Next Article in Journal
Effects of Ti and N Contents on the Characteristic Evolution and Thermal Stability of MC Carbonitrides Holding at 1250 °C in H13 Die Steel
Previous Article in Journal
Properties of Fluorine-Free Steelmaking Flux Prepared Using Red Mud
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

First-Principles Investigation on the Adsorption and Diffusion of Oxygen at the B2(110)–O(001) Interface in Ti2AlNb Alloys

1
Shanghai Key Laboratory of D&A for Metal-Functional Materials, School of Materials Science & Engineering, Tongji University, Shanghai 201804, China
2
Biomaterials R&D Center, Zhuhai Institute of Advanced Technology, Chinese Academy of Sciences, Zhuhai 519000, China
3
Department of Mechanical, Industrial and Mechatronics Engineering, Toronto Metropolitan University, Toronto, ON M5B 2K3, Canada
*
Authors to whom correspondence should be addressed.
These authors contributed equally to this work.
Metals 2024, 14(3), 316; https://doi.org/10.3390/met14030316
Submission received: 13 February 2024 / Revised: 6 March 2024 / Accepted: 7 March 2024 / Published: 9 March 2024
(This article belongs to the Section Computation and Simulation on Metals)

Abstract

:
The adsorption and diffusion of oxygen at the B2(110)[ 1 ¯ 11]||O(001)[1 1 ¯ 0] interface in Ti2AlNb alloys were investigated via first-principles calculations. Only a 2.6% interfacial mismatch indicates that B2(110)–O(001) is basically a stable coherent interface. The calculated adsorption energies and diffusion energy barriers show that oxygen prefers to occupy the Ti-rich interstitial sites, and once trapped, it hardly diffuses to other interstitial sites, thus promoting the preferential formation of Ti oxides. Under the premise of a Ti-rich environment, a Nb-rich environment is more favorable for oxygen adsorption than an Al-rich environment. The electronic structures suggest that O 2p orbitals mainly occupy the energy region below −5 eV, bonding with its coordinated atoms of Ti, Al, and Nb. However, Al 3p and Nb 4d orbitals near the Fermi level couple with sparsely distributed O 2p orbitals, forming anti-bonding, which is not conducive to oxygen adsorption. Because Nb 4d electrons are more localized than Al 3p electrons are, Nb–O anti-bonding is weaker. O–Ti has almost no contribution to anti-bonding, suggesting good bonding between them. This is consistent with the experimental observations that TiO2 is the main oxidation product.

1. Introduction

Ti2AlNb alloys are regarded as the next generation of the most promising lightweight high-temperature aerospace structural materials due to their excellent mechanical properties, such as their good creep resistance, high strength-to-weight ratio, and satisfactory hot workability [1,2,3]. Ti2AlNb alloys are composed of an order of the orthorhombic (O) phase (space group: Cmcm, Pearson symbol: oS16), B2 phase (space group: Pm 3 ¯ m, Pearson symbol: cP2), and α2 phase (space group: P63/mmc, Pearson symbol: hP8) [4,5]. As a high-temperature alloy, the oxidation resistance of Ti2AlNb alloy is not good above 750 °C [6,7,8]. This is because the competitive oxidation of Ti and Al occurs at high temperatures, failing to produce a dense oxide layer to protect the alloy matrix [9,10].
In order to improve the high-temperature oxidation resistance of Ti2AlNb alloys, an in-depth study of the high-temperature oxidation mechanism is necessary. The experimental studies of the high-temperature oxidation behavior of Ti2AlNb alloys currently focus on oxidation kinetic testing, oxidation product and oxidation layer structure analysis, etc. It is shown that the oxide layer of the alloy consists of a mixture of TiO2 and Al2O3, of which TiO2 is the main oxidation product [11], because TiO2 is the preferred oxidation product of B2, α2, and orthorhombic (O) phases [12]. The oxidation process of Ti2AlNb alloys is controlled by the microstructure of the initial B2–O two-phase interface, and the extremely lamellar morphology restricts the diffusion of oxygen [13]. In our previous experiments on the oxidation mechanisms of Ti2AlNb alloys at 800 °C, the TEM observations revealed that the O phase and B2 phase followed a crystallographic-orientation relationship of [ 1 ¯ 11 ]B2//[ 1 1 ¯ 0 ]O and (110)B2//(001)O [14], which was also observed in other studies related to Ti2AlNb alloys [15,16].
Constrained by the experimental characterization scale, some theoretical studies have also been conducted to investigate the oxidation mechanism of Ti–Al–Nb alloys based on the electronic structure calculated via the first-principles density functional theory (DFT). The surface phase diagrams for oxygen adsorption on γ–TiAl low-index surfaces reveal that oxygen may induce surface segregations of Ti and Al in Ti-rich and Al-rich surface environments, respectively [17]. In α2-Ti3Al alloy, the driving force for the adsorption of oxygen atoms on the (0001) surface is provided by the O–Ti bonding strength, and the diffusion coefficients of oxygen in the alloy are anisotropic in different directions [18,19]. The study of oxygen adsorption on Ti2AlNb alloy surfaces indicates that the bonding strength of the O–Ti and O–Nb bonds is stronger than that of the O–Al bond [20]. In addition to the surfaces of the alloy, the interfaces also play an important role in oxidation behavior. Excellent interfacial stability means better mechanical properties and also increases the resistance to oxidation [21,22,23]. In the γ-TiAl–α2-Ti3Al interface, oxygen tends to occupy the Ti-rich octahedral sites, and the occupancy tendency of oxygen in the γ–α2 interface from high to low is α2–Ti3Al to the interface and γ-TiAl. Oxygen adsorption at the interface destabilizes the interface, leading to a reduction in the mechanical properties of the alloy [24].
However, first-principles studies on the oxidation mechanism of Ti2AlNb alloys, especially at the interface between different phases, are lacking. In this study, we investigated the adsorption and diffusion of oxygen at the B2(110)[ 1 ¯ 11 ]||O(001)[ 1 1 ¯ 0 ] interface, which is normally observed in Ti2AlNb alloys.

