According to the equilibrium crystal shape of α-U, the surface of α-U (110) is more stable and has a larger area fraction than α-U (001). Therefore, the adsorption, dissociation, and diffusion behaviors of H atoms and H2 molecules on the α-U (110) surface were systematically studied by first-principles calculations. The results show that there are only two stable adsorption sites for the H atom: the surface short-bridge site (SB1) and the subsurface short-bridge site (SB2). The adsorption of H2 molecules is divided into chemical adsorption of dissociated H2 molecules and physical adsorption of undissociated H2 molecules, and the LB2-ParL adsorption configuration is the most stable adsorption configuration for H2 molecule adsorption, with an adsorption energy of −0.250 eV. The work function and charge transfer show that adsorption of the H atom or H2 molecule leads to an increase in the work function value of the α-U (110) surface, which enhances the electronic stability of the α-U (110) surface. The projected density of states shows that when the H atom or H2 molecule is close to the α-U (110) surface, the 1s orbital electrons of the H atom will hybridize with the 5f/6d orbital electrons of the nearby surface and subsurface U atoms, and new hybridized orbital peaks appear near the −4.5 eV or −7.3 eV energy level. The climbing image nudged elastic band study shows that the surface free H atoms are very easy to diffuse between the surface short-bridge sites and the subsurface short-bridge sites but the diffusion between the short-bridge site and the triangular center site is extremely difficult.

Uranium is an important and irreplaceable material in the nuclear industry for its excellent properties.1,2 In early nuclear reactors, it was used as metal fuel for quite a long time, and even now, many experimental reactors use metal or metal alloy uranium fuel.3 Uranium atoms have a special arrangement of valence electrons (5f36d17s2), in which the electrons in the 5f state easily form special covalent bonds, which ultimately leads to poor corrosion resistance of uranium.4,5 The corrosion of uranium not only greatly reduces the performance of uranium as a nuclear material but also has the potential to cause serious environmental pollution. Therefore, in the past decades, the study of corrosion prevention of uranium has been a very important research topic, and many researchers have invested a lot of energy in the study of corrosion prevention of uranium.6–9 Uranium has three crystal structures at different ambient temperatures:10 orthorhombic α phase (<940 K), body-centered tetragonal (bct) β-phase (941–1048 K), and body-centered cubic (bcc) γ-phase (>1048 K). Since uranium crystallizes in the orthorhombic α-phase at room temperature, it is of great scientific significance to study the corrosion mechanism of α-U surfaces.

The corrosion of uranium is mainly divided into oxidative corrosion and hydrogenation corrosion. Therefore, hydrogenation corrosion of uranium surface is one of the important processes in uranium surface corrosion, and it is a research direction that has been widely considered in the study of uranium surface corrosion.11–16 Therefore, an in-depth understanding of the interaction mechanism between the uranium surface and hydrogen is of great significance for the prevention of hydrogen corrosion of uranium.17 Many experimental studies have revealed the processes of surface corrosion of uranium in hydrogen atmosphere, including thermochemistry, surface diffusion, and nucleation sites.14,17–24 However, because the hydrogenation reaction on the surface of uranium is often masked by the more intense surface oxidation reaction, the resulting oxide layer further limits the study of uranium hydrogenation reaction. Therefore, experimentally, hydrogenated corrosion studies on uranium surfaces have progressed slower than oxidized corrosion studies, and the experimental means can only study the process of uranium–hydrogen corrosion rather than the detailed mechanism. The current experimental means cannot be truly meaningful from the atomic scale to study the corrosion process of the uranium surface, and theoretical research can overcome the weakness of experimental research in this respect.

With the continuous improvement and development of theoretical methods and the rapid development of computer technology, researchers have conducted a large number of theoretical and computational studies to understand the interactions between hydrogen and metal surfaces, which include Fe, Al, Zr, and some transition metal surfaces.25–28 Ferrin et al.29 used the self-consistent density functional theory (DFT) method to calculate the interaction between H and 17 kinds of transition metal surfaces, and the results showed that H prefers to interact at the surface rather than the subsurface in surface reactions. Balasubramanian et al.23 studied the transition from U to UH3 by using the atom-centered method and revealed the interaction between U and H atoms, but they ignored the two important steps of adsorption and dissociation of H2. Taylor et al.30,31 studied in detail the adsorption, dissociation, diffusion, capture, and saturation of H, as well as the phase transition of UH3 and the effect of interstitial impurities on the surface of U, which enhanced the understanding of U–H interaction. Yang et al.32 found that the interaction between the 1s electrons of hydrogen and the 6d electrons of U plays a very important role in the dissociation of H2 after studying the dissociation of H2 on the surface of γ-U (100). Nie et al.33 used the density functional theory approach to study the adsorption and dissociation reactions of H2 molecules on the α-U (001) surface in detail, and the study showed that the 5f orbital electrons of α-U would hybridize with the 1s orbital electrons of H during the reaction of H2 molecules with the U surface. Shi et al.34 systematically studied the adsorption, dissociation, and diffusion behavior of H on clean and Ti-doped α-U (001) surfaces using density functional theory methods, The study showed that Ti atoms contribute to the bond breaking of H2 molecules, hydrogen atoms may gather in the region near Ti atoms, and uranium hydrides may nucleate. Morrison and Ray35 studied the interaction between the H atom and the γ-U (100) surface and showed that the bridge site is most favorable for adsorption, followed by the center site and then the top site. Huang et al.36 compared the adsorption of the H atom and H2 molecule on the α-U (001) surface using DFT calculations, and the study showed that the top site and the short-bridge site are the most active sites for H atom adsorption, the long-bridge site is the most active site for the decomposition reaction of the H2 molecule, and the H atoms preferred the triangular central site as the final location.

Although many researchers have conducted a large number of calculations and theoretical studies on the interaction between hydrogen and uranium surfaces, and the main research focus is on γ-U and the (001) surface of α-U, there are relatively few studies on other surfaces, which has certain limitations for further understanding the corrosion mechanism of uranium. According to the equilibrium crystal shape of α-U, the surface of α-U (110) is more stable and has a larger area fraction than α-U (001). However, there is no comprehensive computational study on the molecular geometry, electronic structure, and energy of the interaction between hydrogen and the α-U (110) surface. Therefore, in this work, we systematically studied the adsorption, dissociation, and diffusion behaviors of the H atom and H2 molecule on the α-U (110) surface, which is of great significance for further understanding the hydrogenation and corrosion of U surfaces.

In this paper, the optimization of the α-U lattice parameters; the structural optimization of the H2 molecule, the α-U (110) surface, and the adsorption system; the work function and charge transfer; and the Climbing Image Nudged Elastic Band (CI-NEB) calculations are all carried out using the Vienna ab initio simulation package (VASP),37 and the projector augmented wave (PAW) method and the generalized gradient approximation (GGA) of the Perdew–Burke–Ernzerhof (PBE) exchange-correlation functional38 are employed to describe the exchange–correlation interactions. Within pseudopotential formalism, the core electrons of the nucleus are replaced by the equivalent potential field, while the valence electrons are treated explicitly. For U atoms, we consider the 6s26p67s25f36d1 electrons as valence electrons. In all calculations, the cut-off energy is set to 500 eV. The k-point grid for integration over the Brillouin zone is generated by the Monkhorst–Pack method.39 A 12 × 6 × 7 k-point grid is used for optimizing the α-U conventional cell and H2 molecule, while a 3 × 3 × 1 k-point grid is used for the α-U surface and the adsorption system. In ionic relaxation, the convergence criteria for the total energy and forces are set to 10−5 eV and 0.02 eV/Å, respectively. The H atom and H2 molecule are placed on one side of the slab, and the induced dipole moment is removed by applying a dipole correction.40 In order to investigate the diffusion properties of hydrogen atoms on the α-U (110) surface, the Climbing Image Nudged Elastic Band (CI-NEB) method41 is used to find the minimum energy paths (MEPs). In this method, a chain of linear interpolation images along an initial pathway between the given initial and final states of a reaction would be relaxed to determine the MEP and its corresponding saddle point, and the adjacent images are connected by a spring force, thus simulating an elastic band.42 The images were relaxed until the maximum residual forces on each atom are less than 0.02 eV/Å.43 

When studying the adsorption of surface H atoms and H2 molecules, to eliminate the self-interaction of surface atoms at the same position, resulting from the periodic reduplicative slab cell along the z direction,44 a vacuum layer of 16 Å is set to obtain a real and reliable surface model. From Fig. S1 of the supplementary material, it can be seen that both the surface energy and the work function converge when the U atomic layer is four layers, so a p(2 × 2) supercell with four layers of surface atoms, which has 32 atoms in total, is used to simulate the real surface. When ion relaxation is performed, the top three layers atoms of α-U are free in all three degrees of freedom, while the U atoms of the bottom layer are fixed. The lattice parameters (a = 2.799, b = 5.841, and c = 4.896) of the optimized α-U conventional cell are consistent with the experimental values (a = 2.836, b = 5.866, and c = 4.936).45 Besides, the bond length of the H2 molecule calculated using the GGA-PBE method is 0.75 Å, only 0.94% different from the experimental value of 0.743 Å.46 These results demonstrate the reliability of the GGA-PBE method.

