The effect of fission-energy Xe ion irradiation on the structural integrity and dissolution of the CeO2 matrix

This work considers the effect of fission fragment damage on the structural integrity and dissolution of the CeO2 matrix in water, as a simulant for the UO2 matrix of spent nuclear fuel. For this purpose, thin films of CeO2 on Si substrates were produced and irradiated by 92 MeV 129Xe23+ ions to a fluence of 4.8 × 1015 ions/cm2 to simulate fission damage that occurs within nuclear fuels along with bulk CeO2 samples. The irradiated and unirradiated samples were characterised and a static batch dissolution experiment was conducted to study the effect of the induced irradiation damage on dissolution of the CeO2 matrix. Complex restructuring took place in the irradiated films and the irradiated samples showed an increase in the amount of dissolved cerium, as compared to the corresponding unirradiated samples. Secondary phases were also observed on the surface of the irradiated CeO2 films after the dissolution experiment.

This material is also proposed as a possible component in inert matrix fuels or as part of high-level nuclear waste forms.
The use of CeO 2 is justified by the facts that it has the same Fm-3m fluorite type structure with similar lattice parameter and cation radii as UO 2 (Table 1) and is considered to be the most appropriate inactive analogue which can serve to gain experience for further work on UO 2 . Table 1 Summary of lattice type, lattice parameter and cation radii for UO2 and CeO2.
Parameter UO 2 CeO 2 lattice type [23] Fm-3m fluorite structure Fm-3m fluorite structure lattice parameter (Å) 5.469 [24] 5.411 [25] crystal cation radius, r cr (Å) [26] 1.14 1.11 However, there are important differences that should be remembered. Uranium is an actinide and has six valence electrons, whereas Ce is a lanthanide and has only four valence electrons. Although, there are some similarities in chemical behaviour between actinides and lanthanides, there are no ideal chemical analogues among lanthanides for Th, Pa, U, Pu and Np [27]. Therefore, it is reasonable to expect that chemical behaviour of UO 2 and CeO 2 will be different. The surface of uranium dioxide tends to oxidise in air to UO 2+x (x ≤ 1) [28], implying that some of U 4+ converts to U +5 M A N U S C R I P T

A C C E P T E D ACCEPTED MANUSCRIPT
3 and U +6 . In contrast, in CeO 2 under air atmosphere trace amount of Ce 3+ tends to be present [1], leading to CeO 2-x composition. Unfortunately, the literature review did not reveal any information on dissolution mechanism of CeO 2 in water, but it is widely accepted that CeO 2 dissolves via reduction of Ce 4+ to Ce 3+ under air atmosphere, whereas UO 2 dissolves via oxidation of U 4+ to U 6+ [29][30]. Work by Ohno et al. [13], Iwase et al. [12] and Kumar et al. [10] showed that ion irradiation of CeO 2 results in an increased proportion of Ce 3+ ions, leading to CeO 2-x , whereas modelling work by Kinoshita et al. [11] showed that the fission tracks in UO 2 can cause several meta-stable configurations for hyperstoichiometric defect structures of UO 2+x . In addition, work by Sonoda et al. [31] showed that the diameter of ion tracks in UO 2 is much less sensitive to the electronic stopping values than in CeO 2 , which indicates that UO 2 has a higher kinetic recovery of the radiation damage than CeO 2 . Weber [32] reported that UO 2 has a better recovery of the radiation damage than CeO 2 : it was observed that UO 2 irradiated by alpha particles showed 12 % recovery of the lattice parameter compared to 10 % recovery for CeO 2 following almost two years of post-irradiation storage at room temperature. In addition, the thermal recovery study showed that complete recovery of the lattice parameter was observed by 500 °C for UO 2 and by 700 °C for CeO 2 .
Electrical properties of UO 2 and CeO 2 are also different. Stoichiometric UO 2 is a Mott-Hubbard insulator that converts to a p-type semiconductor UO 2+x due to oxygen incorporation during oxidation in air [33]. Close to stoichiometric UO 2 has the electrical conductivity values in the range 10 -3 -10 -4 S/cm at room temperature [34]. Stoichiometric CeO 2 is a dielectric [35] and tends to convert into CeO 2-x in air that is an oxygen deficient n-type semiconductor [36]. Polycrystalline thin film CeO 2 with close to stoichiometric ratio of Ce to O has the electrical conductivity values ~10 -10 S/cm at room temperature [37].
All these differences question the suitability of using CeO 2 as an UO 2 analogue. To explore this subject further, experimental work with CeO 2 samples was conducted and the obtained results were compared with the similar work on UO 2 samples by Matzke [38]. Matzke [38]

