actinides physics and chemistry\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775

X-ray absorption spectroscopy and actinide electrochemistry: a setup dedicated to radioactive samples applied to neptunium chemistry

crossmark logo

aCEA, DES, ISEC, DMRC, Université de Montpellier, Marcoule, France, bCEA, DES-Service d'Etudes Analytiques et de Réactivité des Surfaces (SEARS), CEA, Université Paris-Sacly, 911191 Gif-sur-Yvette, France, cESTA–École Supérieure des Technologies et des Affaires, 90004 Belfort Cedex, France, and dSynchrotron SOLEIL, L'Orme des Merisiers Saint-Aubin, BP 48, 91192 Gif-sur-Yvette Cedex, France
*Correspondence e-mail: thomas.dumas@cea.fr, michel.schlegel@cea.fr

Edited by W. Shi, Institute of High Energy Physics, People's Republic of China (Received 15 July 2021; accepted 22 October 2021)

A spectroelectrochemical setup has been developed to investigate radioactive elements in small volumes (0.7 to 2 ml) under oxidation–reduction (redox) con­trolled conditions by X-ray absorption spectroscopy (XAS). The cell design is presented together with in situ XAS measurements performed during neptunium redox reactions. Cycling experiments on the NpO22+/NpO2+ redox couple were applied to qualify the cell electrodynamics using XANES measurements and its ability to probe modifications in the neptunyl hydration shell in a 1 mol l−1 HNO3 solution. The XAS results are in agreement with previous structural studies and the NpO22+/NpO2+ standard potential, determined using Nernst methods, is consistent with measurements based on other techniques. Subsequently, the NpO2+, NpO22+ and Np4+ ion structures in solution were stabilized and measured using EXAFS. The resulting fit parameters are again compared with other results from the literature and with theoretical models in order to evaluate how this spectroelectrochemistry experiment succeeds or fails to stabilize the oxidation states of actinides. The experiment succeeded in: (i) implementing a robust and safe XAS device to investigate unstable radioactive species, (ii) evaluate in a reproducible manner the NpO22+/NpO2+ standard potential under dilute conditions and (iii) clarify mechanistic aspects of the actinyl hydration sphere in solution. In contrast, a detailed comparison of EXAFS fit parameters shows that this method is less appropriate than the majority of the previously reported chemical methods for the stabilization of the Np4+ ion.

1. Introduction

In situ X-ray absorption spectroscopy (XAS) in combination with electrochemical techniques is a powerful combination for determining the coordination of metal ions at a controlled oxidation state in various electrolytes (Nockemann et al., 2009[Nockemann, P., Thijs, B., Lunstroot, K., Parac-Vogt, T. N., Görller-Walrand, C., Binnemans, K., Van Hecke, K., Van Meervelt, L., Nikitenko, S., Daniels, J., Hennig, C. & Van Deun, R. (2009). Chem. Eur. J. 15, 1449-1461.]; Achilli et al., 2016[Achilli, E., Minguzzi, A., Visibile, A., Locatelli, C., Vertova, A., Naldoni, A., Rondinini, S., Auricchio, F., Marconi, S., Fracchia, M. & Ghigna, P. (2016). J. Synchrotron Rad. 23, 622-628.]). Such an in situ approach is nowadays commonly applied for d-block elements and rare earth compounds, but is still very limited for actinide elements (An). These elements display unusual redox properties and complexation behaviour (Brown, 1978[Brown, I. D. (1978). Chem. Soc. Rev. 7, 359-376.]). Understanding the equilibria between oxidized and reduced species, the formation of inter­mediate unstable species and the magnitude of relevant redox couples under the conditions of inter­est is essential for predicting An chemistry in industrial and natural environments. Specifically, the characterization of inter­mediate species or unstable oxidation states is of fundamental inter­est in understanding how the inter­nal structures of An mol­ecular complexes affect their reactivity and the exchange between oxidation states.

To investigate these reactions at the mol­ecular scale using XAS, a spectroelectrochemical cell was designed at Argonne National Laboratory (USA) (Antonio et al., 1997[Antonio, M. R., Soderholm, L. & Song, I. (1997). J. Appl. Electrochem. 27, 784-792.]). The limited volume of this cell (5 ml) allowed for the characterization of radioactive samples using a limited amount of haza­rdous radionuclide with a concentration range (10−2 to 10−4 mol l−1) that corresponds to XAS measurements. Thus, this cell has been used to investigate An redox chemistry under acidic conditions (Soderholm et al., 1999[Soderholm, L., Antonio, M. R., Williams, C. & Wasserman, S. R. (1999). Anal. Chem. 71, 4622-4628.]), alkaline conditions (Williams et al., 2001[Williams, C. W., Blaudeau, J. P., Sullivan, J. C., Antonio, M. R., Bursten, B. & Soderholm, L. (2001). J. Am. Chem. Soc. 123, 4346-4347.]) and on frontier elements such as berkelium (Antonio et al., 2002[Antonio, M. R., Williams, C. W. & Soderholm, L. (2002). Radiochim. Acta, 90, 851-856.]). Subsequent XAS research monitoring actinide and radioactive elements coordination during in situ electroactive processes was performed with less active nucleides, such as U (Hennig et al., 2005[Hennig, C., Tutschku, J., Rossberg, A., Bernhard, G. & Scheinost, A. C. (2005). Inorg. Chem. 44, 6655-6661.]) or Tc (Antonio et al., 2002[Antonio, M. R., Williams, C. W. & Soderholm, L. (2002). Radiochim. Acta, 90, 851-856.]; Poineau et al., 2006[Poineau, F., Fattahi, M. & Grambow, B. (2006). Radiochim. Acta, 94, 559-563.]). Although high-activity samples (Np and Pu) were used in several experiments in recent years, their combination with an electrochemical setup was hampered for various safety reasons. To contribute to this effort and to overcome these limitations, we have developed an electrochemical–XAS setup available for high-activity samples. The setup was designed taking into account specific requirements: (i) a small sample volume to limit radioactivity and permit the study of highly radioactive materials; (ii) the possibility to perform reproducible electrochemical reactions in situ (i.e. directly on the beamline) with highly radioactive material during an XAS experiment; (iii) (moderate) flexibility in the choice of working, reference and counter electrodes; (iv) easy access and handling to limit the risk of spillover during sample preparation; (v) tightness and number of barriers satisfying the synchrotron safety requirements for the handling of radioactive samples; and (vi) a small overall size to facilitate transport between the synchrotron facility and actinide laboratories. The cell has been successfully implemented on the MARS beamline at the SOLEIL synchrotron (LLorens et al., 2014[LLorens, I., Solari, P. L., Sitaud, B., Bes, R., Cammelli, S., Hermange, H., Othmane, G., Safi, S., Moisy, P., Wahu, S., Bresson, C., Schlegel, M. L., Menut, D., Bechade, J. L., Martin, P., Hazemann, J. L., Proux, O. & Den Auwer, C. (2014). Radiochim. Acta, 102, 957-972.]). The first set of experiments was performed in 1 M NHO3 to follow in situ structural and electronic changes of the neptunium ions under potentiometric control. The results demonstrate the capacities and limits of such a microcell setup and are discussed in comparison with purely Np electrochemistry results (i.e. laboratory scale and no X-ray techniques) (Cohen & Hindman, 1952[Cohen, D. & Hindman, J. C. (1952). J. Am. Chem. Soc. 74, 4679-4682.]; Cohen et al., 1954[Cohen, D., Sullivan, J. C. & Hindman, J. C. (1954). J. Am. Chem. Soc. 76, 352-354.]; Takao et al., 2009[Takao, K., Takao, S., Scheinost, A. C., Bernhard, G. & Hennig, C. (2009). Inorg. Chem. 48, 8803-8810.]; Hindman et al., 1958[Hindman, J. C., Sullivan, J. C. & Cohen, D. (1958). J. Am. Chem. Soc. 80, 1812-1814.]; Sornein et al., 2009[Sornein, M. O., Mendes, M., Cannes, C., Le Naour, C., Nockemann, P., Van Hecke, K., Van Meervelt, L., Berthet, J. C. & Hennig, C. (2009). Polyhedron, 28, 1281-1286.]; Kihara et al., 1999[Kihara, S., Yoshida, Z., Aoyagi, H., Maeda, K., Shirai, O., Kitatsuji, Y. & Yoshida, Y. (1999). Pure Appl. Chem. 71, 1771-1807.]; Kim et al., 2004[Kim, S. Y., Asakura, T., Morita, Y., Uchiyama, G. & Ikeda, Y. (2004). J. Radioanal. Nucl. Chem. 262, 311-315.], 2005[Kim, S. Y., Asakura, T. & Morita, Y. (2005). Radiochim. Acta, 93, 767-770.]; Cohen, 1961[Cohen, D. (1961). J. Inorg. Nucl. Chem. 18, 207-210.]; Zielen et al., 1958[Zielen, A. J., Sullivan, J. C. & Cohen, D. (1958). J. Inorg. Nucl. Chem. 7, 378-383.]) or previous structural XAS studies (Bonin et al., 2009[Bonin, L., Guillaumont, D., Jeanson, A., Den Auwer, C., Grigoriev, M., Berthet, J. C., Hennig, C., Scheinost, A. & Moisy, P. (2009). Inorg. Chem. 48, 3943-3953.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Di Giandomenico et al., 2009[Di Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347-353.]; Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Soderholm et al., 1999[Soderholm, L., Antonio, M. R., Williams, C. & Wasserman, S. R. (1999). Anal. Chem. 71, 4622-4628.]; Hennig et al., 2005[Hennig, C., Tutschku, J., Rossberg, A., Bernhard, G. & Scheinost, A. C. (2005). Inorg. Chem. 44, 6655-6661.]; Williams et al., 2001[Williams, C. W., Blaudeau, J. P., Sullivan, J. C., Antonio, M. R., Bursten, B. & Soderholm, L. (2001). J. Am. Chem. Soc. 123, 4346-4347.]; Den Auwer et al., 1999[Den Auwer, C., Revel, R., Charbonnel, M. C., Presson, M. T., Conradson, S. D., Simoni, E., Le Du, J. F. & Madic, C. (1999). J. Synchrotron Rad. 6, 101-104.]; Scheinost et al., 2016[Scheinost, A. C., Steudtner, R., Hübner, R., Weiss, S. & Bok, F. (2016). Environ. Sci. Technol. 50, 10413-10420.]; Antonio et al., 1997[Antonio, M. R., Soderholm, L. & Song, I. (1997). J. Appl. Electrochem. 27, 784-792.], 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.], 2012[Antonio, M. R., Williams, C. W., Sullivan, J. A., Skanthakumar, S., Hu, Y. J. & Soderholm, L. (2012). Inorg. Chem. 51, 5274-5281.]; Ikeda-Ohno et al., 2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.], 2009[Ikeda-Ohno, A., Hennig, C., Tsushima, S., Scheinost, A. C., Bernhard, G. & Yaita, T. (2009). Inorg. Chem. 48, 7201-7210.]).