In this paper, charge transfer is calculated by using Bader charge analysis47 in VASP, which is a method of analyzing the charge transfer of atoms when they are grouped together by dividing the atomic space using the zero-flux surface of the electron density as a dividing interface and solving for the charge carried by each atom. The charge transfer before and after the reaction can be obtained by subtracting the calculated atomic charge from the initial electron number of the atom. The charge density difference (CDD) is obtained by calculating the charge density of the adsorption system minus the charge density of the α-U (110) surface and the adsorbed H atom or H2 molecule. The work function is defined as the minimum energy barrier required to remove an electron from a solid into free space at 0 K. The work function can therefore be expressed as
Φ = V vac E f ,
(1)
where Ef is the Fermi energy and Vvac is the energy level of an electron that is at rest within a “few nanometers” outside the solid.48 
As shown in Table I and Fig. 1, we calculated the surface energies of 19 different surfaces of α-U with a maximum Miller index of 2 and constructed the equilibrium crystal shape of α-U. The equilibrium crystal structure was generated by the Wulff construction, which is based on the principle that the crystal endeavors to minimize its aggregate surface energy while maintaining a fixed volume. The Gibbs–Wulff theorem can be utilized to forecast the crystal morphology following the acquisition of the surface energy for the surface model at various Miller indices. The mathematical expression of Wulff construction is as follows:
E surf r = constants ,
(2)
where, for crystal surfaces with various Miller indices, r is the distance from the crystal center to the surface and Esurf is the surface energy. Under conditions of equilibrium, the Wulff construction can furnish not only the geometrical shape of a single crystal but also the area fraction f slab A . The area fraction of distinct faces can be obtained via the following equation:
f A slab = A slab slab A slab .
(3)
The calculation results showed that the surface energy of the (110) surface was 1.744 J/m2, which is the most stable surface of α-U. At the same time, the area fraction of the (110) surface reaches 15.8%, which is much higher than the 10.9% of the (001) surface, indicating that the (110) surface has a much larger surface proportion in α-U. The results are also consistent with those obtained by Mei et al.49 
TABLE I.

Calculated surface energies Esurf (J/m2) and area fractions of symmetrically distinct facets for α-U with a maximum Miller index of 2. Boldface denotes surface energy and area fraction of α-U (110) surface.

Surface Esurf Area fraction
0 0 1  1.753  10.9% 
0 1 0  1.962  3.7% 
0 1 1  2.098  0% 
0 1 2  2.139  0% 
0 2 1  1.821  21.5% 
1 0 0  2.040  0% 
1 0 1  1.917  0.8% 
1 0 2  1.855  1.7% 
1 1 0  1.744  15.8% 
1 1 1  1.851  3.2% 
1 1 2  1.796  23.4% 
1 2 0  1.972  1.8% 
1 2 1  1.925  7.4% 
1 2 2  1.922  2.2% 
2 0 1  2.007  0% 
2 1 0  1.905  0% 
2 1 1  1.941  0% 
2 1 2  1.890  2.2% 
2 2 1  1.820  5.5% 
Surface Esurf Area fraction
0 0 1  1.753  10.9% 
0 1 0  1.962  3.7% 
0 1 1  2.098  0% 
0 1 2  2.139  0% 
0 2 1  1.821  21.5% 
1 0 0  2.040  0% 
1 0 1  1.917  0.8% 
1 0 2  1.855  1.7% 
1 1 0  1.744  15.8% 
1 1 1  1.851  3.2% 
1 1 2  1.796  23.4% 
1 2 0  1.972  1.8% 
1 2 1  1.925  7.4% 
1 2 2  1.922  2.2% 
2 0 1  2.007  0% 
2 1 0  1.905  0% 
2 1 1  1.941  0% 
2 1 2  1.890  2.2% 
2 2 1  1.820  5.5% 
FIG. 1.

Equilibrium crystal shape of α-U.

FIG. 1.

Equilibrium crystal shape of α-U.

Close modal
As shown in Fig. 2, the α-U (110) surface has two different surface terminations. The distance between surface and subsurface atoms of surface termination 1 is relatively large, while the distance between surface and subsurface atoms of surface termination 2 is relatively small. The following formula was used to calculate the surface energies of the two different surface terminations of the α-U (110) surface:
E surf = E slab N E bulk 2 A .
(4)
FIG. 2.

Schematic diagram of two different surface terminations of the α-U (110) surface, where blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

FIG. 2.

Schematic diagram of two different surface terminations of the α-U (110) surface, where blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

Close modal

Eslab is the total energy of the relaxed slab, N is the total number of atoms in the slab model, Ebulk is the energy of per atom in the bulk α-U, and A is the surface area of the slab model.

Our calculation results show that the surface energy of surface termination 1 and surface termination 2 is 2.258 and 1.744 J/m2, respectively, which is in good agreement with the results achieved by Qu et al. (2.26 J/m2)44 and Mei et al. (1.77 J/m2).49 Surface termination 2 is more stable than surface termination 1, so surface termination 2 is used for our adsorption model. The formula for the adsorption energies is
E ads = E slab+adsorbates E slab E adsorbates ,
(5)
where Eslab+adsorbates is the total energy of the optimized adsorption system, Eslab is the total energy after the optimization of the slab model, and Eadsorbates is the total energy of the isolated H atom or H2 molecule. The energies of the isolated H atom and H2 molecule are calculated in a large cuboid box (13 × 14 × 15 Å3).

As shown in Fig. 3, six highly symmetric adsorption sites were identified in total on the α-U (110) surface for the adsorption of the H atom and H2 molecule, which are the surface top site (T1), surface long-bridge site (LB1), surface short-bridge site (SB1), subsurface top site (T2), subsurface long-bridge site (LB2), and subsurface short-bridge site (SB2). Three adsorption ways [ParS (parallel, pointing to the [010] direction), ParL (parallel, pointing to the [100] direction), and Ver (vertical)] were considered for the adsorption model of the H2 molecule.

FIG. 3.

Schematic diagram of the H atom’s and H2 molecule’s adsorption sites: Top view (left); side view (right) (T1: surface top site, T2: subsurface top site, SB1: surface short-bridge site, SB2: subsurface short-bridge site, LB1: surface long-bridge site, and LB2: subsurface long-bridge site). Yellow spheres represent different adsorption sites, where blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

FIG. 3.

Schematic diagram of the H atom’s and H2 molecule’s adsorption sites: Top view (left); side view (right) (T1: surface top site, T2: subsurface top site, SB1: surface short-bridge site, SB2: subsurface short-bridge site, LB1: surface long-bridge site, and LB2: subsurface long-bridge site). Yellow spheres represent different adsorption sites, where blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

Close modal

The adsorption behavior of the H atom on the α-U (110) surface is investigated in this section. Table II lists the adsorption energies and some structure parameters for the six adsorption sites considered in Sec. III A, and the structure diagram is shown in Table S1 of the supplementary material. Only two stable adsorption sites, namely, SB1 and SB2 sites, were found. After structural optimization, the H atoms adsorbed at the T1 and LB2 sites in the initial adsorption configuration eventually moved to the SB1 site, and the H atoms adsorbed at the T2 and LB1 sites eventually moved to the SB1 site. This is also confirmed by the same adsorption energy and similar structural parameters. The adsorption energies of the two stable adsorption sites are −2.477 eV (SB1) and −2.707 eV (SB2), indicating that they are chemical adsorption, as opposed to physical adsorption. Our calculated adsorption energy values do not differ much from those calculated by Huang et al. (−2.9 eV)36 for the adsorption of the H atom on the α-U (001) surface. Despite the large difference between the structures of the α-U (110) and (001) surface, the surface energies we calculated for these two surfaces according to Table I are almost the same, which means that the two surfaces are similarly active, so the adsorption energies are relatively close to each other.

TABLE II.