Sample production
The bulk samples of CeO 2 were obtained from Sigma-Aldrich in the form of fused pieces 3 -6 mm in size and 99.9 % purity on trace metal basis, as claimed by the supplier.
The CeO 2 thin films were grown by pulsed laser deposition (PLD) in a Neocera PLD system with a Lambda Physik KrF laser (λ = 248 nm) with pulse duration of 50 ns on three (001) oriented p-doped Si substrates with dimensions 10 × 10 × 0.5 mm, secured by silver paste onto the stainless-steel resistive heater at Imperial College London. The target for the PLD system was in-house made from CeO 2 powder (Sigma Aldrich, 99.9 % purity, < 5 µm grain size). X-ray diffraction was used to confirm that there was no change in structure from powder to pellet, both presenting a unit cell size a of 5.41 Å. Thin films were deposited from 20 mm diameter stoichometric CeO 2 target in an oxygen pressure of 100 mTorr. The substrate temperature (T s = 800 K) during deposition was controlled using a thermocouple embedded in the heater. The energy density of the laser spot (2 × 10 mm 2 ) was 1.5 J/cm 2 . From the sample thickness measured using a Dektak 11A, the film growth rate was estimated to be approximately 0.05 nm/pulse. The total number of pulses was 5000 with a repetition rate of 8 Hz. Once the ablation was over, the samples were then cooled down at a rate of 10 °C/min in an oxygen rich environment (760 Torr). The intention was to produce single crystal CeO 2 films in the (111) orientation to utilise the advantages that these samples can offer: idealised simplified system with one crystallographic orientation without grain boundaries and flat surface.
The thin films of CeO 2 were nominally of the same thickness, as they were deposited by the same number of laser pulses, the three samples produced had different colours. This is an indication that the thin films may have had different thicknesses.

Sample irradiation
To simulate the damage produced by fission fragments in nuclear fuel, the samples were irradiated

Sample characterisation
The orientation of the as-produced thin film samples was analysed using PANalytical X'Pert MRD diffractometer with X'Celerator detector.
A bulk sample of the as-supplied CeO 2 was powdered using mortar and pestle and analysed in Bragg-Brentano geometry on a D8 Bruker diffractometer equipped with a primary Ge monochromator for Cu Kα1 and a Sol-X solid state detector to verify identity of the sample and check for other phases.
In addition, the composition of two bulk samples was examined using a Cameca SX-100 electron microprobe analyser. Prior to the analysis, the samples were embedded in a resin, polished and carbon coated to ensure conductivity for the analysis. Calibration of the equipment was performed using a set of rare earth elements.
Surface morphology of the CeO 2 samples was studied using JEOL 820 SEM. No conductive coating was used for the thin film samples to preserve the surface for subsequent studies. A bulk sample of CeO 2 was gold-coated to improve surface conductivity. In addition, uncoated irradiated (to preserve the surface for further studies) and unirradiated bulk samples were studied on a FEI Quanta650F instrument operating at 2 kV with spot size 1 under high vacuum.

Dissolution experiment
Dissolution experiments were conducted to assess the effect of the xenon ion irradiation on the CeO 2 matrix dissolution in water. Table 2 summarises the set of CeO 2 samples used for the dissolution study.  Table 2 were rinsed with deionised water and pre-washed by placing into plastic bottles with ~10 ml of deionised water for a day. The Milli-Q water (18.2 MΩ/cm) was used throughout this experiment.
The aim of this approach was to remove fine CeO 2 particles on the surface of the samples that were observed in SEM (not shown), as they can affect Ce concentration measurements. After a day of prewashing, the pre-washed samples were rinsed with the deionised water and allowed to dry before they were placed into the leaching vessels filled with 4 ml of the deionised water. The leaching vessels consisted of a stainless steel casing with a tight lid on the thread and a PTFE (polytetrafluoroethylene) liner. In addition, two blank leaching vessels were prepared for reference purposes. The leaching vessels were placed in a heater set to 90 °C. The elevated temperature was used to facilitate dissolution, as it is known that CeO 2 is highly insoluble in water.

