This protocol presents the preparation of U2O5 thin films obtained in situ under ultra-high vacuum. The process involves oxidation and reduction of UO2 films with atomic oxygen and atomic hydrogen, respectively.
We describe a method to produce U2O5 films in situ using the Labstation, a modular machine developed at JRC Karlsruhe. The Labstation, an essential part of the Properties of Actinides under Extreme Conditions laboratory (PAMEC), allows the preparation of films and studies of sample surfaces using surface analytical techniques such as X-ray and ultra-violet photoemission spectroscopy (XPS and UPS, respectively). All studies are made in situ, and the films, transferred under ultra-high vacuum from their preparation to an analyses chamber, are never in contact with the atmosphere. Initially, a film of UO2 is prepared by direct current (DC) sputter deposition on a gold (Au) foil then oxidized by atomic oxygen to produce a UO3 film. This latter is then reduced with atomic hydrogen to U2O5. Analyses are performed after each step involving oxidation and reduction, using high-resolution photoelectron spectroscopy to examine the oxidation state of uranium. Indeed, the oxidation and reduction times and corresponding temperature of the substrate during this process have severe effects on the resulting oxidation state of the uranium. Stopping the reduction of UO3 to U2O5 with single U(V) is quite challenging; first, uranium-oxygen systems exist in numerous intermediate phases. Second, differentiation of the oxidation states of uranium is mainly based on satellite peaks, whose intensity peaks are weak. Also, experimenters should be aware that X-ray spectroscopy (XPS) is a technique with an atomic sensitivity of 1% to 5%. Thus, it is important to obtain a complete picture of the uranium oxidation state with the entire spectra obtained on U4f, O1s, and the valence band (VB). Programs used in the Labstation include a linear transfer program developed by an outside company (see Table of Materials) as well as data acquisition and sputter source programs, both developed in-house.
Uranium oxide is the main component of nuclear waste, and its solubility in water is linked to uranium oxidation state, increasing from U(IV) to U(VI). Thus, UO2+x oxidation during geological storage is an important and crucial safety issue1,2. This motivates studies of reaction mechanisms governing the surface interactions between uranium oxides and the environment3,4,5,6. This knowledge is essential to all aspects of treating waste from nuclear fuel cycles.
While tetravalent and hexavalent uranium are well-established and common as solid-state systems, this is not the case for pentavalent uranium, despite its stability in uranyl complexes and occurrence in aqueous solution. In uranium oxides, U(V) is considered a metastable intermediate, and it is not reported as single-state but rather as coexisting with U(IV) and U(VI) species. For this reason, nothing has been reported about the chemical and physical properties of U2O5. This is also due to a common feature of corrosion experiments, in which samples are exposed to a corrosive environment. This creates a steep gradient in oxidation states between the surface (exposed to the oxidants) and bulk of the sample. The change takes place within the analysis depth. Thus, different oxidation states are observed simultaneously, not because of mixed valence, but as an artefact of an incomplete reaction resulting in a heterogeneous layer. These two problems can be resolved by using thin films instead of bulk samples. Large numbers of diverse systems can be prepared with little starting material, and the surface-bulk gradient is avoided because there is no bulk.
The method reported here allows in situ preparation of a very thin layer (some tens of atomic layers deposited on an inert substrate) and analysis of its surface without contact with the atmosphere. This is one of the advantages of the Labstation (Figure 1), which is a modular machine composed of different chambers kept under dynamic ultra-high vacuum (UHV), reaching pressures of 10-9-10-11 mbar. Chambers are dedicated to preparation of the thin films, surface treatment (gas adsorptions), and characterisation by surface spectroscopies techniques [e.g., x-ray photoelectron spectroscopy (XPS), ultra-violet photoelectron spectroscopy (UPS), low energy electron diffraction spectroscopy (LEED)]. Samples are mounted on specific sample holders and transferred between different chambers through a linear transfer chamber using a transport wagon. All chambers are connected to this central chamber through a valve so they can be isolated at any moment (e.g., for gas filling or servicing). Recuperation of the sample holder/sample from the linear transfer chamber is accomplished by a transfer rod mounted on each chamber. The Labstation base system has been manufactured by an external company (see Table of Materials). Extensions and modifications have been added afterwards depending on experimental requirements, resulting in a unique equipment at JRC Karlsruhe. Extensions include the sputter source (a core element for thin film preparation), which has been developed in-house along with the sputter and data acquisition programs. The loading of the sample holder/sample from an ambient atmosphere to ultra-high vacuum is done via a load lock chamber specially designed to perform multiple sample handling and minimize the time to reaching the final pressure of about 10-8-10-9 mbar, thus limiting air contamination of the system. The Labstation is the result of years of experience and expertise in the surface science field at JRC Karlsruhe.