The adsorption energies and related structure parameters for the adsorption of the H atom on the α-U (110) surface. Eads is the adsorption energy, dH-U is the shortest bond length between the H atom and U atoms on the surface, and hH-U is the distance from the H atom to the first layer of U atoms. ΔZ12/d0, ΔZ23/d0, and ΔZ34/d0 represent the rumpling ratios of the first layer atomic spacing ΔZ12, the second layer atomic spacing ΔZ23, and the third layer atomic spacing ΔZ34, respectively, relative to the bulk value d0 after the adsorption reaction. Boldface denotes two stable adsorption sites.

Site Final site Eads (eV) dH-U (Å) hH-U (Å) ΔZ12/d0 (%) ΔZ23/d0 (%) ΔZ34/d0 (%)
T1  SB1  −2.477  2.145  1.353  −8.63  −6.90  −8.98 
LB1  SB2  −2.707  2.201  0.960  −7.41  −7.10  −8.40 
SB1  SB1  −2.477  2.146  1.356  −8.59  −6.91  −8.91 
T2  SB2  −2.707  2.206  0.970  −7.28  −7.13  −8.10 
LB2  SB1  −2.477  2.148  1.360  −8.71  −6.87  −8.96 
SB2  SB2  −2.707  2.206  0.970  −7.18  −7.17  −8.00 
Site Final site Eads (eV) dH-U (Å) hH-U (Å) ΔZ12/d0 (%) ΔZ23/d0 (%) ΔZ34/d0 (%)
T1  SB1  −2.477  2.145  1.353  −8.63  −6.90  −8.98 
LB1  SB2  −2.707  2.201  0.960  −7.41  −7.10  −8.40 
SB1  SB1  −2.477  2.146  1.356  −8.59  −6.91  −8.91 
T2  SB2  −2.707  2.206  0.970  −7.28  −7.13  −8.10 
LB2  SB1  −2.477  2.148  1.360  −8.71  −6.87  −8.96 
SB2  SB2  −2.707  2.206  0.970  −7.18  −7.17  −8.00 

Comparing the adsorption energies of the SB1 site and the SB2 site leads to the conclusion that the SB2 site is the most stable adsorption site for H atom adsorption. Table II and Table S2 of the supplementary material list the relevant parameters for surface relaxation, such as ΔZ12/d0 and ΔZ1, and the spacing between the first, second, and third layers of the α-U (110) surface is 0.497, 2.207, and 0.497 Å, respectively. The results show that the layer spacing of each layer of U atoms is reduced after adsorption of the H atom compared to the α-U (110) clean surface. This shows that the adsorption of the H atom causes some changes in the structure of the α-U (110) surface, but the intralayer rumpling of each layer of U atoms is small, so no surface reconstruction occurs. For all adsorption systems, the first layer spacing has larger shrinkage than the second layer spacing, which indicates that the interaction between the H atom and the surface layer of the U atom is stronger than that of the subsurface layer of the U atom, and the stronger interaction promotes the higher shrinkage for the first layer spacing.

The adsorption of the H atom on the α-U (110) surface is inevitably accompanied by interatomic charge transfer and redistribution, so the interaction between the H atom and the α-U (110) surface can be investigated by the charge transfer before and after adsorption. Table III lists the work function and charge transfer for all adsorption systems, which are similar to those listed in Table II. The adsorption systems that eventually move to SB1 and SB2 sites have a similar work function and charge transfer, respectively. Therefore, it is only necessary to study the two adsorption sites SB1 and SB2. In these two adsorption sites, the H atom obtains charge from U atoms and is negatively charged, and the charge obtained by H atoms is basically the same in all adsorption sites, ranging from −0.49e to −0.51e. Comparing the results obtained by Huang et al. (−0.27e to −0.28e),36 who studied the adsorption of the H atom on the α-U (001) surface, our results are almost twice as much as the results achieved by Huang et al., which indicates that the α-U(110) surface has stronger interactions with the H atom than the α-U (001) surface. In addition, compared with the results calculated (−0.102 to −0.137 eV) by Huda and Ray50 for the adsorption of H on the Pu (100) surface, the charge transfer of H adsorbed on the α-U (110) surface is also much larger than that on the surface of Pu. For all adsorption sites, the surface U atoms on the α-U (110) surface all lose charge, and the number of lost charges is larger than that on the clean surface; although the subsurface U atom obtains charges, the charge number obtained is smaller than that obtained by the clean surface; the third and fourth layer of U atoms gain and lose electrons, respectively, similar to the electron gain or loss situation on the clean surface. This shows that almost all of the charge acquired by the H atom comes from the surface and subsurface U atoms, and this phenomenon is also confirmed by the charge density difference diagram shown in Fig. 4. The region of charge density variation is localized mainly between the H atom and the first and second layers of U atoms.

TABLE III.

Work function and charge transfer for the clean surface and the H atom adsorption on the α-U (110) surface at different adsorption sites. Φ denotes the work function, and QH represents the charge of the H atom. Q1, Q2, Q3, and Q4 denote the total charge of the first, second, third, and fourth U atomic layers, respectively. A negative value represents charge gain, whereas a positive value corresponds to charge loss. Boldface denotes two stable adsorption sites.

Site Φ (eV) QH (e) Q1 (e) Q2 (e) Q3 (e) Q4 (e)
Clean surface  3.578    0.61  −0.62  −0.63  0.65 
T1  3.586  −0.51  0.94  −0.42  −0.56  0.55 
LB1  3.590  −0.50  0.84  −0.28  −0.65  0.59 
SB1  3.590  −0.51  0.95  −0.43  −0.56  0.55 
T2  3.586  −0.49  0.85  −0.30  −0.66  0.60 
LB2  3.590  −0.51  0.94  −0.43  −0.56  0.55 
SB2  3.585  −0.50  0.84  −0.29  −0.64  0.59 
Site Φ (eV) QH (e) Q1 (e) Q2 (e) Q3 (e) Q4 (e)
Clean surface  3.578    0.61  −0.62  −0.63  0.65 
T1  3.586  −0.51  0.94  −0.42  −0.56  0.55 
LB1  3.590  −0.50  0.84  −0.28  −0.65  0.59 
SB1  3.590  −0.51  0.95  −0.43  −0.56  0.55 
T2  3.586  −0.49  0.85  −0.30  −0.66  0.60 
LB2  3.590  −0.51  0.94  −0.43  −0.56  0.55 
SB2  3.585  −0.50  0.84  −0.29  −0.64  0.59 
FIG. 4.

Charge density difference diagram for two stable adsorption configurations of H atoms adsorbed on the α-U (110) surface. (a) and (b) The initial adsorption sites SB1 and SB2, respectively; the yellow spheres represent the H atom, the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively; and the cyan part and the yellow part represent the decrease and increase in the charge density, respectively.

FIG. 4.

Charge density difference diagram for two stable adsorption configurations of H atoms adsorbed on the α-U (110) surface. (a) and (b) The initial adsorption sites SB1 and SB2, respectively; the yellow spheres represent the H atom, the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively; and the cyan part and the yellow part represent the decrease and increase in the charge density, respectively.

Close modal

The work function values at SB1 and SB2 sites are 3.590 and 3.585 eV, respectively, and the work function of the clean surface we calculated is 3.578 eV, which indicates that the H atom adsorbed will enhance the electronic stability of the α-U (110) surface. It is worth noting that the charge obtained by the H atom at the SB1 site is slightly higher than that obtained at the SB2 site, and the Fermi energy level at the SB1 site (1.515 eV) is lower than that at the SB2 site (1.531 eV). The above-mentioned analysis can show that for the adsorption system of the H atom on the α-U (110) surface, more surface charge transfer between the H atom and the surface U atoms will lead to a decrease in the Fermi energy level and ultimately lead to a higher work function value.