2. Computational Methods and Models

2.1. Computational Details

All first-principles calculations were performed using Vienna Ab-initio Simulation Packages (VASP.5.4.4) [25], based on the framework of Kohn–Sham density functional theory (DFT) [26]. The projector-augmented wave (PAW) method proposed by Blöchl and implemented by Kresse and Joubert was used with a cutoff energy of 560 eV [27]. The exchange correlation function was determined via generalized gradient approximation as formulated by Perdew, Burke, and Ernzerhof (GGA-PBE) [28]. Monkhorst–Pack k-point meshes were used for integration over the Brillouin zone [29], where the k-point mesh setting as a uniform mesh grid with a spacing of 0.03 Å was used to sample the complete Brillouin zone and calculate the density of states for all calculated systems. Brillouin zone integrations were carried out with the Methfessel–Paxton technique with a 0.1 eV smearing of the electron levels [30]. The PAW pseudopotentials considered were Ti 3s23p63d34s1, Al 3s23p1, Nb 4s24p64d45s1, and O 2s22p4. Full relaxation structure optimization was used for the bulk and surface of B2 and O phases to obtain the ground state crystal structure. With a precision test, a partial relaxation method with the lattice parameters fixed and internal atomic positions relaxed simultaneously was used for the interface B2(110)–O(001). The total energy convergence parameter during optimization was 1 × 10−5 eV/atom, the Hermann–Feynman force convergence parameter was 0.01 eV/Å, the tolerance shift was less than 0.002 Å, and the stress deviation per atom was less than 0.1 GPa.
We utilized the climbing image nudged elastic band (CI-NEB) method to identify the minimum energy pathway (MEP) for oxygen diffusion at the B2(110)–O(001) interface. The CI-NEB method was derived from the conventional search transition state nudged elastic band (NEB) method combined with climbing image [31]. As one of the most efficient methods, it requires just a few points to locate the transition state accurately. In the framework of the CI-NEB method, the atomic positions of all considered images were relaxed until the force on each atom was less than 0.03 eV/Å. Crystal orbital Hamilton population (COHP) calculations were performed for analyzing the interaction between oxygen atoms and the nearest coordinated atoms by means of LOBSTER [32]. LOBSTER reconstructs the chemical information in terms of local, auxiliary atomic orbitals outputted from DFT computations (VASP.5.4.4). The orbital bonding information between atoms can be visualized very well via COHP analysis; those that contribute positively to the bonding of atoms are called bonding components, and those that contribute negatively are called anti-bonding components [32].

2.2. Structural Models

The orthorhombic (O) phase has an orthogonal structure (Cmcm) (Figure 1a), with the experimental lattice parameters a = 6.090 Å, b = 9.570 Å, and c = 4.670 Å [4]. Ti atoms occupy 8g sites, while Al and Nb atoms are located at 4c1 and 4c2 sites, respectively. The B2 phase has a cubic structure with a space group of Pm 3 ¯ m (Figure 1b), and the experimental lattice parameters are 3.229 Å [4]. Ti and Al atoms occupy 8g sites and 4c sites, respectively. Their chemical formulas are Ti2AlNb and TiAl for the orthorhombic (O) phase and B2 phase, respectively.
Based on the surface energy tests and previous theoretical reports [33,34], a five-layer slab is sufficient to show the bulk characteristics for both B2(110) and O(001) surfaces. Therefore, five layers of B2(110) slabs and O(001) slabs were used to construct the interface structure of B2(110)[ 1 ¯ 11 ]||O(001)[ 1 1 ¯ 0 ], with a 15 Å vacuum layer along the c axis in order to block the periodic effects (Figure 1c). Subsequently, the relative position between B2(110) and O(001) was tested in the three axial directions in order to find the configuration with the lowest energy as the initial interface structure. The final interface configuration was obtained after structural optimization, and had an interfacial distance of 2.2 Å. As shown in Figure 1c, the two-phase interface was divided into five regions: the B2-phase bulk, B2-phase sub-interface, interface, O-phase sub-interface, and O-phase bulk. Only the B2-phase sub-interface, interface, and O-phase sub-interface were considered for oxygen adsorption and diffusion calculations (Figure 1d).