While the main aim was to check the ability of the setup to proceed safely with the combination of electrochemistry and XAS measurements under confined conditions, a more critical evaluation of the structural results is also proposed. The systematic comparison of the structural EXAFS parameters extracted from (i) this work, (ii) formal works and (iii) theoretical calculations is proposed. This allows for a better understanding of the main benefits of such an electrochemical cell concept, but also some limitations that should not be ignored.

2. MARS electrochemical–XAS setup

The setup was designed to facilitate electrochemical control of a limited volume (700 to 2000 µl) of radioactive solution or suspension while at the same time guaranteeing double confinement of these haza­rdous samples and allowing relatively easy handling in the laboratory and on the beamline. The setup was made of an inner cell and an outer envelope (Fig. 1[link]). The inner cell is made of PEEK, a material that is relatively inert, and thus can be used for acidic, basic and even non-aqueous solutions compatible with this material (such as room-temperature ionic liquids and organic solutions like dodecane or hepta­ne). Two X-ray windows are obtained by locally thinning the inner cell walls down to 200 µm. This design limits the risk of leaking at the X-ray windows, while providing a moderate X-ray attenuation (95% of the X-ray flux transmitted at 17 keV). This cell is enclosed in a secondary containment, also made of PEEK, to prevent dispersion in case of inner cell failure. The inner cell is rotated at 45° with respect to the incident beam, and three windows made of 90 µm Kapton film sealed with three screwed clamping rings to the second envelope provide paths for the incident beam, the transmitted beam and the fluorescence signal. The path length of the transmitted X-ray beam in the inner cell is about 11 mm.

[Figure 1]
Figure 1
(Left) Double confinement electrochemical–XAS setup, 0.7 < Vsol < 2 ml (a magnetic stirrer is placed at the bottom of the setup). (Right) The electrolysis cell at the MARS beamline CX3 end-station.

The volumes of liquid that can be introduced in the inner cell vary between 750 (the minimum volume to soak up the windows) and 2000 µl. The bulk solution is steered (600 rpm) by a magnetic bar driven by a stirrer located outside the second envelope. Bulk electrolysis is then performed in the inner cell.

The electrodes are screwed and glued onto the lid of the inner cell to further limit possible leaking of fluids. The nature and distribution of electrodes can be tailored to meet the need of a specific experiment. For example, the working electrode can be made of platinum or any other metallic material, or even carbon. Our conventional reference electrode is an Ag/AgCl microelectrode (World Precision Instruments). How­ever, any other microelectrode could in principle be used, provided it can fit through the inlet (4 mm in diameter) and be reasonably short (a few cm) and sturdy. In a more recent experiment, a homemade reference electrode dedicated to room-temperature ionic liquids was also used in the same cell (Bengio et al., 2018[Bengio, D., Dumas, T., Mendes, E., Solari, P. L., Husar, R., Schlegel, M., Moisy, P. & Pellet-Rostaing, S. (2018). Rare Metal Technology 2018, pp. 99-112. Part of the The Minerals, Metals & Materials Series book series. Switzerland: Springer Nature.], 2020[Bengio, D., Dumas, T., Arpigny, S., Husar, R., Mendes, E., Solari, P. L., Schlegel, M. L., Schlegel, D., Pellet-Rostaing, S. & Moisy, P. (2020). Chem. Eur. J. 26, 14385-14396.]). The counter-electrode can plunge directly into the solution or, alternatively, it can be isolated from the main solution by a tube closed by a porous frit acting as a salt bridge. This setup somewhat hinders parasitic reactions on the electrode surface which might disturb the desired reaction of actinide redox transformation.

The setup performance was assessed by experiments of Np redox in 1 M HNO3. The working electrode is a platinum coiled electrode and the auxiliary electrode is Pt wire separated from the buck by a tube closed by a porous frit. The reference electrode is a 2 mm-diameter Ag/AgCl microelectrode from World Precision Instrument. Preliminary tests with the counter-electrode bathed directly in the experimental solution failed to achieve qu­anti­tative oxidation (or reduction) up to the target oxidation state although the target potentials had been validated previously by spectroelectrochemical UV–Vis absorption spectroscopy. Instead of a progressive change in oxidation state under the applied conditions, competitive side reactions seems to occur and are manifested by important current flow but no significant changes in redox speciation. For the later experiments, the counter-electrode was placed in the frit-sealed tube, which resulted in a reproducible experiment that is presented in the following results.

3. Experimental

3.1. Sample preparation

The sample preparation was performed in a radiochemistry laboratory at the ATALANTE facility (CEA Marcoule, France) in a dedicated glove-box. The purity of the oxidation state of the NpV stock solution was checked with a UV–Vis spectrometer (Carry). A sample of 0.1 mM NpV was transferred to the inner cell via a dedicated holder. The filling hole was then closed with a screw and sealed with ep­oxy glue. Basic tests were performed in the laboratory to check for a correct electrochemical behaviour of the cell. The second envelope was then closed and the cell can be easily shipped due to its relatively small size.