Projected density of states (PDOS) analysis can provide the nature of the electronic interactions between the adsorbed substrate and the adsorbed small molecules. Therefore, in order to further understand the electronic interactions as well as bonding interactions between the H atom and U atomic layers during the adsorption process, the projected density of states (PDOS) of the H atom, clean α-U (110) surfaces, and two stable adsorption sites was calculated (Fig. 5). As previously discussed, the H atom only undergoes charge transfer with the surface and subsurface atoms, so the PDOS contains only the U atoms of the first two layers. It can be seen that after the adsorption of the H atom, the H1s orbital moves from near the Fermi energy level toward lower energy levels and hybridizes with the U5f and U6d orbitals near the −4.5 eV energy level, forming a new hybridized orbital peak, which indicates that both the U6d and U5f orbitals are involved in the bonding during the adsorption process of the H atoms. A similar conclusion was obtained by Huang et al.36 in their study of H atom adsorption on the α-U (001) surface, except that the position of the hybridized orbital peak was slightly different (−5.5 eV). From the intensity of the hybridized orbital peak, it can be seen that the hybridization of the U5f orbitals with the H1s orbitals is not obvious compared to that with the U6d orbitals, which indicates that the chemical activity of the U6d orbital electrons plays a leading role in the reaction between the surface U atoms and the H atom. Compared with the distribution of the U5f electronic states before and after adsorption, it can be seen that the width of the U5f electronic state does not change significantly after adsorption of the H atom, which indicates that the adsorption of the H atom does not affect the localization of the U5f electronic states.

FIG. 5.

PDOS of H atom adsorption on the α-U (110) surface: (a) H atom; (b) clean α-U (110) surface; (c) and (d) SB1 and SB2 sites, respectively. The dashed line shows where the Fermi level is.

FIG. 5.

PDOS of H atom adsorption on the α-U (110) surface: (a) H atom; (b) clean α-U (110) surface; (c) and (d) SB1 and SB2 sites, respectively. The dashed line shows where the Fermi level is.

Close modal

In this section, the adsorption and dissociation behavior of H2 molecules on the α-U (110) surface was mainly studied. As previously discussed in Sec. III A, the six high symmetry adsorption sites, three different ways of adsorption, and a total of 18 different adsorption configurations were considered for H2 molecules. Table S3 of the supplementary material and Table IV list the adsorption energies and some structural parameters of the H2 molecule on the α-U (110) surface. As can be seen from Table IV, except for the two adsorption configurations of SB1-ParL and SB2-ParL, the H2 molecules in the rest of the adsorption configurations were not completely dissociated, and the bond lengths of the H–H bonds ranged from 0.75 to 0.89 Å, with 1.4%–20.3% stretching relative to the experimental bond length of 0.74 Å of H2 molecules. It can be assumed that H2 molecules are only slightly stretched during the adsorption process without breaking the bond. The bond lengths of H–H bonds in the two adsorption configurations of SB1-ParL and SB2-ParL are 2.22 and 2.87 Å, respectively, which are almost 3–4 times the initial bond length of the H2 molecule. Therefore, it can be considered that the H–H bond in the H2 molecule is completely broken and H2 molecule is completely dissociated in these two adsorption configurations. As shown in Fig. 6, the two H atoms after dissociation will move to two different HOL sites (the triangle-center site of two surface U atoms and one subsurface U atom).

TABLE IV.

The final calculated adsorption energies and related structural parameters for the adsorption of H2 molecules on the α-U (110) surface. Eads is the adsorption energy; dH-H is the distance between the two H atoms; dH1-U is the shortest bond length between the first H atom and the surface U atom; dH2-U is the shortest bond length between the second H atom and the surface U atom; ΔZ12/d0, ΔZ23/d0, and ΔZ34/d0 represent the rumpling ratios of the first layer atomic spacing ΔZ12, the second layer atomic spacing ΔZ23, and the third layer atomic spacing ΔZ34, respectively, relative to the bulk value d0 after the adsorption reaction; and φ represents the angle between the H–H bond and the surface normal. Boldface denotes the adsorption configuration of the H2 molecule dissociation.

Configuration Eads (eV) dH-H (Å) dH1-U (Å) dH2-U (Å) ΔZ12/d0 (%) ΔZ23/d0 (%) ΔZ34/d0 (%) φ (deg.)
T1-Ver  −0.158  0.781  2.469  2.451  −8.40  −6.86  −9.17  87.25 
T1-ParS  −0.158  0.781  2.467  2.451  −8.37  −6.84  −9.27  88.67 
T1-ParL  −0.203  0.791  2.417  2.417  −8.02  −6.84  −9.34  90.00 
LB1-Ver  −0.034  0.755  3.942  4.573  −8.90  −6.87  −8.85  16.71 
LB1-ParS  0.013  0.765  2.965  3.155  −9.02  −6.77  −9.62  67.75 
LB1-ParL  −0.250  0.888  2.198  2.198  −9.27  −6.94  −7.92  90.00 
SB1-Ver  −0.037  0.757  3.383  4.059  −8.53  −6.92  −8.79  4.18 
SB1-ParS  −0.036  0.754  3.700  4.224  −8.81  −6.87  −8.79  26.06 
SB1-ParL  −1.028  2.215  2.200  2.200  −7.10  −7.48  −5.75  90.00 
T2-Ver  −0.032  0.755  3.656  4.411  −8.62  −6.92  −8.79  0.45 
T2-ParS  −0.125  0.859  2.248  2.264  −9.78  −6.83  −8.65  74.08 
T2-ParL  −0.249  0.885  2.207  2.207  −9.65  −6.88  −8.23  90.00 
LB2-Ver  −0.157  0.779  2.497  2.480  −8.11  −6.89  −9.14  86.77 
LB2-ParS  −0.157  0.779  2.489  2.474  −8.33  −6.9  −9.02  87.25 
LB2-ParL  −0.183  0.863  2.310  2.310  −8.08  −6.96  −8.23  90.00 
SB2-Ver  −0.031  0.756  3.825  4.505  −8.59  −6.92  −8.59  5.15 
SB2-ParS  −0.033  0.755  4.055  4.700  −8.70  −6.92  −8.60  20.15 
SB2-ParL  −1.057  2.872  2.235  2.235  −7.55  −7.36  −6.05  90.00 
Configuration Eads (eV) dH-H (Å) dH1-U (Å) dH2-U (Å) ΔZ12/d0 (%) ΔZ23/d0 (%) ΔZ34/d0 (%) φ (deg.)
T1-Ver  −0.158  0.781  2.469  2.451  −8.40  −6.86  −9.17  87.25 
T1-ParS  −0.158  0.781  2.467  2.451  −8.37  −6.84  −9.27  88.67 
T1-ParL  −0.203  0.791  2.417  2.417  −8.02  −6.84  −9.34  90.00 
LB1-Ver  −0.034  0.755  3.942  4.573  −8.90  −6.87  −8.85  16.71 
LB1-ParS  0.013  0.765  2.965  3.155  −9.02  −6.77  −9.62  67.75 
LB1-ParL  −0.250  0.888  2.198  2.198  −9.27  −6.94  −7.92  90.00 
SB1-Ver  −0.037  0.757  3.383  4.059  −8.53  −6.92  −8.79  4.18 
SB1-ParS  −0.036  0.754  3.700  4.224  −8.81  −6.87  −8.79  26.06 
SB1-ParL  −1.028  2.215  2.200  2.200  −7.10  −7.48  −5.75  90.00 
T2-Ver  −0.032  0.755  3.656  4.411  −8.62  −6.92  −8.79  0.45 
T2-ParS  −0.125  0.859  2.248  2.264  −9.78  −6.83  −8.65  74.08 
T2-ParL  −0.249  0.885  2.207  2.207  −9.65  −6.88  −8.23  90.00 
LB2-Ver  −0.157  0.779  2.497  2.480  −8.11  −6.89  −9.14  86.77 
LB2-ParS  −0.157  0.779  2.489  2.474  −8.33  −6.9  −9.02  87.25 
LB2-ParL  −0.183  0.863  2.310  2.310  −8.08  −6.96  −8.23  90.00 
SB2-Ver  −0.031  0.756  3.825  4.505  −8.59  −6.92  −8.59  5.15 
SB2-ParS  −0.033  0.755  4.055  4.700  −8.70  −6.92  −8.60  20.15 
SB2-ParL  −1.057  2.872  2.235  2.235  −7.55  −7.36  −6.05  90.00 
FIG. 6.

Top view of the adsorption configuration of H2 molecules adsorbed and dissociated on the α-U (110) surface. (a) and (b) The structure before and after adsorption; the yellow spheres represent H atoms, and the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

FIG. 6.