3. Results and Discussion

3.1. B2(110)/O(001) Interface

The lattice parameters of the full-relaxed crystal structures are a = 3.187 Å for the B2 phase, and a = 6.053 Å, b = 9.520, Å and c = 4.679 Å for the orthorhombic (O) phase, all being in good agreement with the experimental values. The interfacial mismatch parameter ξ is usually used to describe the stability of interface structure [35]:
ξ = 1 2 Ω A 1 + A 2 ,
where Ω is the area of the B2(110)/O(001) interface structure, and A1 and A2 are the areas of B2(110) slabs and O(001) slabs, respectively. Based on the optimized interface and slab structures, we obtained Ω = 57.588 Å2, A1 = 60.636 Å2, and A2 = 57.634 Å2, leading to a value of 2.6% for ξ. It is usually assumed that the interface satisfies the stabilizing relationship characteristics when the mismatch is less than 5%. The small interfacial mismatch degree also implies that the B2(110)/O(001) interface is basically coherent, which is often observed in Ti2AlNb alloys in experiments [14].
In order to investigate the bonding characteristics of the B2(110)/O(001) interface, the partial density of states (PDOS), charge density differences, and corresponding planar average charge density were calculated. The charge density difference (e/Å3) is defined as Δ ρ = Δ ρ i n t e r f a c e Δ ρ B 2 - s l a b Δ ρ O - s l a b ; here, Δ ρ i n t e r f a c e is the charge density of the B2–O two-phase interface, and Δ ρ B 2 - s l a b and Δ ρ O - s l a b represent the charge density of the isolated B2(110) and O(001) slabs, respectively. The planar average charge density difference (e/Å) is obtained by taking a planar average of Δ ρ in the ab plane (parallel to the contacting interface), which can reflect the characteristic charge fluctuation vertical to the interfacial direction (the c direction).
One can find from Figure 2a,b that the PDOS of Ti 3d, Al 3p, and Nb 4d orbitals of the B2-phase sub-interface and O-phase sub-interface have a smaller distribution at the Fermi level than those of B2-phase bulk and O-phase bulk, indicating much stronger bonding for the atoms in the sub-interfaces. This can also be observed from the charge density differences in Figure 2c. The red part represents the regions of electron accumulation, and the blue part represents those of electron depletion. Combining the charge density difference and the corresponding planar-averaged charge density difference plot, we can see that electron accumulation and depletion mainly happen at the interface with a little occurring in the next layers adjacent to it. In other words, only the atoms in the B2-phase sub-interface and O-phase sub-interface take part in bonding in forming the B2(110)–O(001) interface.
Compared with O phase, the electrons in B2 phase are more delocalized, i.e., Ti 3d orbitals cover −5 eV to Ef and Al 3p orbitals extend from Ef to −7.5 eV, creating strong hybridization with 3s orbitals (Figure 2a). As a result, Ti 3d orbitals interact with Al sp hybridized orbitals in the B2 phase. However, Al 3s and 3p orbitals are almost separated in the O phase, so the main interaction in the O phase takes place among the Al 3p, Ti 3d, and Nb 4d orbitals near Ef.