3.2. XAS experiment

The electrochemical cell was installed on the CX3 station of the MARS beamline, which is dedicated to XAS (Jeanson et al., 2009[Jeanson, A., Dahou, S., Guillaumont, D., Moisy, P., Auwer, C. D., Scheinost, A., Hennig, C., Vidaud, C., Subra, G. & Solari, P. L. (2009). J. Phys. Conf. Ser. 190, 012185.]; LLorens et al., 2014[LLorens, I., Solari, P. L., Sitaud, B., Bes, R., Cammelli, S., Hermange, H., Othmane, G., Safi, S., Moisy, P., Wahu, S., Bresson, C., Schlegel, M. L., Menut, D., Bechade, J. L., Martin, P., Hazemann, J. L., Proux, O. & Den Auwer, C. (2014). Radiochim. Acta, 102, 957-972.]). A 13-element HPGe solid-state detector (ORTEC) was used to collect the Np signal in fluor­escence mode. The energy calibration of the mono­chrom­ator was performed at the yttrium K-edge. All measurements were performed at room temperature.

3.3. Bulk electrolysis

Bulk electrolysis was performed over a period of 30 min using a remote-controlled potentiostat (μAutolab, Metrohm) located in the experimental hutch. Chrono-amperometric sequences of 30 min were alternated with acquisition of XAS data. It is worth mentioning that shorter electrolysis times or simultaneous electrolysis–XAS acquisition may artificially suggest hysteresis on Np redox couples under the chosen conditions, a consequence of incomplete or ongoing elec­trol­ysis. Moreover, the recorded chronoamperogram, even indicating the exponential shape expected for dilute solution electrolysis, never reached a null current during the experiment, indicating parasitic reactions.

Data processing was carried out using the Athena code (Ravel & Newville, 2005[Ravel, B. & Newville, M. (2005). J. Synchrotron Rad. 12, 537-541.]). The E0 energy was set at the maximum of the absorption edge. The EXAFS signal was extracted by subtracting a linear pre-edge background and a combination of cubic spline functions for the atomic absorption background and for normalizing the signal by the Lengeler–Eisenberg procedure. Fourier transforms (FT) were obtained by Fourier transform in k3χ(k) between 2.5 and 11 Å−1. Selected FT contributions were fitted in R-space over individual radial distances and Debye–Waller factors (σ2) for every considered distance, using backscattering amplitude and phase shift functions obtained with FEFF8.2 (Rehr & Albers, 2000[Rehr, J. J. & Albers, R. C. (2000). Rev. Mod. Phys. 72, 621-654.]) performed on structures optimized by density functional theory (DFT) calculations (see §4[link], DFT calculations). The amplitude reduction factor E02 was set at 0.9. All fitting operations were performed. The R factor (%) and errors in distances were provided by ARTEMIS (Ravel & Newville, 2005[Ravel, B. & Newville, M. (2005). J. Synchrotron Rad. 12, 537-541.]).

4. DFT calculations

The geometry and frequency calculations were performed with GAUSSIAN16 (Frisch et al., 2016[Frisch, M. J., et al. (2016). GAUSSIAN16. Revision B.01. GaussView Version 5.0. Gaussian Inc., Wallingford, CT, USA.]) at the DFT level of theory. A small core quasi-relativistic effective core potential (RECP-60 electrons) (Cao & Dolg, 2004[Cao, X. Y. & Dolg, M. (2004). J. Mol. Struct. Theochem. 673, 203-209.]; Küchle et al., 1994[Küchle, W., Dolg, M., Stoll, H. & Preuss, H. (1994). J. Chem. Phys. 100, 7535-7542.]) by the Stuttgart–Cologne group and its corresponding TZ-valence basis set were used for the neptunium ion. The PBE0 functional was used with the def-TZVP (Schäfer et al., 1994[Schäfer, A., Huber, C. & Ahlrichs, R. (1994). J. Chem. Phys. 100, 5829-5835.]) basis sets for O and H atoms. Aqueous solvation effects were taken into account using two explicit hydration shells. Effects beyond the second hydration shell were described through an implicit solvation model. The Integral Equation Formalism Polarizable Continuum Model (IEFPCM) was used as implemented in GAUSSIAN16.

The ab initio Debye–Waller factors (σ2) were calculated at 300 K for each scattering path from the dynamical matrix extracted from the DFT frequency calculations with the DMDW module of FEFF9 (Rehr et al., 2010[Rehr, J. J., Kas, J. J., Vila, F. D., Prange, M. P. & Jorissen, K. (2010). Phys. Chem. Chem. Phys. 12, 5503-5513.]; Rehr & Albers, 2000[Rehr, J. J. & Albers, R. C. (2000). Rev. Mod. Phys. 72, 621-654.]).

5. Results

5.1. NpVI/NpV

The first electrolysis experiment was performed with an initial NpV solution. XANES and EXAFS spectra were recorded after 30 min of electrolysis for each potential step at 850, 950, 975, 1000, 1025 and 1150 mV/(Ag/AgCl) in the oxidation (anodic) direction, and at 1150, 1050, 975, 950, 925, 900, 875, 800 and 750 mV/(Ag/AgCl) in the reduction (cathodic) direction (Fig. 2[link]). All XANES spectra display a white line, which is shifted from 17616 to 17619.5 eV as NpV is oxidized to NpVI. The high-energy shoulder of the absorption edge, located near 17628 eV and attributed to multiple scattering in the neptunyl moiety, is also shifted to higher energy, i.e. 17634 eV, consistent with a trans-dioxo structure for both +V and the +VI oxidation states. Conversely, a shift of the white line and of the shoulder back to their initial values is measured during the reduction steps of NpVI back to NpV. From NpV to NpVI, the white line increases in intensity. An isosbestic point at 17617.7 eV repeats itself in both the oxidation and the reduction sequences, suggesting that only two components were simultaneously present in solution for each sequence. Transitional spectra were reproduced by linear combination fits using the two extrema spectra for NpV and NpVI, resulting in NpIV/NpV ratios for each potential. As pointed out by Soderholm et al. (1999[Soderholm, L., Antonio, M. R., Williams, C. & Wasserman, S. R. (1999). Anal. Chem. 71, 4622-4628.]), a direct estimation of the redox species is only possible if there is no change in the experimental setup during the data acquisition except for the imposed potential in the solution.

[Figure 2]
Figure 2
Normalized Np L3-edge XANES spectrum for (a) oxidation and (b) reduction experiments. The red spectrum is the pure NpV XANES spectrum and the blue spectrum is the pure NpVI XANES spectrum. The grey spectrum resamples inter­mediate states.

The relative NpVI/NpV concentrations are determined with a linear combination fit from the two extrema spectra. This can be used in a Nernst plot as log([NpVI/NpV]) plotted as a function of the applied potential (Fig. 3[link]). Both data sets from the oxidation (red) and reduction (blue) experiments are well aligned within error bars. This evidences the reversibility of the NpVI/NpV redox system, confirming that the electrochemical setup operates as expected. From the Nernst plot, a formal potential of the NpVI/NpV redox couple in 1 M NHO3 is determined by linear regression (Fig. 3[link], red and blue dotted lines), according to

[E = E^{\,0\,\prime} + \left( 2.3RT / nF \right) \log \left( \left[ {\rm Np}^{\rm VI} \right] / \left[ {\rm Np}^{\rm V} \right]\right)]

where E is the potential, E0′ is the apparent standard potential, R is the perfect gas constant (J mol−1 K−1), T is temperature (K), n is the number of electrons in the reaction, F is the Faraday constant and [NpVI] and [NpV] are the relative con­centrations in NpVI and NpV, respectively, as determined by the linear combination fit of the neptunium L3-edge spectra.

[Figure 3]
Figure 3
Nernst plots obtained from XANES measurements. Red diamonds cor­respond to the oxidizing experiment and blue diamonds to the reducing experiment. The dashed lines are the linear regression.