Top view of the adsorption configuration of H2 molecules adsorbed and dissociated on the α-U (110) surface. (a) and (b) The structure before and after adsorption; the yellow spheres represent H atoms, and the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

Close modal

As shown in Table IV, after adsorption of H2 molecules, there is a significant shrinkage of the layer spacing between each U atom layer compared to the clean α-U (110) surface, where the layer spacing is expressed by the average height difference of all U atoms in each layer. This phenomenon shows that the adsorption of the H atom causes some changes in the structure of the α-U (110) surface. From Table S3 of the supplementary material, it can be seen that after adsorption of the H2 molecule, the intralayer rumpling of each layer of U atoms is very small. Therefore, no surface reconstruction occurs. Comparison of the structural parameters after the adsorption of H atoms in Table II reveals that the shrinkage of the H2 molecular adsorption system is significantly higher than that of the H atom adsorption system, and this result indicates that the effect of the H2 molecule on the structure of the α-U (110) surface during the adsorption process is greater than that of the adsorption of the H atom. Similar to the H atom adsorption shown in Sec. III B, for almost all adsorption systems, the first layer spacing has much larger shrinkage than the second layer spacing, which indicates that the interaction between the H atom and the surface layer of the U atom is stronger than that of the subsurface layer of the U atom, and the stronger interaction promotes the higher shrinkage for the first layer spacing.

As shown in Table IV, for all H2 molecular adsorption configurations, the adsorption energy values range from 0.013 to −1.057 eV. Except for the two adsorption configurations, SB1-ParL and SB2-ParL, the adsorption energies of the other adsorption configurations are less than or equal to −0.250 eV. The results show that the two adsorption configurations of SB1-ParL and SB2-ParL are chemical adsorption with complete dissociation of H2 molecules, while the rest of the adsorption configurations are physical adsorption. Among all the adsorption configurations in which the H2 molecules are not dissociated, the most stable adsorption configuration is LB1-ParL, with an adsorption energy of −0.250 eV. In the study by Huang et al.36 only three adsorption configurations of H2 molecules were undissociated during adsorption on the α-U (001) surface, while the H2 molecules in the remaining nine adsorption configurations were completely dissociated, which indicates that the α-U (110) surface is more unfavorable for the dissociative adsorption of H2 molecules than that on the α-U (001) surface. At the same adsorption sites, the adsorption energies of H2 molecules adsorbed as ParL is the largest, and both dissociated adsorption configurations are also adsorbed as ParL. This phenomenon indicates that the adsorption mode of ParL is more favorable for adsorption and dissociation of H2 molecules on the α-U (110) surface.

As shown in Table S4 of the supplementary material, before and after adsorption, the H2 molecule will undergo certain position movements or rotations. As shown in Table IV, when the H2 molecule is placed in the vertical adsorption configuration Ver, the H2 molecules will rotate from 0.45° to 87.25° after structural optimization, and the angle of rotation varies from one adsorption site to another. In addition to the rotation of vertically adsorbed H2 molecules to the α-U (110) plane, we also observed that the adsorption configurations of H2 molecules adsorbed parallel to the α-U (110) surface rotate toward the normal direction of the α-U (110) surface, which occurs for both adsorptions at the short-bridge sites in the manner of ParS, that is, the two adsorption configurations of SB1-ParL and SB2-ParL, with angles of rotation of 63.94° and 69.85°, respectively. With the exception of the adsorption configurations in which the H2 molecules rotate and dissociate, the H2 molecules in the rest of the adsorption configurations undergo only a small amount of translation based on the initial position.

As shown in Table S3 of the supplementary material, the distances from H atoms to the first U atomic layer in the two adsorption configurations of SB1-ParL and SB2-ParL are 1.147 and 1.157 Å, respectively, which are significantly smaller than the rest of the adsorption configurations and are relatively close to the hH-U value shown in Table II. This suggests that similar to H atom adsorption, the dissociated H atoms in the two adsorption configurations of SB1-ParL and SB2-ParL strongly interact with the α-U (110) surface and ultimately form U–H bonds. The experimentally obtained U–H bond length in the UH3 crystals of the α-phase is 2.310 Å,51 and that of the β-phase is 2.300–3.710 Å.52 As shown in Table IV, the distances between the H atom and the nearest U atom on the surface are 2.200 and 2.235 Å in the two H2 molecule-dissociated adsorption configurations, respectively, and it can be seen that our calculations are close to the experimental values.

The work function values and charge transfer conditions of 18 adsorption configurations are listed in Table V. It can be seen that on the clean surface without adsorbed H2 molecules, the values of Q1 and Q4 are positive, while the values of Q2 and Q3 are negative. After adsorption of H2 molecules, the values of Q3 and Q4 change little, so the charge transfer mainly occurs between the first and second layer of U atoms and the H atom. It can be considered that H2 molecules mainly interact with the first and second layer of U atoms. This is further confirmed by the charge density difference diagram in Fig. 7. Regardless of adsorption configurations of dissociated H2 molecules or adsorption configurations of undissociated H2 molecules, the region of charge density variation is mainly concentrated between the H atom or H2 molecule and the top two layers of U atoms. The charge transfer between the H atom and the α-U (110) surface is relatively small in all adsorption configurations except for the two adsorption configurations with dissociated H2 molecules of SB1-ParL and SB2-ParL, in which the charge obtained by the H atoms ranges from −0.01e to 0.21e, which suggests once again that the interactions of the H2 molecules with the α-U (110) surface are weak and the adsorption mode is physical adsorption except the two H2 molecular dissociation adsorption configurations. In the adsorption configurations of SB1-ParL and SB2-ParL, the charge obtained by the H atoms is 0.48e and 0.5e, respectively, which is similar to the charge obtained by the H atom adsorbed on the α-U (110) surface in Table III. This further shows that the dissociated H atom in the two adsorption configurations of SB1-ParL and SB2-ParL interacts strongly with the α-U (110) surface. Comparing the results achieved by Huang et al. (−0.27e to −0.28e),36 who studied the adsorption of H2 molecules on the α-U (001) surface, for the same adsorption configurations of the dissociated H2 molecules, our results are almost double the results achieved by Huang et al., which once again illustrates the fact that the α-U (110) surface has a stronger interaction with the H atom than the α-U (001) surface.

TABLE V.

Work function and charge transfer for the clean surface and the H atom adsorption on the α-U (110) surface at different adsorption sites. Φ denotes the work function, QH1 represents the charge of the first H atom, and QH2 represents the charge of the second H atom. Q1, Q2, Q3, and Q4 denote the total charge of the first, second, third, and fourth U atomic layers, respectively. A negative value represents charge gain, whereas a positive value corresponds to charge loss. Boldface denotes the adsorption configuration of the H2 molecule dissociation.

Configuration Φ (eV) QH1 (e) QH2 (e) Q1 (e) Q2 (e) Q3 (e) Q4 (e)
Clean surface  3.578      0.61  −0.62  −0.63  0.65 
T1-Ver  3.582  −0.05  −0.03  0.65  −0.57  −0.60  0.60 
T1-ParS  3.581  −0.11  0.03  0.64  −0.56  −0.59  0.60 
T1-ParL  3.581  0.02  −0.12  0.65  −0.55  −0.60  0.59 
LB1-Ver  3.580  −0.01  −0.02  0.61  −0.55  −0.62  0.58 
LB1-ParS  3.585  0.01  −0.07  0.59  −0.52  −0.62  0.60 
LB1-ParL  3.590  −0.21  −0.09  0.75  −0.41  −0.59  0.55 
SB1-Ver  3.579  −0.10  0.06  0.64  −0.58  −0.62  0.60 
SB1-ParS  3.580  0.00  −0.02  0.64  −0.57  −0.63  0.59 
SB1-ParL  3.575  −0.48  −0.48  1.23  −0.21  −0.59  0.53 
T2-Ver  3.583  −0.06  0.03  0.61  −0.58  −0.59  0.60 
T2-ParS  3.587  −0.19  −0.06  0.74  −0.41  −0.62  0.55 
T2-ParL  3.588  −0.21  −0.08  0.77  −0.42  −0.60  0.54 
LB2-Ver  3.583  −0.09  0.01  0.65  −0.56  −0.60  0.59 
LB2-ParS  3.583  −0.01  −0.07  0.64  −0.57  −0.60  0.60 
LB2-ParL  3.582  −0.19  −0.06  0.73  −0.50  −0.56  0.56 
SB2-Ver  3.579  −0.03  −0.01  0.65  −0.59  −0.63  0.61 
SB2-ParS  3.579  −0.08  0.05  0.65  −0.57  −0.64  0.59 
SB2-ParL  3.586  −0.50  −0.50  1.27  −0.25  −0.57  0.54 
Configuration Φ (eV) QH1 (e) QH2 (e) Q1 (e) Q2 (e) Q3 (e) Q4 (e)
Clean surface  3.578      0.61  −0.62  −0.63  0.65 
T1-Ver  3.582  −0.05  −0.03  0.65  −0.57  −0.60  0.60 
T1-ParS  3.581  −0.11  0.03  0.64  −0.56  −0.59  0.60 
T1-ParL  3.581  0.02  −0.12  0.65  −0.55  −0.60  0.59 
LB1-Ver  3.580  −0.01  −0.02  0.61  −0.55  −0.62  0.58 
LB1-ParS  3.585  0.01  −0.07  0.59  −0.52  −0.62  0.60 
LB1-ParL  3.590  −0.21  −0.09  0.75  −0.41  −0.59  0.55 
SB1-Ver  3.579  −0.10  0.06  0.64  −0.58  −0.62  0.60 
SB1-ParS  3.580  0.00  −0.02  0.64  −0.57  −0.63  0.59 
SB1-ParL  3.575  −0.48  −0.48  1.23  −0.21  −0.59  0.53 
T2-Ver  3.583  −0.06  0.03  0.61  −0.58  −0.59  0.60 
T2-ParS  3.587  −0.19  −0.06  0.74  −0.41  −0.62  0.55 
T2-ParL  3.588  −0.21  −0.08  0.77  −0.42  −0.60  0.54 
LB2-Ver  3.583  −0.09  0.01  0.65  −0.56  −0.60  0.59 
LB2-ParS  3.583  −0.01  −0.07  0.64  −0.57  −0.60  0.60 
LB2-ParL  3.582  −0.19  −0.06  0.73  −0.50  −0.56  0.56 
SB2-Ver  3.579  −0.03  −0.01  0.65  −0.59  −0.63  0.61 
SB2-ParS  3.579  −0.08  0.05  0.65  −0.57  −0.64  0.59 
SB2-ParL  3.586  −0.50  −0.50  1.27  −0.25  −0.57  0.54 
FIG. 7.