3.2. Oxygen Adsorption at B2(110)–O(001) Interface

We considered all possible octahedral interstitial sites for single-oxygen-atom adsorption at the B2(110)–O(001) interface. As shown in Figure 1d, they are O1a-b (2Ti–4Al coordination) and O2a-c (4Ti–2Al) at the B2-phase sub-interface, O3a-b (2Ti–3Al–1Nb), O4a-b (4Ti–2Al) and O5a-b (4Ti–1Al–1Nb) at the interface, and O6a-b (2Ti–2Al–2Nb), O7a-b (4Ti–1Al–1Nb), O8a (4Ti–2Al), and O9a (4Ti–2Nb) at the O-phase sub-interface.
Table 1 lists the total energies and adsorption energies of all oxygen-absorbed interface structures. The adsorption energies of oxygen atoms were calculated via the following equation:
E a d s = E ( O - s y s t e m ) ( E ( s y s t e m ) 1 2 E ( O 2 ) ) ,
where E(O-system) and E(system) are the total energies of the interfaces with and without oxygen, respectively, and E ( O 2 ) is the total energy of an oxygen molecule, which was calculated in a cubic box with 12 Å edges. Our calculated E ( O 2 ) is −9.854 eV, in which is good agreement with that of −9.860 eV in the previous report [19]. The obtained value of E(system) is −539.703 eV.
Lower adsorption energy means a more stable oxygen adsorption site. In the B2-phase sub-interface (Figure 1d), the most stable oxygen adsorption site is O2c, as shown in Table 1. It is followed by O2b, with the adsorption energy being 10 meV higher. Both of them are coordinated by four Ti atoms and two Al atoms. The small energy difference probably originates from the influence of the adjacent interface layer. The O2a site is also coordinated by four Ti atoms and two Al atoms but has a total energy that is about 0.3 eV higher. The value of the adsorption energy is strongly linked to the bonding strength of the oxygen atom with the nearest neighboring atoms in the site. The local structures show that for O2b site, the average bond lengths of Al–O and Ti–O bonds are 2.031 Å and 2.129 Å, while for the O2a site, they are 1.879 and 2.207 Å. Obviously, a shorter Ti–O bond supports a higher stability of oxygen atom adsorption. The O1a and O1b sites coordinated by two Ti atoms and four Al atoms were too unstable to shift to the 4Ti–2Al coordination sites after optimization. Obviously, oxygen atoms prefer the 4Ti–2Al coordination environment in the B2-phase sub-interface, and a shorter Ti–O bond is more supportive for oxygen adsorption.
In the O-phase sub-interface (Figure 1d), O9a (4Ti-2Nb) is the most stable oxygen adsorption site with the lowest Eads of −4.287 eV (Table 1), and is also most stable in the whole B2(110)–O(001) two-phase interface structure. The next site is the 4Ti–2Al-coordinated O8a site, with an Eads that is 0.17 eV higher than that of O9a. Apparently, the adsorption sites with 4Ti–2Al coordination (O8a) in the O-phase sub-interface have a much lower Eads than do those (O2a-c) in the B2-phase sub-interface, and the max energy difference is up to 0.6 eV (between O2a and O8a). The local structures show that the average bond lengths of T–-O at the O8a site are 2.075 Å, which is much shorter than that of 2.207 Å at the O2c site. 4Ti–1Al–1Nb-coordinated O7a and O7b sites have similar Eads values of around −3.7 eV, which is very close to the values of 4Ti–2Al-coordinated O2b and O2c sites in the B2-phase sub-interface. The most unstable sites in the O-phase sub-interface are 2Ti–2Al–2Nb-coordinated O6a and O6b. The oxygen atoms absorbed at the O6a site move to the 4Ti–2Nb site after optimization. Obviously, oxygen atoms prefer Ti-rich and Nb-rich coordination environments in the O-phase sub-interface.
At the interface (Figure 1d), the 4Ti–1Al–1Nb-coordinated O5b site is the most stable site with an Eads of −4.071 eV. Next is 4Ti–2Al coordinated O4b with an Eads of −3.966 eV. Here, O5a and O4a sites, which have the same coordination environment as O5b and O4b, respectively, have higher adsorption energies with differences of up to 0.3 eV and 0.4 eV. The local structures show that the average bond length of Ti–O at O4b is 2.130 Å, which is shorter than that of 2.150 Å at O4a, as mentioned above. The shorter Ti–O bond is more constructive for oxygen adsorption, and so the adsorption energy of O4b is 0.4 eV, which is lower than that of O4a. Comparing the local structures of O5a and O5b (4Ti-1Al-1Nb), we find that the bond lengths of Al–O and Nb–O at the O5b site (Eads = −4.071 eV) are 1.854 Å and 2.126 Å, respectively, which are shorter than those of 1.879 and 2.134 Å at the O5a site (Eads = −3.749 eV), while the average Ti–O bond length at O5b is 2.173 Å, which is longer than that of 2.148 Å at O5a. Apparently, Ti–O bonds no longer play a decisive role in the stability of oxygen adsorption at the 4Ti–1Al–1Nb-coordinated site at the interface.
Generally, the adsorption stability of oxygen atoms at the octahedral interstitial positions is mainly determined by their coordination environment, i.e., oxygen atoms prioritize occupying the Ti-rich and Nb-rich interstitial positions. Next, the local structure also affects the stability of oxygen adsorption, e.g., for the 4Ti–2Al-coordinated site, Ti–O bond strength plays a more important role than does the Al–O bond. As a whole, for the same coordination environment, the stability of the oxygen adsorption site is O-phase sub-interface > interface > B2-phase sub-interface in the B2(110)–O(001) interface system.