The slopes, ranging between 55.5 and 58.2 mV (±8 mV), are in reasonable agreement with a single electron transfer (expected value of 59 mV at T = 298 K). The two formal potentials obtained from the oxidation and reduction experiments are 918.7 and 918.4 mV/(Ag/AgCl), respectively. Herein, the NpVI/NpV formal potential given with a maximum uncertainty of ±10 mV is in good agreement with similar studies performed in perchloric acid (Soderholm et al., 1999[Soderholm, L., Antonio, M. R., Williams, C. & Wasserman, S. R. (1999). Anal. Chem. 71, 4622-4628.]; Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]) or recent work from Chatterjee et al. (2017[Chatterjee, S., Bryan, S. A., Casella, A. J., Peterson, J. M. & Levitskaia, T. G. (2017). Inorg. Chem. Front. 4, 581-594.]) in nitric acid using equivalent spectrophotometric methods.

5.1.1. EXAFS analysis

From electrolysis experiments and analysis of XANES spectra, we demonstrated that the setup provides a fairly robust and reliable method to isolate NpVI and NpV oxidation states by electrolysis. With the potentials set at 1150 and 750 mV/(Ag/AgCl), we expect to stabilize pure species for EXAFS analysis. This controlled potential may stabilize the Np oxidation state upon long EXAFS measurements, thereby balancing for the in situ photooxidation due to beam damage. The purpose of the following section is to evaluate whether the electrochemically purified solution results in a more reliable solution to isolate both neptunium oxidation-state ions in a simple solution. In order to compare the results with previous EXAFS measurements on similar systems, we propose to compare the significant fit parameters (CN and DWF) on a single plot. First, Fig. 4[link] shows the k3-weighted EXAFS oscillations and Fourier transform (FT) for both NpV- and NpVI-stabilized solutions. Neptunyl ion hydrates show two main oscillations corresponding to two FT peaks typical for hydrated actinyl oxocation (Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Bolvin et al., 2001[Bolvin, H., Wahlgren, U., Moll, H., Reich, T., Geipel, G., Fanghänel, T. & Grenthe, I. (2001). J. Phys. Chem. A, 105, 11441-11445.]; Di Giandomenico et al., 2009[Di Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347-353.]; Duvail et al., 2019[Duvail, M., Dumas, T., Paquet, A., Coste, A., Berthon, L. & Guilbaud, P. (2019). Phys. Chem. Chem. Phys. 21, 7894-7906.]). The structural parameters from simple two oxygen shell EXAFS fits are summarized in Table 1[link]. The coordination spheres of NpVI and NpV are formed by two O atoms from the neptunyl moiety (O­yl) at short R(Np–O­yl) distances of 1.75 and 1.82 Å, respectively, and by about 4.4 and 4.9 equatorial O atoms (Oeq) from water mol­ecules at longer R(Np–Oeq) distances of 2.41 and 2.51 Å, respectively. Overall, these results are in line with published values and confirm that a single oxidation state predominates in solution. Distances correspond to the values reported in the literature within a maximum deviation of 0.02 Å (Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Antonio et al., 1997[Antonio, M. R., Soderholm, L. & Song, I. (1997). J. Appl. Electrochem. 27, 784-792.], 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]; Den Auwer et al., 1999[Den Auwer, C., Revel, R., Charbonnel, M. C., Presson, M. T., Conradson, S. D., Simoni, E., Le Du, J. F. & Madic, C. (1999). J. Synchrotron Rad. 6, 101-104.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Bolvin et al., 2001[Bolvin, H., Wahlgren, U., Moll, H., Reich, T., Geipel, G., Fanghänel, T. & Grenthe, I. (2001). J. Phys. Chem. A, 105, 11441-11445.]; Williams et al., 2001[Williams, C. W., Blaudeau, J. P., Sullivan, J. C., Antonio, M. R., Bursten, B. & Soderholm, L. (2001). J. Am. Chem. Soc. 123, 4346-4347.]; Kim et al., 2004[Kim, S. Y., Asakura, T., Morita, Y., Uchiyama, G. & Ikeda, Y. (2004). J. Radioanal. Nucl. Chem. 262, 311-315.]; Denecke et al., 2005[Denecke, M. A., Dardenne, K. & Marquardt, C. M. (2005). Talanta, 65, 1008-1014.]; Kim et al., 2005[Kim, S. Y., Asakura, T. & Morita, Y. (2005). Radiochim. Acta, 93, 767-770.]; Ikeda-Ohno et al., 2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.]; Di Giandomenico et al., 2009[Di Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347-353.]; Hennig et al., 2009[Hennig, C., Ikeda-Ohno, A., Tsushima, S. & Scheinost, A. C. (2009). Inorg. Chem. 48, 5350-5360.]; Takao et al., 2009[Takao, K., Takao, S., Scheinost, A. C., Bernhard, G. & Hennig, C. (2009). Inorg. Chem. 48, 8803-8810.]). However, the hydration numbers (NOeq) are known to be much less accurately determined from the EXAFS fits and are still actively discussed. For NpVI, the NOeq values from previous EXAFS measurements range between 4.6 and 5.3. For NpV, the NOeq values are between 3.6 and 5.2.

Table 1
Metrical parameters from EXAFS fits

Asterisks (*) indicate fixed parameters.

  Path N R σ2
NpVI ΔE0 5 eVR factor 2.1% O­yl 2* 1.75 (1) 0.0022 (10)
OH2O 4.6 (7) 2.41 (2) 0.0035 (15)
 
NpV ΔE0 −1 eVR factor 1.8% O­yl 2* 1.82 (1) 0.0019 (10)
OH2O 4.9 (6) 2.51 (2) 0.0060 (15)
[Figure 4]
Figure 4
Experimental FT EXAFS signal (line) and best fits (open circles) obtained after 30 min electrolysis at 1150 mV/(Ag/AgCl) (blue) and at 750 mV/(Ag/AgCl) (red). The inset is the corresponding k3-weighted EXAFS oscillations.

The large dispersions reported in the literature are likely due to the mathematical correlation observed in a fitting procedure between the number of scattering atoms in a given shell (N), the Debye–Waller factor (σ2) and the value of the total amplitude reduction factor ( #!S0 2). Although there is no physical meaning between these physical parameters, upon an EXAFS fit the three parameters N, #!S0 2 and σ2 result in similar effects on the oscilation amplitudes. It is therefore difficult to address coordination numbers from a single EXAFS signal with perfect accuracy, especially if the k range is short.

In order to estimate the magnitude of these correlations and allow a good comparison between previous EXAFS fits performed under different conditions (i.e. fixed or floating CN values and σ2), a more complete EXAFS data analysis is proposed hereafter. This analysis follows a method proposed by Ikeda-Ohno et al. (2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.]). NOeq is successively fixed from 3 to 6 and, for each coordination number step, a best fit is performed. This results in the determination of a conditional σ2 variation as a function of N. The results can be displayed on a σ2N plot, as proposed by Ikeda-Ohno et al. (2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.]) (Fig. 5[link]). The best NOeq and σ2 combinations (corresponding to Table 1[link]; values for NpV and NpVI) are plotted as triangles in Fig. 5[link].

[Figure 5]
Figure 5
σ2N plots from the NpVI and NpV fits. Blue and red dashed lines are the σ2 variation as a function of NOeq for NpVI and NpV, respectively. Previous reported single-point fit values are given as grey diamonds (Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]; Ikeda-Ohno et al., 2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.], 2009[Ikeda-Ohno, A., Hennig, C., Tsushima, S., Scheinost, A. C., Bernhard, G. & Yaita, T. (2009). Inorg. Chem. 48, 7201-7210.]; Hennig et al., 2009[Hennig, C., Ikeda-Ohno, A., Tsushima, S. & Scheinost, A. C. (2009). Inorg. Chem. 48, 5350-5360.]; Di Giandomenico et al., 2009[Di Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347-353.]; Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Scheinost et al., 2016[Scheinost, A. C., Steudtner, R., Hübner, R., Weiss, S. & Bok, F. (2016). Environ. Sci. Technol. 50, 10413-10420.]; Denecke et al., 2005[Denecke, M. A., Dardenne, K. & Marquardt, C. M. (2005). Talanta, 65, 1008-1014.]). The open circles are DFT-calculated parameters.