Charge density difference diagram of the H2 molecules adsorbed on the α-U (110) surface. (a) The most stable adsorption configuration LB1-ParL among the undissociated H2 adsorption configurations; (b) and (c) the adsorption configurations SB1-ParL and SB2-ParL of H2 molecular dissociation, respectively. The yellow spheres represent H atoms; the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively; and the cyan part and the yellow part represent the decrease and increase in charge density, respectively.

FIG. 7.

Charge density difference diagram of the H2 molecules adsorbed on the α-U (110) surface. (a) The most stable adsorption configuration LB1-ParL among the undissociated H2 adsorption configurations; (b) and (c) the adsorption configurations SB1-ParL and SB2-ParL of H2 molecular dissociation, respectively. The yellow spheres represent H atoms; the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively; and the cyan part and the yellow part represent the decrease and increase in charge density, respectively.

Close modal

As shown in Table V, the work function values of all H2 molecular adsorption configurations range from 3.575 to 3.590 eV, among which the work function values of SB1-ParL and SB2-ParL adsorption configurations are 3.575 and 3.586 eV, respectively. We calculated the work function value of 3.578 eV for the clean surface, which indicates that the adsorption of H2 molecules enhances the electron stability of the α-U (110) surface for all adsorption configurations except for the SB1-ParL adsorption configuration (work function value of 3.575 eV). We find that for the undissociated adsorption configuration of the H2 molecule, more charge transfer between the H2 molecule and U atom tends to lead to higher work function values, which is similar to the phenomenon of H atom adsorption.

As shown in Fig. 8, in order to further investigate the electronic interactions and bonding between the H2 molecule and the U atoms on the α-U (110) surface during adsorption, we calculated the projected density of states (PDOS) of the H2 molecule, the α-U (110) clean surface, and the three adsorption configurations, where (d) and (e) are the adsorption configurations of H2 molecular dissociation SB1-ParL and SB2-ParL, respectively, Since the adsorption of H2 molecules is divided into chemical adsorption of dissociated H2 molecules and physical adsorption of undissociated H2 molecules, we also calculated the most stable adsorption configuration of H2 molecular undissociated LB1-ParL, as a comparison. Similar to the PDOS of the H atom adsorbed as shown in Sec. III B, our PDOS still contains only the first two layers of U atoms, and it can be seen that there are a large number of electron orbitals near the Fermi energy level, indicating that the U atoms on the α-U (110) surface have strong metallic properties.

FIG. 8.

PDOS of H2 molecule adsorption on the α-U (110) surface: (a) H2 molecule; (b) clean α-U (110) surface; (c) the most stable adsorption configuration LB1-ParL among the undissociated H2 adsorption configurations; (d) and (e) the adsorption configurations SB1-ParL and SB2-ParL of H2 molecular dissociation, respectively. The dashed line shows where the Fermi level is.

FIG. 8.

PDOS of H2 molecule adsorption on the α-U (110) surface: (a) H2 molecule; (b) clean α-U (110) surface; (c) the most stable adsorption configuration LB1-ParL among the undissociated H2 adsorption configurations; (d) and (e) the adsorption configurations SB1-ParL and SB2-ParL of H2 molecular dissociation, respectively. The dashed line shows where the Fermi level is.

Close modal

It can be seen from Figs. 8(d) and 8(e) that the PDOS of the H2 molecule dissociated is similar to the PDOS of the H atom adsorbed as shown in Sec. III B, in which there are new 5f and 6d orbital electron peaks of the U atoms near the energy level of about −4.5 eV and they overlap with the 1s orbitals of the H atom. The difference is that the SB1-ParL adsorption configuration has two orbital electron peaks, while the SB2-ParL adsorption configuration has only one. This phenomenon is also similar to that found by Huang et al.36 when studying the adsorption of H2 molecules on the α-U (001) surface, except that the number and positions of hybridized orbital peaks are different. This indicates that the H1s orbital electrons hybridize with the U6d and U5f orbital electrons and form U–H bonds after dissociation of the H2 molecule. From the intensities of the hybridized orbital peaks, it can be seen that the U5f orbitals do not have a significant hybridization with the H1s orbitals compared to the U6d orbitals, but it overlaps with both H1s and U6d electron states, which indicates that both U6d and U5f orbital electrons are involved in bonding and the chemical activity of the U6d orbital electrons dominates the interactions of the surface U atoms with the H atom. This is similar to the conclusion obtained by adsorption of the H atom in Sec. III B. In the adsorption configuration of the undissociated H2 molecule [Fig. 8(c)], the 1s orbital of the H2 molecule moves toward a lower energy level and hybridizes with the U5d and U5f orbitals near −7.3 eV, but the intensities of the hybrid orbital peaks of U5d and U5f are much smaller than the adsorption of the H atom, which further indicates that the adsorption of the H2 molecule on the α-U (110) surface is weak physical adsorption. Compared with the distribution of the U5f electronic states before and after adsorption, it can be seen that the width of the U5f electronic state does not change significantly after adsorption of the H2 molecule, which indicates that the adsorption of the H2 molecule does not affect the localization of the U5f electronic states, which is also consistent with the electronic properties of hydrogen adsorbed on the Am (001) surface.53 

As shown in Fig. 9, since there are only two stable adsorption sites (SB1 and SB2) for H atom adsorption, the two H atoms after adsorption and dissociation of the H2 molecule will occupy two HOL sites, which are, respectively, recorded as the HOL1 site and HOL2 site. Therefore, in order to study the diffusion behavior of the H atom on the α-U (110) surface, four stable adsorption sites—SB1, SB2, HOL1, and HOL2—were selected. Meanwhile, a total of six different surface diffusion paths were set up based on these four adsorption sites: path1 (transfer from the SB1 site to the SB2 site), path2 (transfer from the SB2 site to the HOL2 site), path3 (transfer from the HOL2 site to the HOL1 site), path4 (transfer from the HOL1 site to the SB1 site), path5 (transfer from the SB1 site to the HOL2 site), and path6 (transfer from the SB2 site to the HOL1 site). As can be seen from the results in Fig. 10(a), the energy barriers of the three diffusion paths path1–path3 are 0.069, 0.767, and 0.446 eV, respectively. This shows that H atoms only need to overcome a small energy barrier to move from SB1 to SB2 sites, which indicates that H atoms are still able to diffuse rapidly between the SB1 and SB2 sites, even under room temperature conditions. The diffusion energy barrier of H atoms between two HOL sites is 0.446 eV, which is much more difficult than the diffusion between two short-bridge sites.

FIG. 9.

Schematic diagram of six kinds of diffusion paths of H atoms on the α-U (110) surface. The yellow spheres represent H atoms, and the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

FIG. 9.

Schematic diagram of six kinds of diffusion paths of H atoms on the α-U (110) surface. The yellow spheres represent H atoms, and the blue, green, and gray spheres represent the first, second, and third and fourth layers of uranium atoms, respectively.