3.3. Oxygen Diffusion at B2(110)–O(001) Interface

Based on Table 1, possible diffusion pathways were designed for the interface, B2-phase, and O-phase sub-interfaces, as well as the place between the interface and sub-interface, in order to understand the oxygen diffusion at the B2(110)/O(001) interface, as summarized in Table 2.
It can be seen from Table 2 that the coordination environment of oxygen adsorption sites plays an important role in oxygen diffusion. Whether at the interface, the B2-phase sub-interface and O-phase sub-interface, or between the interface and sub-interface, the diffusion between Ti-rich sites has a high energy barrier, such as O2b → O2c (1.844 eV), O4a → O4b (1.698 eV), O8a → O9a (1.667 eV), O2c → O5b (1.632 eV), and O5b → O9a (1.773 eV). However, the diffusion from Ti-poor sites to Ti-rich sites has a much lower energy barrier, e.g., 0.199 eV for O3b (2Ti–3Al–1Nb) → O5a (4Ti–1Al–1Nb) at the interface, and 0.568 eV for O6b (2Ti–2Al–2Nb) → O7a (4Ti–1Al–1Nb) at the O-phase sub-interface. Moreover, the oxygen atom at O3b (2Ti–3Al–1Nb) diffuses more easily than that at O6b (2Ti–2Al–2Nb) to a Ti-rich site. This is consistent with the result of adsorption energy in Table 1, where O3b shows a higher Eads than O6b does, and the Ti-rich and Nb-rich coordination environments are more attractive for the adsorption of oxygen atoms. It is clear that the diffusion from the Ti-rich site to other sites requires crossing a high energy barrier, which makes the Ti-rich site an oxygen trap and hinders the diffusion of oxygen. As a result, oxygen atoms trapped at the Ti-rich site react with Ti atoms to produce Ti oxides preferentially.
Figure 3 shows the minimum energy paths of oxygen diffusion from a Ti-rich site, O8a (4Ti–2Al), to another Ti-rich site, O9a (4Ti–2Nb), and from a Ti-poor site, O3b (2Ti–3Al–1Nb), to Ti-rich site, O5a (4Ti–1Al–1Nb), with the corresponding partial density of states of the oxygen atom and its coordinated atoms of Ti, Al, and Nb. In order to obtain insights into bonding and anti-bonding, which are closely related to the stability of the local structure of oxygen adsorption, crystal orbital Hamilton population (COHP) calculations were performed for the bond between the oxygen atom and the nearest coordinated atom; the results are shown in Figure 4.
It can be seen from Figure 3a that it is hard for oxygen diffusion to happen between two stable oxygen adsorption sites because of the presence of a too-high energy barrier. The PDOS at the O8a site shows that O 2p orbitals divide into two peaks at −8.5 eV and −7.3 eV, which mainly interact with Al 3s orbitals, and Ti 3p, 3d, and 4s orbitals, respectively, while the PDOS of O 2p orbitals at the O9a site only have one peak at −7.3 eV and couples with Ti 3s, 3p, and 3d orbitals, and Nb 4d orbitals. As shown in Figure 4a,b, the bonding for both O8a and O9a sites’ oxygen adsorption originates from the O 2p orbital and its coordinated atoms’ orbitals in the energy region away from the Fermi level (<−5.0 eV). The anti-bonding for O8a site oxygen adsorption mainly stems from the O–Al bond, which is very delocalized, distributed from −7.0 eV to Ef, while the anti-bonding for O9a site oxygen adsorption comes from a more localized O–Nb bond and is weaker than the O–Al anti-bonding at O8a. For both O8a and O9a sites, the O–Ti bond has almost no contribution to anti-bonding, suggesting good bonding between them. This is consistent with the idea that oxygen atoms prefer a Ti-rich adsorption site.
Figure 3d shows that oxygen diffuses much more easily from a Ti-poor site (O3b, 2Ti–3Al–1Nb) to a Ti-rich site (O5a, 4Ti–1Al–1Nb) with a low energy barrier of 0.199 eV. The PDOS in Figure 3e shows that compared to those at O8a and O9a, O 2p electrons at the O8a site are more delocalized, with a wide peak ranging from −9.1 eV to −5 eV, and couple with Al 3s electrons, and Ti 3d, 3p, and 4s electrons. As shown in Figure 3f, the PDOS of O 2p orbitals at O5a has two peaks located at −8.2 eV and −7.3 eV, but these are different from those at O8a, which are not completely separated. The former mainly interact with Al 3s orbitals, but the latter couples with Ti 3d, 3p, and 4d orbitals, and Nb 4d orbitals. The COHP in Figure 4c,d indicates that for both O3b and O5a sites, bonding originates from the O 2p orbital and its coordinated atom orbitals in the energy region away from the Fermi level (<−5.0 eV), while anti-bonding mainly stems from the delocalized O–Al bond distributed from −7.0 eV to the Fermi level and the localized O–Nb bond distributed near the Fermi level. Also, the O–Ti bond has a very small contribution to anti-bonding just as in the case of the O8a and O9a sites. Too-large O–Al anti-bonding makes the O3b (2Ti–3Al–1Nb) site less stable than the O5a site (4Ti–1Al–1Nb).
Clearly, for oxygen adsorption at the B2(110)–O(001) interface, the oxygen atom at the Ti-rich site is more stable because of the strong bonding and very weak anti-bonding between Ti and oxygen atoms. Compared with Ti, more delocalized Al 3p electrons create strong anti-bonding with O 2p electrons, which is not conducive to oxygen adsorption. The Nb 4d electrons near Ef also create anti-bonding with O 2p electrons, but this is much weaker than Al–O anti-bonding because of the more localized 4d electrons.