For NpV (Fig. 5[link], red dashed line) and NpVI (Fig. 5[link], blue dashed line) the σ2 values were found to vary approximately linearly with NOeq. The NpV trend is similar to that reported by Ikeda-Ohno et al. (2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.]) for NpV. The corresponding R factors follow a parabolic variation (not shown) with a minimum at NOeq = 4.9 for NpV and a minimum at NOeq = 4.6 for NpVI. However, for both NpVI and NpV, reasonable fit parameters were obtained for all tested NOeq values (i.e. RF < 5% and σ2 < 0.015 Å2). This is the reason why this semi-empirical methodology was applied to compare the data with previous results whatever the method used to fit the data. Since the fit can be performed assuming multiple NOeq values, a relevant comparison with any single data point from previous reports needs to be achieved toward the full NOeq range.

On Fig. 5[link], the grey diamonds are previous fit parameters determined under similar chemical conditions (i.e. noncomplexing and acidic solution). Both ex situ measurements (i.e. chemical preparation) (Hennig et al., 2009[Hennig, C., Ikeda-Ohno, A., Tsushima, S. & Scheinost, A. C. (2009). Inorg. Chem. 48, 5350-5360.]; Ikeda-Ohno et al., 2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.]; Di Giandomenico et al., 2009[Di Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347-353.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]) and in situ measurements by spectroelectrochemistry (Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]) are compared with NpV and NpVI fit parameters from this work. For a given NOeq value, the σ2 values (dashed blue and red lines) determined in this work are comparable with or lower than previously published σ2 parameters. This is a good indication for a lower disorder/higher purity in this work where NpV and NpVI were purified electrochemically.

Concomitantly, a comparison of the NpV and NpVI structures is in agreement with the expected changes in the neptunyl ion coordination. The σ2 values tend to be lower for NpVI than for NpV for a given NOeq. This result agrees well with the decrease in atomic charge and bonding strength between water and neptunyl units from NpVI to NpV (Choppin & Rao, 1984[Choppin, G. R. & Rao, L. F. (1984). Radiochimica Acta, 37, 143-146.]; Denning, 2007[Denning, R. G. (2007). J. Phys. Chem. A, 111, 4125-4143.]).

At this stage, we conclude that the sample preparation, as well as the continuous application of electrochemical potential, is a good way to maintain a pure oxidation state during X-ray measurements resulting in a low conformational σ2. However, this better oxidation-state purification does not solve the difficult question of the hydration structure of actinyl ions in solution. To evaluate this point further, additional information may be extracted to narrow down the actual NOeq values. With this same aim, Ikeda-Ohno et al. (2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.]) suggested that the NOeq values be derived from the more accurately determined inter­atomic distances applying the bond valence model (Brown, 1978[Brown, I. D. (1978). Chem. Soc. Rev. 7, 359-376.]). Although this approach proved to be satisfying for the well-known coordination chemistry of UVI, application to Np is plagued by the limited number of crystallographic references and by uncertainties in the valence unit and the charges of the neptunyl moieties, so that NOeq values still range between 4 and 6. Another way to constrain the accurate coordination number is to associate EXAFS analysis with quantum chemical calculations (Vila et al., 2012[Vila, F. D., Lindahl, V. E. & Rehr, J. J. (2012). Phys. Rev. B, 85, 024303.]). For some actinide mol­ecular compounds, it is possible to generate EXAFS metrical parameters from a DFT calculation (Acher et al., 2016[Acher, E., Hacene Cherkaski, Y., Dumas, T., Tamain, C., Guillaumont, D., Boubals, N., Javierre, G., Hennig, C., Solari, P. L. & Charbonnel, M. C. (2016). Inorg. Chem. 55, 5558-5569.], 2017[Acher, E., Dumas, T., Tamain, C., Boubals, N., Solari, P. L. & Guillaumont, D. (2017). Dalton Trans. 46, 3812-3815.]; Dalodière et al., 2018[Dalodière, E., Virot, M., Dumas, T., Guillaumont, D., Illy, M. C., Berthon, C., Guerin, L., Rossberg, A., Venault, L., Moisy, P. & Nikitenko, S. I. (2018). Inorg. Chem. Front. 5, 100-111.]). The calculation provides distances and σ2 values ab initio from a selected structure. The NpO2(H2O)4(H2O)82+, NpO2(H2O)5(H2O)102+, NpO2(H2O)4(H2O)8+ and NpO2(H2O)5(H2O)10+ complexes in the presence of a continuum solvent model were selected to produce the parameters reported in Table 2[link]. The accuracy of the optimized geometry was checked by direct comparison of Np—Oeq bond lengths with the measurements. This type of calculation provides an estimated value for the thermal σ2 for a given coordination number and for both NpV and NpVI. It was possible to compare it with the experimental fit values in Fig. 5[link] (red and blue open circles). For both NpVI and NpV, inter­estingly, σ2 follows the same trend as already observed for experimental fit parameters. However, this trend is obviously not due to the correlation of EXAFS fit parameters since the two values (CN and σ2) are determined independently. This inquires a physical inter­relation between a bonding strength and the actinyl coordination number in actinyl solvation. This effect fundamentally impedes the direct determination of actinyl solvatation using the amplitude of the equatorial scattering shell in the EXAFS signal. Again, using distances from both EXAFS and DFT to discriminate between penta- and tetra­hydrated neptunyl ions seems a more reliable method. For NpV with five water mol­ecules, both the calculation and the fit converge for a 2.51 Å bond. For NpVI with five water mol­ecules, 2.41 and 2.42 Å bond lengths are obtained from the fit and the calculation, respectively. Overall, the results support the fact that five water mol­ecules coordinate both the NpVI and the NpV ions.

Table 2
DFT calculations of Np—O bond distances (average values in Å) and Debye–Waller factor σ22) in NpV and NpVI hydrated clusters with four and five coordinated water mol­ecules

  Np—O­yl σ2 Np—Owat σ2
NpV(H2O)4(H2O)8+ 1.80 0.0014 2.47 0.0048
NpV(H2O)5(H2O)10+ 1.82 0.0016 2.51 0.0068
NpVI(H2O)4(H2O)82+ 1.73 0.0012 2.32 0.0033
NpVI(H2O)5(H2O)102+ 1.73 0.0013 2.42 0.0050

5.2. NpV/NpIV

To evaluate further the electrochemistry setup, a second electrolysis experiment was performed following the previous one. After stabilizing the +V oxidation state by a 1 h electrolysis at 750 mV/(Ag/AgCl), the NpV purity was checked again using XANES and EXAFS. Next, the NpV reduction was investigated by a stepwise decrease of the applied potential to 0, −100, −150, −200 and −250 mV/(Ag/AgCl). In HNO3 solution, the NpIV ion is difficult to stabilize due to its reoxidation with HNO2 in equilibrium with nitrate ions (Taylor et al., 1998[Taylor, R. J., Koltunov, V. S., Savilova, O. A., Zhuravleva, G. I., Denniss, I. S. & Wallwork, A. L. (1998). J. Alloys Compd. 271-273, 817-820.]; Koltunov et al., 1997[Koltunov, V. S., Taylor, R. J., Savilova, O. A., Zhuravleva, G. I., Denniss, I. S. & Wallwork, A. L. (1997). Radiochim. Acta, 76, 45-54.]). The corresponding XANES spectra are displayed on Fig. 4[link]. Consistent with previous XANES measurements, the Np L3-edge drastically changes due to cleavage of the two Np—O bonds in the neptunyl moiety (Bonin et al., 2009[Bonin, L., Guillaumont, D., Jeanson, A., Den Auwer, C., Grigoriev, M., Berthet, J. C., Hennig, C., Scheinost, A. & Moisy, P. (2009). Inorg. Chem. 48, 3943-3953.]; Denecke et al., 2005[Denecke, M. A., Dardenne, K. & Marquardt, C. M. (2005). Talanta, 65, 1008-1014.]; Hennig et al., 2009[Hennig, C., Ikeda-Ohno, A., Tsushima, S. & Scheinost, A. C. (2009). Inorg. Chem. 48, 5350-5360.]; Scheinost et al., 2016[Scheinost, A. C., Steudtner, R., Hübner, R., Weiss, S. & Bok, F. (2016). Environ. Sci. Technol. 50, 10413-10420.]). Actinide reduction from +V to +IV results in an increase of the white line intensity and a decrease in the amplitude of the higher-energy shoulder at 17637 eV. However, the edge inflection point is almost stabilized by the reduction. XANES spectra are consistent with previous ex situ studies and indicate an almost complete reduction of NpV to NpIV. More unexpected is the corresponding potential at which this electroreduction occurred. As shown in Fig. 6[link], it was mostly NpV that was detected in solution at the potential step of 0 mV. At −100 and −150 mV, the spectra indicate mixed NpV/NpIV contributions. Qu­anti­tative NpV reduction is eventually achieved at −200 and −250 mV.