Close modal
FIG. 10.

Energy profile for the H atom diffusing on the α-U (110) surface. (a) Path1–path3; (b) path4–path6.

FIG. 10.

Energy profile for the H atom diffusing on the α-U (110) surface. (a) Path1–path3; (b) path4–path6.

Close modal

Since path 4–path6 are further away than path1–path3, in order to ensure consistency and accuracy of the results, as shown in Fig. 10(b), we inserted more images for path4–path6. The energy barriers of the three diffusion paths path4–path6 are 4.708, 4.152, and 4.639 eV, respectively. Combined with the energy barrier of path2 shown in Fig. 10(a), it is 0.767 eV. Such high energy barriers indicate that diffusion between the two short-bridging sites and the two triangular center sites is extremely difficult. Therefore, the diffusion of H atoms on the α-U (110) surface mainly occurs between the two sites of SB1 and SB2.

In this paper, the adsorption, dissociation, and diffusion behaviors of H atoms and H2 molecules on the α-U (110) surface are systematically studied by first-principles calculations. The results show that the adsorption energies range from −2.477 to −2.707 eV for the H atom adsorption system and from 0.013 to −0.250 eV for the H2 molecule adsorption system. There are only two stable adsorption sites for the H atom on the α-U (110) surface: surface short-bridge site (SB1) and subsurface short-bridge site (SB2), in which the subsurface short-bridge site (SB2) is the most stable adsorption site for the H atom. The adsorption of H2 molecules is divided into chemical adsorption of dissociated H2 molecules and physical adsorption of undissociated H2 molecules. Only two adsorption configurations of SB1-ParL and SB2-ParL are chemical adsorption of dissociated H2 molecules, and the LB2-ParL adsorption configuration is the most stable adsorption configuration for H2 molecule adsorption.

The work function and charge transfer show that H atoms and H2 molecules accept electrons from the U atoms of the surface and subsurface. In the undissociated adsorption configuration of the H2 molecule, the charge transfer between the H2 molecule and the nearby U atom is relatively small. The adsorption of H atoms or H2 molecules leads to an increase in the work function value of the α-U (110) surface, which in turn enhances the electronic stability of the α-U (110) surface.

The PDOS of the H atom and H2 molecule adsorption system shows that when the H atom or H2 molecule is close to the α-U (110) surface, the 1s orbital electrons of the H atom or H2 molecule will move to a lower energy level and hybridize with the 5f/6d orbital electrons of U atoms in the nearby surface and subsurface layers, forming a new hybridized orbital peak, which is formed near −4.5 eV for H atom adsorption or H2 molecule dissociative adsorption and near −7.3 eV for H2 molecule undissociative adsorption. The adsorption of the H atom or H2 molecule does not affect the localization of the U5f electronic states.

The CI-NEB study shows that the surface-free H atoms are very easily diffused between the surface short-bridge sites (SB1) and the subsurface short-bridge sites (SB2). The calculated diffusion energy barrier is only 0.069 eV. However, the diffusion energy barrier between the short-bridge sites and the triangular center sites (HOL) ranges from 0.767 to 4.708 eV, which indicates that the diffusion of H atoms between the short-bridge site and the triangular center site is extremely difficult.

See the supplementary material for the adsorption structure diagram and some structural parameters.

This work was supported by the National Natural Science Foundation of China (Grant No. 12304274) and the Sichuan University Postdoctoral Interdisciplinary Innovation Fund.

The authors have no conflicts to disclose.