4. Conclusions

First-principles density functional theory (DFT) was employed to study the adsorption and diffusion of oxygen at the B2(110)–O(001) interface in Ti2AlNb alloys. The extremely low interfacial mismatch parameter suggests that B2(110)[ 1 ¯ 11 ]||O(001)[ 1 1 ¯ 0 ] is a fairly stable coherent interface, consistent with the experimental report. The calculated adsorption energies indicate that the stability of the oxygen atom at the adsorption site is mainly determined by its coordination environment, i.e., the Ti-rich and Nb-rich interstitial sites are more supportive for oxygen adsorption. Moreover, for the same coordination environment, the stability of oxygen adsorption sites is O phase > interface > B2 phase in the B2(110)/O(001) interface system. The coordination environment also plays an important role in oxygen diffusion. In the B2(110)–O(001) interface, the diffusion between the Ti-rich sites needs to overcome a high energy barrier of over 1.5 eV, while only a much lower energy barrier of ~0.5 eV needs to be surmounted from Ti-poor sites to Ti-rich sites. The partial density of states with the corresponding crystal orbital Hamilton population suggests that O 2p orbitals mainly occupy the energy region below −5 eV and bond well with their coordinated atoms of Ti, Al, and Nb, supporting the stability of the oxygen atom at the absorption site. However, Al 3p electrons and Nb 4d electrons near the Fermi level interact with sparsely distributed O 2p electrons to create anti-bonding, weakening the stability of oxygen adsorption. Because Al 3p electrons are more delocalized, Al–O anti-bonding is stronger than Nb–O anti-bonding.

Author Contributions

Conceptualization, H.X. and A.F.; methodology, L.X.; software, M.Z. and H.X.; validation, H.X. and A.F.; formal analysis, L.X. and S.Q.; investigation, M.Z., H.X. and A.F.; resources, S.Q. and D.C.; data curation, L.X. and M.Z.; writing—original draft preparation, M.Z. and H.X.; writing—review and editing, A.F., S.Q. and D.C.; visualization, H.X.; supervision, D.C.; project administration, A.F.; funding acquisition, H.X. and S.Q. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundations of China (Grant Nos: 51971159, 52271012, and 51871168), Zhuhai Science Technology Department Project (2220004000049).

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding authors.