To pass from one chamber to another, the sample is mounted on a transport wagon driven by an external magnet, controlled by a computer program (Figure 2) and moving along the linear transfer chamber of about 7 m to predefined stop positions in front of the chambers.
Without a similar or close installation, the experiment might be difficult to reproduce. However, this installation contributes to the PAMEC laboratory that contributes to the open access program at JRC, in which external users are invited to submit proposals reviewed by a panel of international scientific experts. Their evaluation then allows users to access the infrastructure operated by the JRC. After requests and in the frame of collaborations, thin films can be prepared for external users for analyses and experiments performed outside JRC Karlsruhe.
In this report, we provide a detailed protocol of the growth of single-valence U2O5 thin films, obtained by successive steps involving oxidation and reduction of UO2 with atomic oxygen and atomic hydrogen, respectively. Unlike UO2 and UO2+x, direct deposition of U2O5 and UO3 films by DC sputtering cannot be done. Therefore, we first proceed to the deposition of a UO2 film, oxidize it into UO3 using atomic oxygen, then reduce it back to U2O5 with atomic hydrogen. The oxidation and reduction times and sample temperature during the process have effects on the result and are important to master. Correct composition was verified with high-resolution X-ray photoelectron spectroscopy, which provides direct and quantitative evidence for uranium 5f1 electronic configuration, as expected for the U(V).
1. Sample Holder Preparation
NOTE: Handling of the sample holder outside of the Labstation under an ambient atmosphere should be performed with gloves and clean tweezers.
2. Thin Film Preparation
NOTE: Uranium oxide thin films are prepared in situ by direct current (DC) sputtering using a uranium metal target and gas mixture of Ar (6 N) and O2 (4.5 N) partial pressure.
The identification of U(V) can be easily done by a characteristic energy of the shake-up satellite accompanying the characteristic U4f doublet. The binding energy at which appears the satellite, associated with intrinsic energy loss processes, depends on the uranium oxidation state.
Uranium 4f core level X-ray photoemission spectra are recorded for U(IV) in UO2 (red curve), U(V) in U2O5 (green curve), and U(VI) in UO3 (pink curve), then compared to U(0) in uranium metal (black curve) in the left part of Figure 11. The corresponding O1s core level spectra are superposed and reported in the right part of Figure 11.
In the center part of Figure 11, the U4f7/2 core level peaks have been shifted to superpose the main lines (upper half), allowing visualization of the energy separation (ΔE) between the satellite and main line (lower half). With increasing oxidation state, the energy separation increases, while satellite intensity decreases. Spectra were obtained on thin films of about 20 monolayers in thickness. The satellite energy peak and the 4f5/2 (4f7/2) emission line were used as a fingerprint for the oxidation state of the uranium atoms. The valence band spectra of UO2, U2O5, and UO3 obtained on the same films are reported in Figure 12.
The spectra described in the protocol are corresponding to UO2 films (Figure 5) obtained after deposition in the preparation chamber B2. This film is then oxidized with atomic oxygen. Depending on the oxidation time, the result can be UO2+x (as reported in Figure 6) or UO3 (as reported in Figure 7). Also, if atomic reduction with hydrogen on UO3 is too long, it will return to UO2 as reported in Figure 8. In this case, reoxidation to UO3 as reported in Figure 9 should take place before reducing it again with an appropriate time to get U2O5, as displayed in Figure 10. The results show that the oxidation and reduction processes are completely reversible.