Sample characterisation
It is expected that the produced films have different thickness as they have different colours. Film thickness was not measured due to technical limitations but the target film thickness was 250 nm.
Crystallographic orientation of the as-produced thin films was examined by XRD (Fig. 1) Electron probe microanalysis was performed for two bulk CeO 2 samples to assess purity of the samples. The analysis indicated that Gd impurity was present at ~6 wt% and there were some traces of Sm, Eu and La. However, X-ray diffraction of a powdered bulk CeO 2 sample produced a diffraction pattern identical to the reference one and no Gd containing phase was observed, since Gd is soluble in CeO 2 up to ~36 wt% and does not change much the lattice parameter especially for low loadings [45].
The surface topography of the samples was studied using SEM. The fused pellet exhibited cracks at the grain boundaries, which indicate incomplete sintering or crack formation during the cooling stage (not shown), and sub-micron particles present at the surface (Fig. 2a). The ion irradiation caused formation of a wavy pattern on the surface of the irradiated bulk samples (Fig. 2b).    The presence of a maximum in the dissolution curves indicates that precipitation of cerium containing secondary phases is likely to take place -dissolution-precipitation behaviour is expected. The concentration values of Ce in the pre-wash solutions were 19, 1.2 and 0.7 times the values after one day of leaching for thin film CeO 2 samples Ce-AP1, Ce-AP2* and Ce-AP3*, respectively. Gadolinium ions were also detected in solutions (up to 2.5 × 10 -9 mol/l for sample Ce-AP2*), indicating that the stock CeO 2 powder used for production of the thin films had also gadolinium as an impurity. Again, the irradiated bulk sample showed higher Ce concentration in water than the unirradiated bulk samples -14 times higher for irradiated sample Ce-AP5*, as compared to unirradiated sample Ce-SLS3 on the 27 th day of leaching. Again, this indicates that the Xe ion irradiation increased CeO 2 dissolution in water. Irradiated bulk sample Ce-AP5* showed a dissolution curve with different gradients that were positive for all days of leaching. This indicates that the initial dissolution mechanism was altered and that equilibrium was not attained in the system. The dissolution mechanism can be altered by secondary phases precipitating at the surface of the sample and limiting the access of the sample's surface to water or it might be the case that the next stage of the dissolution is dominated by the dissolution of these secondary phases. samples had loose cerium containing fine particles, which is consistent with the SEM study (Fig. 2a).

Dissolution results
The Ce concentration values in acid wash solutions were ~200 times higher than the values on the last day of leaching for CeO 2 bulk samples Ce-SLS1 and Ce-SLS2. Again, gadolinium ions were detected in solutions (~10 -11 mol/l) supporting the electron microprobe results.
The pH values of the solutions were measured at the end of leaching once the samples were removed and were in the range 5.1 -5.6. The pH value of the deionised water used for the leaching was slightly acidic (pH = 5.7), more likely, due to absorption of atmospheric CO 2 . Hence, dissolution of CeO 2 resulted in a decrease of the pH values, as compared to the value of the deionised water.
The most pronounced change in the pH value was observed for irradiated bulk sample Ce-AP5* (pH = 5.1), as compared to the value of the deionised water.

Post-dissolution results
Unirradiated thin film sample Ce-AP1 after the dissolution experiment did not show any noticeable surface alternations, as indicated by SEM study (not shown). SEM image of the surface of irradiated thin film sample Ce-AP2* after the dissolution experiment (Fig. 7) suggests that the holes, which were observed before the leaching experiment (Fig. 3), increased in size to ~30 µm and a secondary phase precipitated in the middle of the holes and between them. The surface topography of irradiated thin film sample Ce-AP3* after the dissolution experiment did not show any significant changes (Fig. 8). The small circular satellites around the larger features, observed in Fig. 4, disappeared, more likely, as a result of dissolution. Instead, rod-shape particles with a length of 0.2 µm and a width of 0.05 µm appeared attributed to cerium secondary phases.