Acknowledgments

D.C. would like to thank the Natural Sciences and Engineering Research Council of Canada for their financial support.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Goyal, K.; Sardana, N. Mechanical Properties of the Ti2AlNb Intermetallic: A Review. Trans. Indian Inst. Met. 2021, 74, 1839–1853. [Google Scholar] [CrossRef]
  2. Goyal, K.; Bera, C.; Sardana, N. Temperature-dependent structural, mechanical, and thermodynamic properties of B2-phase Ti2AlNb for aerospace applications. J. Mater. Sci. 2022, 57, 19553–19570. [Google Scholar] [CrossRef]
  3. Wang, Y.; Wang, X.; Yang, Y.; Lan, X.; Zhang, Z.; Li, H. Study on Creep-Fatigue Mechanical Behavior and Life Prediction of Ti2AlNb-Based Alloy. Materials 2022, 15, 6238. [Google Scholar] [CrossRef] [PubMed]
  4. Banerjee, D. The intermetallic Ti2AlNb. Prog. Mater. Sci. 1997, 42, 135–158. [Google Scholar] [CrossRef]
  5. Zhang, H.Y.; Yan, N.; Liang, H.Y.; Liu, Y.C. Phase transformation and microstructure control of Ti2AlNb-based alloys: A review. J. Mater. Sci. Technol. 2021, 80, 203–216. [Google Scholar] [CrossRef]
  6. Dai, J.; Sun, C.; Wang, A.; Zhang, H.; Li, S.; Zhang, H. High temperature oxidation and hot corrosion behaviors of Ti2AlNb alloy at 923 K and 1023 K. Corros. Sci. 2021, 184, 109336. [Google Scholar] [CrossRef]
  7. He, Y.S.; Hu, R.; Luo, W.Z.; He, T.; Liu, X.H. Oxidation behavior of a novel multi-element alloyed Ti2AlNb-based alloy in temperature range of 650–850 degrees C. Rare Met. 2018, 37, 838–845. [Google Scholar] [CrossRef]
  8. Chen, W.; Huang, L.; Liu, Y.Y.; Zhao, Y.F.; Wang, Z.; Xie, Z.W. Oxidative Corrosion Mechanism of Ti2AlNb-Based Alloys during Alternate High Temperature-Salt Spray Exposure. Coatings 2022, 12, 1374. [Google Scholar] [CrossRef]
  9. Qu, S.J.; Tang, S.Q.; Feng, A.H.; Feng, C.; Shen, J.; Chen, D.L. Microstructural evolution and high-temperature oxidation mechanisms of a titanium aluminide based alloy. Acta Mater. 2018, 148, 300–310. [Google Scholar] [CrossRef]
  10. Cheng, J.; Li, J.; Rao, Q. Time-resolved in-situ XRD study on oxidation evolution of Ti2AlNb-based alloys. Mater. Today Commun. 2023, 36, 106660. [Google Scholar] [CrossRef]
  11. Malecka, J. Investigation of the oxidation behavior of orthorhombic Ti2AlNb alloy. J. Mater. Eng. Perform. 2015, 24, 1834–1840. [Google Scholar] [CrossRef]
  12. Chen, J.; Chen, Q.; Qu, S.J.; Xiang, H.P.; Wang, C.; Gao, J.B.; Feng, A.H.; Chen, D.L. Oxidation mechanisms of an intermetallic alloy at high temperatures. Scr. Mater. 2021, 199, 113852. [Google Scholar] [CrossRef]
  13. Ralison, A.; Dettenwanger, F.; Schutze, M. Oxidation of orthorhombic Ti2AlNb alloys in the temperature range 550-1000 degrees C in air. Mater. High Temp. 2003, 20, 607–629. [Google Scholar] [CrossRef]
  14. Xiang, J.M.; Mi, G.B.; Qu, S.J.; Huang, X.; Chen, Z.; Feng, A.H.; Shen, J.; Chen, D.L. Thermodynamic and microstructural study of Ti2AlNb oxides at 800 °C. Sci. Rep. 2018, 8, 12761. [Google Scholar] [CrossRef] [PubMed]
  15. Cai, Q.; Li, M.C.; Zhang, Y.R.; Liu, Y.C.; Ma, Z.Q.; Li, C.; Li, H.J. Precipitation behavior of Widmanstatten O phase associated with interface in aged Ti2AlNb-based alloys. Mater. Charact. 2018, 145, 413–422. [Google Scholar] [CrossRef]
  16. Zheng, Y.P.; Zeng, W.D.; Li, D.; Xu, J.W.; Ma, X.; Liang, X.B.; Zhang, J.W. Orthorhombic precipitate variant selection in a Ti2AlNb based alloy. Mater. Des. 2018, 158, 46–61. [Google Scholar] [CrossRef]
  17. Wang, L.; Shang, J.X.; Wang, F.H.; Chen, Y.; Zhang, Y. Oxygen adsorption on γ-TiAl surfaces and the related surface phase diagrams: A density-functional theory study. Acta Mater. 2013, 61, 1726–1738. [Google Scholar] [CrossRef]
  18. Wei, L.J.; Guo, J.X.; Dai, X.H.; Guan, L.; Wang, Y.L.; Liu, B.T. First-principles calculations of oxygen adsorption on the Ti3Al (0001) surface. Surf. Interface Anal. 2016, 48, 1337–1340. [Google Scholar] [CrossRef]
  19. Bakulin, A.V.; Kulkov, S.S.; Kulkova, S.E. Diffusion properties of oxygen in the α2-Ti3Al alloy. Intermetallics 2021, 137, 107281. [Google Scholar] [CrossRef]
  20. Li, Y.; Dai, J.H.; Song, Y.; Yang, R. Adsorption properties of oxygen atom on the surface of Ti2AlNb by first principles calculations. Comput. Mater. Sci. 2017, 139, 412–418. [Google Scholar] [CrossRef]