[Figure 6]
Figure 6
Normalized Np L3-edge XANES spectrum for the NpV reduction experiment at 750, 0, −100, −150, −200 and −250 mV/(Ag/AgCl).

According to recent work (Chatterjee et al., 2017[Chatterjee, S., Bryan, S. A., Casella, A. J., Peterson, J. M. & Levitskaia, T. G. (2017). Inorg. Chem. Front. 4, 581-594.]), reduction to NpIV is expected to occur at approximately 50 mV/(Ag/AgCl) in 1 M HNO3. Moreover, one would also expect a reduction to the NpIII oxidation state below −100 mV/(Ag/AgCl). The NpIV/NpIII redox potential was determined at approximately −50 mV (Guillaumont et al., 2003[Guillaumont, R., Fanghänel, T., Fuger, J., Grenthe, I., Neck, V., Palmer, D. A. & Rand, M. H. (2003). Update on the Chemical Thermodynamics of U, Np, Pu, Am and Tc, Vol. 5. Amsterdam: Elsevier.]). In such a case, the resulting neptunium L3 spectra would be a combination of the two redox species. However, NpIII is unstable under aerobic conditions and only NpIV seems to be present in the solution after this enforced electrolysis (Hindman et al., 1949[Hindman, J. C., Magnusson, L. B. & LaChapelle, T. J. (1949). J. Am. Chem. Soc. 71, 687-693.]; Sjoblom & Hindman, 1951[Sjoblom, R. & Hindman, J. C. (1951). J. Am. Chem. Soc. 73, 1744-1751.]; Sullivan et al., 1976[Sullivan, J. C., Gordon, S., Cohen, D., Mulac, W. & Schmidt, K. H. (1976). J. Phys. Chem. 80, 1684-1686.]; Cohen, 1976[Cohen, D. (1976). Abstr. Pap. Am. Chem. Soc. 172, 28.]).

5.2.1. EXAFS analysis

From electrolysis experiments and analysis of XANES spectra, we demonstrated that it is possible to obtain NpIV by electrolysis at −250 mV. The k3-weighted Np L3-edge EXAFS spectrum of NpIV hydrate shows a single-frequency oscillation (Fig. 7[link]) and the corresponding FT displays a single peak which can be reliably modelled with a single Np—O coordination shell. The structural parameters from this fit are summarized in Table 3[link]. The best fit values correspond to approximately 9.5 O atoms (from water mol­ecules), with an average R(Np–O) distance of 2.39 Å.

Table 3
Parameters from NpIV EXAFS fits

Asterisks (*) indicate fixed parameters.

  Path CN R σ2
#!S0 2 1* ΔE0 −4 eV R factor 3.8% OH2O 9.5 (16) 2.39 (2) 0.0095 (15)
[Figure 7]
Figure 7
Experimental k3-weighted EXAFS oscillations (line) and best fits (open circles) obtained after 30 min electrolysis at −250 mV (green).

The Np—O distance corresponds to the values reported in the literature of 2.37–2.41 Å (Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Williams et al., 2001[Williams, C. W., Blaudeau, J. P., Sullivan, J. C., Antonio, M. R., Bursten, B. & Soderholm, L. (2001). J. Am. Chem. Soc. 123, 4346-4347.]; Bolvin et al., 2001[Bolvin, H., Wahlgren, U., Moll, H., Reich, T., Geipel, G., Fanghänel, T. & Grenthe, I. (2001). J. Phys. Chem. A, 105, 11441-11445.]; Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]). The NH2O value also falls within the range of previous EXAFS results for NpIV aqua complexes, i.e. between 8.7 and 11.6 (Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Williams et al., 2001[Williams, C. W., Blaudeau, J. P., Sullivan, J. C., Antonio, M. R., Bursten, B. & Soderholm, L. (2001). J. Am. Chem. Soc. 123, 4346-4347.]; Bolvin et al., 2001[Bolvin, H., Wahlgren, U., Moll, H., Reich, T., Geipel, G., Fanghänel, T. & Grenthe, I. (2001). J. Phys. Chem. A, 105, 11441-11445.]; Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]).

As for NpV and NpVI, a clear correlation between σ2 and NO can be drawn with restricted NO fits. Fixing NO from 8 to 11 allows the determination of the corresponding σ2. These values approximately align as a function of NO (Fig. 8[link], green dashed line), together with the best fit values (triangle). The corresponding data determined previously on analogous systems are compared (grey diamond). Inter­estingly, almost all previous studies report a lower σ2 value for a given NO compared with the present report. The more distorted geometry of the 8 to 11 water mol­ecule coordination shell results in σ2 values (from 0.0065 to 0.012 Å2) much larger than observed for the neptunyl hydration shell. This time the comparison with previous studies clearly indicates a residual disorder in the neptunium coordination sphere for the in situ prepared Np4+. This may be explained by either (i) a residual NpO2+ contribution in the EXAFS spectrum, (ii) over-reduction resulting in the formation of NpIII with a coordination shell at about 2.5 Å (Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]; Brendebach et al., 2009[Brendebach, B., Banik, N. L., Marquardt, C. M., Rothe, J., Denecke, M. A. & Geckeis, H. (2009). Radiochim. Acta, 97, 701-708.]) or (iii) complexation of NpIV by nitrate (by analogy with PuIV chemistry in nitric acid this must be quite insignificant; Allen et al., 1996[Allen, P. G., Veirs, D. K., Conradson, S. D., Smith, C. A. & Marsh, S. F. (1996). Inorg. Chem. 35, 2841-2845.]), or a combination of points (i), (ii) and (iii).

[Figure 8]
Figure 8
Plot of the σ2 values as a function of coordination number from the EXAFS fit (dashed line) and comparison with previously reported values (Antonio et al., 2001[Antonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17-26.]; Ikeda-Ohno et al., 2008[Ikeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294-8305.], 2009[Ikeda-Ohno, A., Hennig, C., Tsushima, S., Scheinost, A. C., Bernhard, G. & Yaita, T. (2009). Inorg. Chem. 48, 7201-7210.]; Hennig et al., 2009[Hennig, C., Ikeda-Ohno, A., Tsushima, S. & Scheinost, A. C. (2009). Inorg. Chem. 48, 5350-5360.]; Di Giandomenico et al., 2009[Di Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347-353.]; Combes et al., 1992[Combes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376-382.]; Allen et al., 1997[Allen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676-4683.]; Reich et al., 2000[Reich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633-638.]; Scheinost et al., 2016[Scheinost, A. C., Steudtner, R., Hübner, R., Weiss, S. & Bok, F. (2016). Environ. Sci. Technol. 50, 10413-10420.]; Denecke et al., 2005[Denecke, M. A., Dardenne, K. & Marquardt, C. M. (2005). Talanta, 65, 1008-1014.]) (grey diamonds). The triangle shows the best fit results from this work.

6. Conclusions

The main aim of this study was to develop a spectroelectrochimical cell and to qualify this setup for further applications on radioactive samples. During the experiment, it was possible to investigate the neptunium coordination during redox processes in a simple 1 M HNO3 solution.