Discussion
The radiation damage induced by Xe ion irradiation resulted in the increased cerium concentration values as was shown by the CeO 2 thin film and bulk samples. The effect is most likely caused by an increased proportion of Ce +3 ions in the CeO 2 matrix due to the Xe ion irradiation [10,[12][13]. Ce 3+ can be more easily removed from the CeO 2-x surface than Ce 4+ . In addition, the expected increase in Ce 3+ fraction should result in hypo-stoichiometry, CeO 2-x , to maintain the charge balance [10]. The electrical conductivity of CeO 2-x tends to increase with increase in x up to x = 0.1 [46]. Hence, the dissolution rate is also expected to increase [30]. However, the overall effect of the radiation damage on the electrical conductivity in CeO 2 can be more complex, as was the case for UO 2 (Fig. 2 in [34]).
Hence, the exact effect of the radiation damage on the electrical conductivity in CeO 2 remains unknown. Incorporation of Si from the silicon substrate into the CeO 2 lattice to form a substitutional solid solution, as a result of the irradiation induced mixing, is unlikely, as a Si 4+ ion is half the size of a Ce 4+ ion [26]. Hence, stabilisation or distortion of the CeO 2 lattice by the substrate Si is not expected.
The dissolution data from the thin films should be treated with some caution. The difference in the film thickness might imply that thinner samples have more Ce 3+ on the surface that is more soluble than Ce 4+ . Hence, the effect of the radiation damage enhanced dissolution might interfere with the enhanced dissolution caused by a higher proportion of Ce 3+ ions due to smaller film thickness.
Horlait et al. [5] showed that incorporation of Gd into CeO 2 results in an increased dissolution rate of The pre-wash concentration results show the significance of a more soluble material at the surface of the samples that dissolves in the first instance. An initial burst of leaching is a very common phenomenon that is observed for many materials, although the reasons for this (surface defects, surface oxidation/reduction?) are still unclear.
The observed decrease of the pH values as a result of the CeO 2 dissolution, which correlates with the extent of dissolution, can be potentially explained in terms of the hydrolysis of Ce 4+ and Ce 3+ as suggested by Hayes at al. [47]: The acid leaching of the vessels, in which CeO 2 dissolution took place, indicates that significant precipitation of the secondary Ce containing phases was taking place on the walls of the liners and on the surface of the samples. The extensive secondary phase precipitation in this dissolution experiment is facilitated by two mechanisms: 1) the intrinsic dissolution-precipitation behaviour of CeO 2 in water that is indicated by the presence of a maximum in the dissolution curves; 2) the cooling of solution from 90 °C to an ambient temperature of ~20 -25 °C that might result in reaching the saturation concentration of dissolved cerium. Further work is required to identify these secondary phases.
The observed microstructural response in the CeO 2 thin films to the ion irradiation is a microscale cumulative effect of the irradiation damage. The difference in microstructural response of thin film Ce-AP2 and Ce-AP3, more likely, can be attributed to different thickness of the films.
Despite the differences between CeO 2 and UO 2 outlined in the introduction section, the observed dissolution response for the irradiated CeO 2 thin film and bulk samples is in qualitative agreement with work by Matzke [38], where it was observed that the leach rate of the ion irradiated UO 2 and UO 2based simfuel samples increased compared the corresponding unirradiated samples. For a better comparison, thin films of CeO 2 and UO 2 could be produced with the same thickness on the same substrates and irradiated under the same conditions along with the polished bulk samples and subsequently characterised to reveal any differences in structural and chemical responses to the irradiation damage.

Conclusions
It was observed that the high energy, high fluence ion irradiation resulted in significant microstructural rearrangements of the CeO 2 thin films. It was also suggested that the microstructural rearrangement due to an ion irradiation depends on thickness of the film being irradiated.
An increase in the measured Ce concentration values for the irradiated bulk and thin film CeO 2 samples was observed, as compared to the unirradiated samples. This observation is in qualitative M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT agreement with work by Matzke [38], where an increase in the leach rate of the UO 2 matrix was observed as a result of radiation damage by ion irradiations.
Secondary phases were observed on the surface of the irradiated thin film samples after the dissolution experiment.

Supporting data
Supporting data will be available in A.J. Popel's PhD thesis (University of Cambridge) published online.

M A N U S C R I P T
A C C E P T E D ACCEPTED MANUSCRIPT  Ion irradiation induced microstructural rearrangements in CeO 2 thin films.  Ion irradiation reduced aqueous durability of bulk and thin film CeO 2 samples.  Secondary phases observed from dissolution of irradiated CeO 2 films in di-water.