Zihan Xu: Writing – original draft (lead); Writing – review & editing (lead). Chenglong Qin: Methodology (equal). Yushu Yu: Investigation (equal). Gang Jiang: Conceptualization (equal). Liang Zhao: Project administration (equal); Supervision (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
Q. G.
Sheng
,
B. R.
Cooper
, and
S. P.
Lim
, “
First-principle study of hybridization effects and magnetic ordering in correlated-electron uranium systems
,”
Phys. Rev. B
50
(
13
),
9215
(
1994
).
2.
R.
Springell
,
B.
Detlefs
,
G. H.
Lander
,
R. C. C.
Ward
,
R. A.
Cowley
,
N.
Ling
,
W.
Goetze
,
R.
Ahuja
,
W.
Luo
, and
B.
Johansson
, “
Elemental engineering: Epitaxial uranium thin films
,”
Phys. Rev. B
78
(
19
),
193403
(
2008
).
3.
A.
Banos
,
R.
Burrows
, and
T. B.
Scott
, “
A review of the mechanisms, reaction products and parameters affecting uranium corrosion in water
,”
Coord. Chem. Rev.
439
,
213899
(
2021
).
4.
J.
Petiau
,
G.
Calas
,
D.
Petitmaire
,
A.
Bianconi
,
M.
Benfatto
, and
A.
Marcelli
, “
Delocalized versus localized unoccupied 5f states and the uranium site structure in uranium oxides and glasses probed by x-ray-absorption near-edge structure
,”
Phys. Rev. B
34
(
10
),
7350
7361
(
1986
).
5.
W. D.
Schneider
and
C.
Laubschat
, “
5f-electron localization in uranium compounds
,”
Phys. Rev. Lett.
46
(
15
),
1023
1027
(
1981
).
6.
R.
Arkush
,
M. H.
Mintz
, and
N.
Shamir
, “
Passivation of uranium towards air corrosion by N2+ and C+ ion implantation
,”
J. Nucl. Mater.
281
,
182
190
(
2000
).
7.
A.
Raveh
,
R.
Arkush
,
S.
Zalkind
, and
M.
Brill
, “
Passivation of uranium metal by radio-frequency plasma nitriding against gas phase (H2, H2O) corrosion
,”
Surf. Coat. Technol.
82
(
1–2
),
38
41
(
1996
).
8.
G. C.
Allen
and
N. R.
Holmes
, “
The passivation of uranium metal surfaces by nitrogen bombardment—The formation of uranium nitride
,”
J. Nucl. Mater.
152
(
2–3
),
187
193
(
1988
).
9.
F. P. A.
Robinson
and
L.
Golante
, “
Minimizing corrosion in uranium extraction solutions by application of anodic protection
,”
Corrosion
20
(
8
),
239t
244t
(
1964
).
10.
D. A.
Young
,
Phase Diagrams of the Elements
(
Lawrence Livermore National Lab.(LLNL)
,
Livermore, CA
,
1975
).
11.
J. B.
Condon
and
E. A.
Larson
, “
Kinetics of the uranium-hydrogen system
,”
J. Chem. Phys.
59
(
2
),
855
865
(
1973
).
12.
A. L.
DeMint
and
J. H.
Leckey
, “
Effect of silicon impurities and heat treatment on uranium hydriding rates
,”
J. Nucl. Mater.
281
(
2–3
),
208
212
(
2000
).
13.
J. B.
Condon
, “
Alternative model for nonstoichiometry in uranium hydride
,”
J. Phys. Chem.
79
(
1
),
42
48
(
1975
).
14.
G. L.
Powell
,
W. L.
Harper
, and
J. R.
Kirkpatrick
, “
The kinetics of the hydriding of uranium metal
,”
J. Less-Common Met.
172-174
,
116
123
(
1991
).
15.
J. R.
Kirkpatrick
, “
Diffusion with chemical reaction and a moving boundary
,”
J. Phys. Chem.
85
(
23
),
3444
3448
(
1981
).
16.
J.
Bloch
and
M. H.
Mintz
, “
Kinetics and mechanism of the U-H reaction
,”
J. Less-Common Met.
81
(
2
),
301
320
(
1981
).
17.
C. P.
Jones
,
T. B.
Scott
,
J. R.
Petherbridge
, and
J.
Glascott
, “
A surface science study of the initial stages of hydrogen corrosion on uranium metal and the role played by grain microstructure
,”
Solid State Ionics
231
,
81
86
(
2013
).
18.
J. B.
Condon
, “
Nucleation and growth in the hydriding reaction of uranium
,”
J. Less-Common Met.
73
(
1
),
105
112
(
1980
).
19.
J.
Bloch
and
M. H.
Mintz
, “
Kinetics and mechanisms of metal hydrides formation—A review
,”
J. Alloys Compd.
253–254
,
529
541
(
1997
).
20.
R.
Arkush
,
M.
Brill
,
S.
Zalkind
,
M. H.
Mintz
, and
N.
Shamir
, “
The effect of N2+ and C+ implantation on uranium hydride nucleation and growth kinetic
,”
J. Alloys Compd.
330–332
,
472
475
(
2002
).
21.
J. F.
Bingert
,
R. J.
Hanrahan
, Jr.
,
R. D.
Field
, and
P. O.
Dickerson
, “
Microtextural investigation of hydrided α-uranium
,”
J. Alloys Compd.
365
(
1-2
),
138
148
(
2004
).
22.
P.
Shi
,
L.
Shen
,
B.
Bai
,
D.
Lang
,
L.
Lu
,
G.
Li
,
X.
Lai
,
P.
Zhang
, and
X.
Wang
, “
Preferred hydride growth orientation of U-0.79wt.%Ti alloy with β+U2Ti microstructure
,”
J. Nucl. Mater.
441
(
1-3
),
1
5
(
2013
).
23.
K.
Balasubramanian
,
W. J.
Siekhaus
, and
W.
McLean
, “
Potential energy surfaces for the uranium hydriding reaction
,”
J. Chem. Phys.
119
(
12
),
5889
5900
(
2003
).
24.
A.
Loui
, “
The hydrogen corrosion of uranium: Identification of underlying causes and proposed mitigation strategies
,”
Report LLNL-TR-607653
, October 2012.
25.
H. R.
Hoekstra
,
S.
Siegel
,
L. H.
Fuchs
, and
J. J.
Katz
, “
The uranium–oxygen system: UO2.5 to U3O8
,”
J. Phys. Chem.
59
(
2
),
136
138
(
1955
).
26.
M.
Carboni
,
C. W.
Abney
,
S.
Liu
, and
W.
Lin
, “
Highly porous and stable metal-organic frameworks for uranium extraction
,”
Chem. Sci.
4
(
6
),
2396
2402
(
2013
).
27.
C. A.
Partin
,
S. V.
Lalonde
,
N. J.
Planavsky
,
A.
Bekker
,
O. J.
Rouxel
,
T. W.
Lyons
, and
K. O.
Konhauser
, “
Uranium in iron formations and the rise of atmospheric oxygen
,”
Chem. Geol.
362
,
82
90
(
2013
).
28.
O.
Jankovský
,
P.
Šimek
,
D.
Sedmidubský
,
S.
Matějková
,
Z.
Janoušek
,
F.
Šembera
,
M.
Pumera
, and
Z.
Sofer
, “
Water-soluble highly fluorinated graphite oxide
,”
RSC Adv.
4
(
3
),
1378
1387
(
2014
).
29.
P.
Ferrin
,
S.
Kandoi
,
A. U.
Nilekar
, and
M.
Mavrikakis
, “
Hydrogen adsorption, absorption and diffusion on and in transition metal surfaces: A DFT study
,”
Surf. Sci.
606
(
7–8
),
679
689
(
2012
).
30.
C. D.
Taylor
and
R.
Scott Lillard
, “
Ab-initio calculations of the hydrogen–uranium system: Surface phenomena, absorption, transport and trapping
,”
Acta Mater.
57
(
16
),
4707
4715
(
2009
).
31.
C. D.
Taylor
,
T.
Lookman
, and
R. S.
Lillard
, “
Ab initio calculations of the uranium–hydrogen system: Thermodynamics, hydrogen saturation of α-U and phase-transformation to UH3
,”
Acta Mater.
58
(
3
),
1045
1055
(
2010
).
32.
Y.
Yang
,
P.
Zhang
,
P.
Shi
, and
X.
Wang
, “
s-d electronic interaction induced H2 dissociation on the γ-U (100) surface and influences of niobium doping
,”
J. Phys. Chem. C
115
(
47
),
23381
23386
(
2011
).
33.
J. L.
Nie
,
H. Y.
Xiao
,
X. T.
Zu
, and
F.
Gao
, “
Hydrogen adsorption, dissociation and diffusion on the α-U(001) surface
,”
J. Phys.: Condens. Matter
20
(
44
),
445001
(
2008
).
34.
P.
Shi
,
Y.
Yang
,
B.
Ao
,
P.
Zhang
, and
X.
Wang
, “
Influences of surface substitutional Ti atom on hydrogen adsorption, dissociation, and diffusion behaviors on the α-U(001) surface
,”
J. Phys. Chem. C
118
(
46
),
26634
26640
(
2014
).
35.
D. R.
Morrison
and
A. K.
Ray
, “
Atomic H interaction with the γ-U (100) surface
,”
Phys. Status Solidi B
250
(
10
),
2221
2231
(
2013
).
36.
S. Q.
Huang
,
X. L.
Zeng
,
F. Q.
Zhao
, and
X. H.
Ju
, “
Density functional study of H2 molecule and H atom adsorption on α-U (001) surface
,”
Surf. Interface Anal.
48
(
6
),
328
333
(
2016
).
37.
G.
Kresse
and
J.
Furthmüller
, “
Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set
,”
Phys. Rev. B
54
,
11169
11186
(
1996
).
38.
J. P.
Perdew
,
K.
Burke
, and
M.
Ernzerhof
, “
Generalized gradient approximation made simple
,”
Phys. Rev. Lett.
77
(
18
),
3865
(
1996
).
39.
H. J.
Monkhorst
and
J. D.
Pack
, “
Special points for Brillouin-zone integrations
,”
Phys. Rev. B
13
(
12
),
5188
(
1976
).
40.
J.
Neugebauer
and
M.
Scheffler
, “
Adsorbate-substrate and adsorbate-adsorbate interactions of Na and K adlayers on Al(111)
,”
Phys. Rev. B
46
(
24
),
16067
(
1992
).
41.
G.
Henkelman
,
B. P.
Uberuaga
, and
H.
Jónsson
, “
A climbing image nudged elastic band method for finding saddle points and minimum energy paths
,”
J. Chem. Phys.
113
(
22
),
9901
9904
(
2000
).
42.
E. L.
Kolsbjerg
,
M. N.
Groves
, and
B.
Hammer
, “
An automated nudged elastic band method
,”
J. Chem. Phys.
145
,
094107
(
2016
).
43.
Q.
Su
,
H.
Deng
,
B.
Ao
,
S.
Xiao
,
P.
Chen
, and
W.
Hu
, “
First-principles study of nitrogen adsorption and dissociation on α-uranium (001) surface
,”
RSC Adv.
4
(
100
),
57308
57321
(
2014
).
44.
X.
Qu
,
R. S.
Li
,
B.
He
,
F.
Wang
, and
K. L.
Yuan
, “
Co-adsorption of O2 and H2O on α-uranium (110) surface: A density functional theory study
,”
Chin. Phys. B
27
(
7
),
076501
(
2018
).
45.
W. A.
Curtin
and
R. E.
Miller
, “
Atomistic/continuum coupling in computational materials science
,”
Modell. Simul. Mater. Sci. Eng.
11
(
3
),
R33
(
2003
).
46.
I. F.
Silvera
and
V. V.
Goldman
, “
The isotropic intermolecular potential for H2 and D2 in the solid and gas phase
,”
J. Chem. Phys.
69
(
9
),
4209
4213
(
1978
).
47.
W.
Tang
,
E.
Sanville
, and
G.
Henkelman
, “
A grid-based Bader analysis algorithm without lattice bias
,”
J. Phys.: Condens. Matter
21
(
8
),
084204
(
2009
).
48.
A.
Kahn
, “
Fermi level, work function and vacuum level
,”
Mater. Horiz.
3
(
1
),
7
10
(
2016
).
49.
Z. G.
Mei
,
L.
Liang
, and
A. M.
Yacout
, “
First-principles study of the surface properties of U-Mo system
,”
Comput. Mater. Sci.
142
,
355
360
(
2018
).
50.
M. N.
Huda
and
A. K.
Ray
, “
A density functional study of atomic hydrogen adsorption on plutonium layers
,”
Physica B
352
(
1–4
),
5
17
(
2004
).
51.
S. T.
Lin
and
A. R.
Kaufmann
, “
Magnetic properties of beta-uranium hydride
,”
Phys. Rev.
102
(
3
),
640
(
1956
).
52.
R. J.
Cave
and
M. D.
Newton
, “
Generalization of the Mulliken-Hush treatment for the calculation of electron transfer matrix elements
,”
Chem. Phys. Lett.
249
(
1-2
),
15
19
(
1996
).
53.
P. P.
Dholabhai
,
R.
Atta-Fynn
, and
A. K.
Ray
, “
A density functional study of atomic hydrogen and oxygen chemisorption on the relaxed (0001) surface of double hexagonal close packed americium
,”
Eur. Phys. J. B
61
,
261
270
(
2008
).