The technical parameters of the microcell design are presented herein and its first application to transuranic elements, performed safely, are reported. The cell design minimizes the volume of radioactive solution and facilitates transport and handling on the beamline. Low actinide concentration and volume minimize the sample activity and allows XANES and EXAFS measurements within a reasonable timescale. From an electrochemistry point of view, such a static confinement concept is not ideal. On one hand, cycling the NpO22+/NpO2+ redox couple was quite possible and resulted in reliable standard potential determination. It was then possible to measure both NpV and NpVI EXAFS spectra under fully consistent conditions. This was a good opportunity to compare the NpV and NpVI hydration spheres, and to implement a statistical comparison with formal literature results, as well as theoretical models. The evolution in the equatorial actinyl hydration is well characterized and is consistent with theoretical expectations. Moreover, a detailed comparison with previous structural data using EXAFS established that the stabilization of NpV and NpVI oxidation by in situ electrochemistry appears to be the most reliable way to maintain a pure oxidation state during X-ray measurements. On the other hand, the reduction to NpIV never fully succeeded in maintaining pure NpIV in solution. From the point of view of electrochemistry, a clear offset in the expected standard potential must be applied to begin the NpV reduction. The resulting NpIV EAXFS spectra reveal hydrated Np4+ as the main species, but comparison with a previous chemically stabilized NpIV solution indicates a more disordered coordination sphere. Altogether, it seems that, while the results are fully satisfying for the oxidized neptunium species, the cell must be used with care to study reduced neptunium forms. As a final point, this cell, primarily designed to confine radioactive samples for user safety, was also efficient for performing measurements on nonradioactive samples. Its design for radioactive samples appears to be versatile and extremely convenient for preventing moisture and oxygen contaminating air-sensitive samples in the reverse direction. By doing so, the in situ XAS stabilization and characterization of LnII in room-temperature ionic liquid samples was made possible with this cell (Bengio et al., 2018[Bengio, D., Dumas, T., Mendes, E., Solari, P. L., Husar, R., Schlegel, M., Moisy, P. & Pellet-Rostaing, S. (2018). Rare Metal Technology 2018, pp. 99-112. Part of the The Minerals, Metals & Materials Series book series. Switzerland: Springer Nature.], 2020[Bengio, D., Dumas, T., Arpigny, S., Husar, R., Mendes, E., Solari, P. L., Schlegel, M. L., Schlegel, D., Pellet-Rostaing, S. & Moisy, P. (2020). Chem. Eur. J. 26, 14385-14396.]) and other applications of this kind are expected.