Figure 1: Photograph and schematic of the Labstation machine developed at JRC Karlsruhe to enable in situ surface science studies. Please click here to view a larger version of this figure.
Figure 2: Screenshot of the linear transfer control program. The program enables transfer of the wagon carrying the samples (I to V) along the linear transfer chamber at different chamber positions. Please click here to view a larger version of this figure.
Figure 3: Screenshot of the acquisition program. Once the measurement conditions are introduced, a series of measurements can be performed automatically after switching on the X-ray generator. The sample position window enables positioning of the sample in the analyses chamber. The adjustment along x, y, and z can be done to optimize the intensity of the signal. Please click here to view a larger version of this figure.
Figure 4: Screenshot of the sputter control program. The sputtering conditions can be selected with this program developed in-house. Among the variables to be defined are the heating and working voltages of the filament and the voltages of up to two targets. Please click here to view a larger version of this figure.
Figure 5: U4f, O1s, and valence band spectra after deposition of a UO2 film, measured by high-resolution X-ray photoemission spectroscopy. The peak and satellite positions are characteristic of a UO2 sample. Please click here to view a larger version of this figure.
Figure 6: U4f, O1s, and valence band spectra after oxidation of UO2 with atomic oxygen, measured by high-resolution X-ray photoemission spectroscopy. The oxidation time is too short, thus the oxidation to UO3 is incomplete. The satellite and peak positions are characteristic of UO2+x and not of the UO3 reported in Figure 7. Please click here to view a larger version of this figure.
Figure 7: U4f, O1s, and valence band spectra measured after oxidation of UO2 film with atomic oxygen using high-resolution X-ray photoemission spectroscopy. The peak and satellite positions are characteristic of a UO3 sample. Please click here to view a larger version of this figure.
Figure 8: U4f, O1s, and valence band spectra measured after reduction of UO3 with atomic hydrogen. The reduction time is too long, thus the U2O5 is further reduced to UO2. The satellite and peak positions are characteristic of a UO2 and not the U2O5 sample reported in Figure 10. Please click here to view a larger version of this figure.
Figure 9: U4f, O1s, and valence band of the sample obtained in Figure 8 and re-oxidized with atomic oxygen to UO3. The satellite and peak positions are characteristic of a UO3 sample. The processes of reduction and oxidation are thus reversible. Please click here to view a larger version of this figure.
Figure 10: U4f, O1s, and valence band spectra after reduction of UO3 film with atomic hydrogen, measured by high-resolution X-ray photoemission spectroscopy. The peak and satellite positions are characteristic of a U2O5 sample. Please click here to view a larger version of this figure.
Figure 11: U4f and O1s core level X-ray photoemission spectra of U(IV) in UO2 (red curve), U(V) in U2O5 (green curve), and U(VI) in UO3 (pink curve), then compared to U(0) in uranium metal (black curve). Please click here to view a larger version of this figure.
Figure 12: Valence band spectra of U(IV) in UO2 (red curve), U(V) in U2O5 (black curve), and U(VI) in UO3 (pink curve). Please click here to view a larger version of this figure.
The initial results obtained on the thin films of U2O5 of about 30 monolayers (ML) in thickness, together with the corresponding core level spectroscopy obtained with high-resolution X-ray photoemission spectroscopy, have been reported in a previous publication7. The evolution of the uranium state during the oxidation process of UO2 into UO3 was reported through X-ray photoelectron spectra obtained on thin films of two to 50 layers in thickness in a wide range of the O:U ratio (Figure 11, Figure 12). Film oxidation and film reduction were obtained by exposing the films to atomic oxygen and atomic hydrogen, respectively. The homogeneity of the films with uranium oxidation states from IV to VI could be confirmed due to their small thickness and the reaction temperatures. Thin films of uranium oxides are deposited on a substrate using direct current sputtering with a sputter source developed at JRC Karlsruhe. The sputter source is installed in a chamber kept under ultra-high vacuum, like all the chambers of the Labstation. While UO2 can be obtained directly, UO3 and U2O5 films are only obtained after further treatment with atomic oxygen and atomic hydrogen. The binding energy of the main peaks and their satellites positions allow differentiation between the oxidation states of uranium in uranium oxide films produced in situ. High-resolution spectroscopy is necessary to differentiate the different oxidation states, as the satellite binding energies are close and have low intensities.
