bands in the TBP/n-dodecane solution were slightly different than those in the aqueous nitrate solution, which reflects expected changes in the inner coordination environment of the Pu(IV) ion upon transport into the organic solvent.

To quantitatively determine Pu(IV) in the extraction phase, a series of solutions containing 0.2-2 mM Pu(IV) in 30% TBP/n-dodecane was mea­sured by vis-NIR spectroscopy. The resulting spectral overlay is shown in Fig. 4.5 (left). The linear standard curves of the resulting Pu(IV) data are displayed in Fig. 4.5 (right). The detection limit for Pu(IV) in 30% TBP/n- dodecane was determined to be 0.014 mM using the 640 nm band.

Neptunium is present in the dissolved spent fuel predominantly as Np(V, VI) in the NpO2+ and NpO22+ chemical forms (Madic 1984). However, its redox equilibrium Np(IV) ^ Np(V) ^ Np(VI) highly depends on the multiple factors including solution composition, temperature, etc. (Guillaume 1981). This redox chemistry determines Np distribution into 30% TBP/n — dodecane phase, and therefore, its spectroscopic monitoring is desirable. The aqueous speciation of Np(V) is complex because of its coordination with nitrate anion and formation of cation-cation complexes in accord with reactions 4.2 and 4.3 (Colston 2001):

NpO2+ + n NO3- ^ [NpO2+ ■ n NO3-] 4.2

NpO2+ + Mn+ ^ [NpO2+ ■ Mn+] 4.3

where Mn+ is transition or f-metal cation. In the dissolved spent fuel, the complex species NpO2+ ■ UO22+ formed as described by reaction 4.3 are the most prominent.

The single NpO2+ band observed in 1 M HNO3 solution splits into two bands in the presence of U(VI). A second neptunium band nm emerges in the presence of U(VI) at the concentration ranges typical to the dissolved spent fuel streams (Steele 2007). As a result, Np(V) spectroscopic proper­ties are highly mobile and dependent on solution composition. To this end, understanding and quantification of Np(V) chemistry is needed to correctly interpret its vis-NIR spectra.

To investigate the spectral nature of Np in the UREX process, a series of solutions of variable Np(V) concentration in 1.33 M UO2(NO3)2 and 0.8 M HNO3 were prepared; the vis-NIR spectra are shown in Fig. 4.6 (left).

0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0

0 0.2 0.4 0.6 0.8 1 1.2 1.4

U concentration, M

4.7 Spectral layout of 0.3 mM NpO2+ titrated with variable UO2(NO3)2 solution, in 2.75 M HNO3 (left), and corresponding relative concentrations of the "free" and "uranyl-bound" NpO2+ species in solution (right).

Two bands observed for the Np(V) in the UREX feed solution located at 981 and 992 nm in agreement with previous reports (Steele 2007). Figure 4.6 (right) shows the standard curves for the Np(V) vis-NIR data in the Simple Feed solution (Fig. 4.6, left). To illustrate the interaction of the [NpO2+ • UO22+] cation-cation complex, a spectrophotometric titration was carried out in which uranyl nitrate was titrated into a constant concentra­tion of Np(V). Figure 4.7 (left) shows the spectral overlay of the resulting vis-NIR spectra, revealing the gradual depletion of the “free” NpO2+ ion and the gradual increase of the “uranyl-bound” [NpO2+ • UO22+] complex. The speciation diagram for the relative concentrations of the “free” and “bound” Np species are shown in Fig. 4.7 (right).

To evaluate the feasibility of Np(V) quantification in solutions containing significant concentrations of Pu(IV), a series of feed simulant solutions containing 0.1 mM Np(V) and a variable concentration of Pu(IV) (0.1-10 mM) in 1.33 M UO2(NO3)2 and 0.8 M HNO3 matrix was measured using vis-NIR (Fig. 4.8). The spectral overlay shown in this figure illustrates that the Np(V) is detected in the presence of large excesses of Pu(IV) and U(VI); Np was 100 times less concentrated than Pu, and 13 000 times less concentrated than U in solution (Levitskaia 2008). A conservative measure of the detection limit of Np(V) can be established at < 0.1 mM under UREX flowsheet conditions. An initial distribution experiment using a UREX feed containing Np(V), 10 mM Pu(IV) and 1.3 M UO2(NO3)2 in 0.8 M HNO3 showed no detectible Np(V) in the loaded 30% TBP/n — dodecane solvent by UV-vis-NIR as would be expected for the known low extractability of Np(V).

Initial chemometric analysis of vis-NIR spectral data was undertaken using separate Simple Feed solutions containing variable Np(V), U(VI),

4.8 Absorption spectra of 0.1 mM Np(V) at variable 0.1-10 mM Pu(IV) in Simple Feed (1.3 M UO2(NO3)2 in 0.8 M HNO3).

and Pu(IV). PLS analysis of vis-NIR spectral data (containing variable UO22+/Np(V)/nitric acid concentrations) yielded linear response over 0-0.5 mM range for Np(V). A PLS model of the vis-NIR spectral region for Pu(IV) in UREX feed solution (Pu(IV) in 1.3 M UO2(NO2)2 in variable HNO3) showed a linear response over the 0-10 mM range for Pu. Figure

4.9 contains the results of the PLS analysis for Np(V) (left) and Pu(IV) (right).