References

First citationAcher, E., Dumas, T., Tamain, C., Boubals, N., Solari, P. L. & Guillaumont, D. (2017). Dalton Trans. 46, 3812–3815.  Web of Science CrossRef CAS PubMed Google Scholar
First citationAcher, E., Hacene Cherkaski, Y., Dumas, T., Tamain, C., Guillaumont, D., Boubals, N., Javierre, G., Hennig, C., Solari, P. L. & Charbonnel, M. C. (2016). Inorg. Chem. 55, 5558–5569.  Web of Science CSD CrossRef CAS PubMed Google Scholar
First citationAchilli, E., Minguzzi, A., Visibile, A., Locatelli, C., Vertova, A., Naldoni, A., Rondinini, S., Auricchio, F., Marconi, S., Fracchia, M. & Ghigna, P. (2016). J. Synchrotron Rad. 23, 622–628.  Web of Science CrossRef IUCr Journals Google Scholar
First citationAllen, P. G., Bucher, J. J., Shuh, D. K., Edelstein, N. M. & Reich, T. (1997). Inorg. Chem. 36, 4676–4683.  CrossRef PubMed CAS Web of Science Google Scholar
First citationAllen, P. G., Veirs, D. K., Conradson, S. D., Smith, C. A. & Marsh, S. F. (1996). Inorg. Chem. 35, 2841–2845.  CrossRef CAS Web of Science Google Scholar
First citationAntonio, M. R., Soderholm, L. & Song, I. (1997). J. Appl. Electrochem. 27, 784–792.  CrossRef CAS Web of Science Google Scholar
First citationAntonio, M. R., Williams, C. W. & Soderholm, L. (2002). Radiochim. Acta, 90, 851–856.  Web of Science CrossRef CAS Google Scholar
First citationAntonio, M. R., Williams, C. W., Sullivan, J. A., Skanthakumar, S., Hu, Y. J. & Soderholm, L. (2012). Inorg. Chem. 51, 5274–5281.  Web of Science CrossRef CAS PubMed Google Scholar
First citationAntonio, R., Soderholm, L., Williams, C. W., Blaudeau, J. P. & Bursten, B. E. (2001). Radiochim. Acta, 89, 17–26.  CrossRef CAS Google Scholar
First citationBengio, D., Dumas, T., Arpigny, S., Husar, R., Mendes, E., Solari, P. L., Schlegel, M. L., Schlegel, D., Pellet–Rostaing, S. & Moisy, P. (2020). Chem. Eur. J. 26, 14385–14396.  Web of Science CrossRef CAS PubMed Google Scholar
First citationBengio, D., Dumas, T., Mendes, E., Solari, P. L., Husar, R., Schlegel, M., Moisy, P. & Pellet-Rostaing, S. (2018). Rare Metal Technology 2018, pp. 99–112. Part of the The Minerals, Metals & Materials Series book series. Switzerland: Springer Nature.  Google Scholar
First citationBolvin, H., Wahlgren, U., Moll, H., Reich, T., Geipel, G., Fanghänel, T. & Grenthe, I. (2001). J. Phys. Chem. A, 105, 11441–11445.  Web of Science CrossRef CAS Google Scholar
First citationBonin, L., Guillaumont, D., Jeanson, A., Den Auwer, C., Grigoriev, M., Berthet, J. C., Hennig, C., Scheinost, A. & Moisy, P. (2009). Inorg. Chem. 48, 3943–3953.  Web of Science CSD CrossRef PubMed CAS Google Scholar
First citationBrendebach, B., Banik, N. L., Marquardt, C. M., Rothe, J., Denecke, M. A. & Geckeis, H. (2009). Radiochim. Acta, 97, 701–708.  Web of Science CrossRef CAS Google Scholar
First citationBrown, I. D. (1978). Chem. Soc. Rev. 7, 359–376.  CrossRef CAS Web of Science Google Scholar
First citationCao, X. Y. & Dolg, M. (2004). J. Mol. Struct. Theochem. 673, 203–209.  Web of Science CrossRef CAS Google Scholar
First citationChatterjee, S., Bryan, S. A., Casella, A. J., Peterson, J. M. & Levitskaia, T. G. (2017). Inorg. Chem. Front. 4, 581–594.  Web of Science CrossRef CAS Google Scholar
First citationChoppin, G. R. & Rao, L. F. (1984). Radiochimica Acta, 37, 143–146.  CrossRef CAS Google Scholar
First citationCohen, D. (1961). J. Inorg. Nucl. Chem. 18, 207–210.  CrossRef CAS Web of Science Google Scholar
First citationCohen, D. (1976). Abstr. Pap. Am. Chem. Soc. 172, 28.  Google Scholar
First citationCohen, D. & Hindman, J. C. (1952). J. Am. Chem. Soc. 74, 4679–4682.  CrossRef CAS Web of Science Google Scholar
First citationCohen, D., Sullivan, J. C. & Hindman, J. C. (1954). J. Am. Chem. Soc. 76, 352–354.  CrossRef CAS Web of Science Google Scholar
First citationCombes, J. M., Chisholm-Brause, C. J., Brown, G. E., Parks, G. A., Conradson, S. D., Eller, P. G., Triay, I. R., Hobart, D. E. & Miejer, A. (1992). Environ. Sci. Technol. 26, 376–382.  CrossRef CAS Web of Science Google Scholar
First citationDalodière, E., Virot, M., Dumas, T., Guillaumont, D., Illy, M. C., Berthon, C., Guerin, L., Rossberg, A., Venault, L., Moisy, P. & Nikitenko, S. I. (2018). Inorg. Chem. Front. 5, 100–111.  Google Scholar
First citationDen Auwer, C., Revel, R., Charbonnel, M. C., Presson, M. T., Conradson, S. D., Simoni, E., Le Du, J. F. & Madic, C. (1999). J. Synchrotron Rad. 6, 101–104.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationDenecke, M. A., Dardenne, K. & Marquardt, C. M. (2005). Talanta, 65, 1008–1014.  Web of Science CrossRef PubMed CAS Google Scholar
First citationDenning, R. G. (2007). J. Phys. Chem. A, 111, 4125–4143.  Web of Science CrossRef PubMed CAS Google Scholar
First citationDi Giandomenico, M. V., Naour, C. L., Simoni, E., Guillaumont, D., Moisy, P., Hennig, C., Conradson, S. D. & Den Auwer, C. (2009). Radiochim. Acta, 97, 347–353.  Web of Science CrossRef CAS Google Scholar
First citationDuvail, M., Dumas, T., Paquet, A., Coste, A., Berthon, L. & Guilbaud, P. (2019). Phys. Chem. Chem. Phys. 21, 7894–7906.  Web of Science CrossRef CAS PubMed Google Scholar
First citationFrisch, M. J., et al. (2016). GAUSSIAN16. Revision B.01. GaussView Version 5.0. Gaussian Inc., Wallingford, CT, USA.  Google Scholar
First citationGuillaumont, R., Fanghänel, T., Fuger, J., Grenthe, I., Neck, V., Palmer, D. A. & Rand, M. H. (2003). Update on the Chemical Thermodynamics of U, Np, Pu, Am and Tc, Vol. 5. Amsterdam: Elsevier.  Google Scholar
First citationHennig, C., Ikeda-Ohno, A., Tsushima, S. & Scheinost, A. C. (2009). Inorg. Chem. 48, 5350–5360.  Web of Science CrossRef PubMed CAS Google Scholar
First citationHennig, C., Tutschku, J., Rossberg, A., Bernhard, G. & Scheinost, A. C. (2005). Inorg. Chem. 44, 6655–6661.  Web of Science CrossRef PubMed CAS Google Scholar
First citationHindman, J. C., Magnusson, L. B. & LaChapelle, T. J. (1949). J. Am. Chem. Soc. 71, 687–693.  CrossRef CAS Web of Science Google Scholar
First citationHindman, J. C., Sullivan, J. C. & Cohen, D. (1958). J. Am. Chem. Soc. 80, 1812–1814.  CrossRef CAS Web of Science Google Scholar
First citationIkeda-Ohno, A., Hennig, C., Rossberg, A., Funke, H., Scheinost, A. C., Bernhard, G. & Yaita, T. (2008). Inorg. Chem. 47, 8294–8305.  Web of Science PubMed CAS Google Scholar
First citationIkeda-Ohno, A., Hennig, C., Tsushima, S., Scheinost, A. C., Bernhard, G. & Yaita, T. (2009). Inorg. Chem. 48, 7201–7210.  Web of Science PubMed CAS Google Scholar
First citationJeanson, A., Dahou, S., Guillaumont, D., Moisy, P., Auwer, C. D., Scheinost, A., Hennig, C., Vidaud, C., Subra, G. & Solari, P. L. (2009). J. Phys. Conf. Ser. 190, 012185.  CrossRef Google Scholar
First citationKihara, S., Yoshida, Z., Aoyagi, H., Maeda, K., Shirai, O., Kitatsuji, Y. & Yoshida, Y. (1999). Pure Appl. Chem. 71, 1771–1807.  Web of Science CrossRef CAS Google Scholar
First citationKim, S. Y., Asakura, T. & Morita, Y. (2005). Radiochim. Acta, 93, 767–770.  Web of Science CrossRef CAS Google Scholar
First citationKim, S. Y., Asakura, T., Morita, Y., Uchiyama, G. & Ikeda, Y. (2004). J. Radioanal. Nucl. Chem. 262, 311–315.  Web of Science CrossRef CAS Google Scholar
First citationKoltunov, V. S., Taylor, R. J., Savilova, O. A., Zhuravleva, G. I., Denniss, I. S. & Wallwork, A. L. (1997). Radiochim. Acta, 76, 45–54.  CrossRef CAS Google Scholar
First citationKüchle, W., Dolg, M., Stoll, H. & Preuss, H. (1994). J. Chem. Phys. 100, 7535–7542.  Google Scholar
First citationLLorens, I., Solari, P. L., Sitaud, B., Bes, R., Cammelli, S., Hermange, H., Othmane, G., Safi, S., Moisy, P., Wahu, S., Bresson, C., Schlegel, M. L., Menut, D., Bechade, J. L., Martin, P., Hazemann, J. L., Proux, O. & Den Auwer, C. (2014). Radiochim. Acta, 102, 957–972.  Web of Science CrossRef CAS Google Scholar
First citationNockemann, P., Thijs, B., Lunstroot, K., Parac-Vogt, T. N., Görller-Walrand, C., Binnemans, K., Van Hecke, K., Van Meervelt, L., Nikitenko, S., Daniels, J., Hennig, C. & Van Deun, R. (2009). Chem. Eur. J. 15, 1449–1461.  Web of Science CSD CrossRef PubMed CAS Google Scholar
First citationPoineau, F., Fattahi, M. & Grambow, B. (2006). Radiochim. Acta, 94, 559–563.  Web of Science CrossRef CAS Google Scholar
First citationRavel, B. & Newville, M. (2005). J. Synchrotron Rad. 12, 537–541.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationRehr, J. J. & Albers, R. C. (2000). Rev. Mod. Phys. 72, 621–654.  Web of Science CrossRef CAS Google Scholar
First citationRehr, J. J., Kas, J. J., Vila, F. D., Prange, M. P. & Jorissen, K. (2010). Phys. Chem. Chem. Phys. 12, 5503–5513.  Web of Science CrossRef CAS PubMed Google Scholar
First citationReich, T., Bernhard, G., Geipel, G., Funke, H., Hennig, C., Roßberg, A., Matz, W., Schell, N. & Nitsche, H. (2000). Radiochim. Acta, 88, 633–638.  Web of Science CrossRef CAS Google Scholar
First citationSchäfer, A., Huber, C. & Ahlrichs, R. (1994). J. Chem. Phys. 100, 5829–5835.  Google Scholar
First citationScheinost, A. C., Steudtner, R., Hübner, R., Weiss, S. & Bok, F. (2016). Environ. Sci. Technol. 50, 10413–10420.  Web of Science CrossRef CAS PubMed Google Scholar
First citationSjoblom, R. & Hindman, J. C. (1951). J. Am. Chem. Soc. 73, 1744–1751.  CrossRef CAS Web of Science Google Scholar
First citationSoderholm, L., Antonio, M. R., Williams, C. & Wasserman, S. R. (1999). Anal. Chem. 71, 4622–4628.  Web of Science CrossRef CAS Google Scholar
First citationSornein, M. O., Mendes, M., Cannes, C., Le Naour, C., Nockemann, P., Van Hecke, K., Van Meervelt, L., Berthet, J. C. & Hennig, C. (2009). Polyhedron, 28, 1281–1286.  Web of Science CSD CrossRef CAS Google Scholar
First citationSullivan, J. C., Gordon, S., Cohen, D., Mulac, W. & Schmidt, K. H. (1976). J. Phys. Chem. 80, 1684–1686.  CrossRef CAS Web of Science Google Scholar
First citationTakao, K., Takao, S., Scheinost, A. C., Bernhard, G. & Hennig, C. (2009). Inorg. Chem. 48, 8803–8810.  Web of Science CrossRef PubMed CAS Google Scholar
First citationTaylor, R. J., Koltunov, V. S., Savilova, O. A., Zhuravleva, G. I., Denniss, I. S. & Wallwork, A. L. (1998). J. Alloys Compd. 271–273, 817–820.  Web of Science CrossRef CAS Google Scholar
First citationVila, F. D., Lindahl, V. E. & Rehr, J. J. (2012). Phys. Rev. B, 85, 024303.  Web of Science CrossRef Google Scholar
First citationWilliams, C. W., Blaudeau, J. P., Sullivan, J. C., Antonio, M. R., Bursten, B. & Soderholm, L. (2001). J. Am. Chem. Soc. 123, 4346–4347.  Web of Science CrossRef PubMed CAS Google Scholar
First citationZielen, A. J., Sullivan, J. C. & Cohen, D. (1958). J. Inorg. Nucl. Chem. 7, 378–383.  CrossRef CAS Web of Science Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775
Follow J. Synchrotron Rad.
Sign up for e-alerts
Follow J. Synchrotron Rad. on Twitter
Follow us on facebook
Sign up for RSS feeds