In 1948, pure pentavalent uranium, U2O5, was identified for the first time8. Later, its synthesis was described based on high temperature (673-1,073 K) and high pressure (30-60 kbar) of a mixture of UO2 and U3O89. However, the existence and stability of U2O5 at ambient temperature and pressure conditions have been questioned, suggesting a lower limit of x = 0.56-0.6 for the single-phase region below U3O810. So far, preparation of U2O5 at high pressure and temperature or during a thermo-reduction process was not reproducible; often, it was not possible to assign a single oxidation state to obtained samples. Some of a U2O5 bulk sample preparation appeared as mixtures of UO2 or UO3 with coexistence of U(V) with U(IV) or U(VI), as for U4O9 and U3O8. For instance, Teterin et al.11 reported the leaching process of U3O8 in sulphuric acid followed by thermal treatment in a helium atmosphere, claiming that the results were related to U2O5. This conclusion could be easily excluded due to a resulting two-peak structure in their XPS spectra. A mixture of U(V) and U(VI) species could explain the result, excluding the formation of a single U(V) oxidation state expected for U2O5.
Our method of preparation allows preparation of thin films of uranium oxide with single oxidation states of U(IV), U(VI), and U(V). The entire process of sample preparation takes place in situ within an instrument maintained at ultra-high vacuum. It was found that reduction of UO3 by atomic hydrogen does not proceed to UO2 but can be stopped at U(V). The time factor is very important as well as the temperature of the sample during the reduction process. With the high-resolution photoemission spectrometer, it was shown that a pure sample of U2O5 can be prepared in situ. Preparing thicker films should be a next step in looking at the crystallographic structure and bulk properties with ex situ techniques.
The authors have nothing to disclose.
The authors have no acknowledgements.
Name | Company | Catalog Number | Comments |
1ary dry scroll vacuum pump | Agilent | SH-100 | All chambers except B1 |
1ary pump | EDWARDS | nXDS10i 100/240V | B1 chamber |
Acetone | |||
Acquisition programme | Developed in-house | ||
Analyser | Specs | Phoibos 150 hemispherical | A4 chamber |
Argon | BASI | 6N | |
Atomic source | GenII plasma source | Tectra | B3 chamber |
Au foil | Goodfellow | ||
CasaXPS programme | CasaXPS | ||
Gauge 1ary vacuum | PFEIFFER | TPR 280 (2011/10) | All chambers |
Gauge 2ary vacuum | VACOM | ATMION ATS40C | All chambers |
Hydrogen gas | BASI | 6N | |
Ion gun source | Specs | IG10/35 | B1 chamber |
Linear transfer programme | Specs | Program delivered with the station | |
Origin programme | Origin | OriginPro 8.1SRO | |
Oxygen gas | 6N | ||
Sampler e-beam heater power supply | Specs | SH100 | B1 chamber |
Sampler resistance heater | Made in-house | power supply + Eurotherm | B3 chamber |
Sputtering programme | Developed in-house | ||
Stainless steal or Molybdenum substrate | in house | ||
Ta wire | Goodfellow | ||
turbo pump | PFEIFFER | TC 400 | All chambers |
Uranium target | in house | in house | Natural uranium target |
Vacuum gauge controller | VACOM | MVC-3 | All chambers |
X-ray source | Specs | XRC-1000 MF | Equipped with a monochromator |
Request permission to reuse the text or figures of this JoVE article
Request PermissionThis article has been published
Video Coming Soon
Copyright © 2025 MyJoVE Corporation. All rights reserved