  21. Li, D.H.; Wang, B.B.; Luo, L.S.; Li, X.W.; Xu, Y.J.; Li, B.Q.; Wang, L.; Liu, W.Y.; Han, B.S.; Su, Y.Q.; et al. The interface structure and its impact on the mechanical behavior of TiAl/ Ti2AlNb laminated composites. Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process. 2021, 827, 142095. [Google Scholar] [CrossRef]
  22. Zou, C.X.; Li, J.S.; Zhu, L.; Zhang, Y.; Yao, G.; Tang, B.; Wang, J.; Kou, H.C.; Song, H.F.; Wang, W.Y. Electronic structures and properties of TiAl/Ti2AlNb heterogeneous interfaces: A comprehensive first-principles study. Intermetallics 2021, 133, 107173. [Google Scholar] [CrossRef]
  23. Ishkildin, A.D.; Kistanov, A.A.; Izosimov, A.A.; Korznikova, E.A. The nitriding effect on the stability and mechanical properties of the iron titan phase: First-principles investigation. Phys. Chem. Chem. Phys. 2023, 25, 24060–24068. [Google Scholar] [CrossRef]
  24. Wei, Y.; Zhou, H.B.; Zhang, Y.; Lu, G.H.; Xu, H. Effects of O in a binary-phase TiAl-Ti3Al alloy: From site occupancy to interfacial energetics. J. Phys. Condens. Matter. 2011, 23, 225504. [Google Scholar] [CrossRef] [PubMed]
  25. Kresse, G.; Furthmuller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 1996, 54, 11169–11186. [Google Scholar] [CrossRef] [PubMed]
  26. Kohn, W.; Sham, L. Self-consistent equations including exchange and correlation effects. Phys. Rev. A 1965, 140, 1133–1138. [Google Scholar] [CrossRef]
  27. Blochl, P.E. Projector augmented-wave method. Phys. Rev. B 1994, 50, 17953–17979. [Google Scholar] [CrossRef]
  28. Perdew, J.P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865–3868. [Google Scholar] [CrossRef]
  29. Monkhorst, H.J.; Pack, J.D. Special points for Brillouin-zone integrations. Phys. Rev. B 1976, 13, 5188–5192. [Google Scholar] [CrossRef]
  30. Methfessel, M.; Paxton, A.T. High-precision sampling for Brillouin-zone integration in metals. Phys. Rev. B 1989, 40, 3616–3621. [Google Scholar] [CrossRef]
  31. Henkelman, G.; Uberuaga, B.P.; Jonsson, H. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J. Chem. Phys. 2000, 113, 9901–9904. [Google Scholar] [CrossRef]
  32. Nelson, R.; Ertural, C.; George, J.; Deringer, V.L.; Hautier, G.; Dronskowski, R. LOBSTER: Local orbital projections, atomic charges, and chemical-bonding analysis fromprojector-augmented-wave-baseddensity-functional theory. J. Comput. Chem. 2020, 41, 1931–1940. [Google Scholar] [CrossRef] [PubMed]
  33. Wang, A.Q.; Liu, P.; Xie, J.P.; Ma, D.Q.; Mao, Z.P. First-principles investigation on the atomic structure, stability and electronic property of O(001)/B2(110) interface in Ti2AlNb alloys. J. Alloys Compd. 2020, 817, 152734. [Google Scholar] [CrossRef]
  34. Li, Y.; Dai, J.H.; Song, Y. First-principles investigation on stability and oxygen adsorption behavior of a O/B2 interface in Ti2AlNb alloys. J. Alloys Compd. 2020, 818, 152926. [Google Scholar] [CrossRef]
  35. Liu, P.; Han, X.; Sun, D.; Chen, Z.; Wang, Q. Adhesion, stability and electronic properties of Ti2AlN(0001)/TiAl(111) coherent interface from first-principles calculation. Intermetallics 2018, 96, 49–57. [Google Scholar] [CrossRef]
Figure 1. The crystal structures of the (a) O phase, (b) B2 phase, (c) B2(110)[ 1 ¯ 11 ]||O(001)[ 1 1 ¯ 0 ] interface, and (d) adsorption sites of oxygen, named according to O + number + letter, such as O1a, where the number and letter stand for the coordination environment and position, respectively (see Table 1). As shown in (c), the two-phase interface is divided into five regions: the B2-phase bulk, B2-phase sub-interface, interface, O-phase sub-interface, and O-phase bulk. Only the B2-phase sub-interface, interface, and O-phase sub-interface are considered for oxygen adsorption.
Figure 1. The crystal structures of the (a) O phase, (b) B2 phase, (c) B2(110)[ 1 ¯ 11 ]||O(001)[ 1 1 ¯ 0 ] interface, and (d) adsorption sites of oxygen, named according to O + number + letter, such as O1a, where the number and letter stand for the coordination environment and position, respectively (see Table 1). As shown in (c), the two-phase interface is divided into five regions: the B2-phase bulk, B2-phase sub-interface, interface, O-phase sub-interface, and O-phase bulk. Only the B2-phase sub-interface, interface, and O-phase sub-interface are considered for oxygen adsorption.
Metals 14 00316 g001
Figure 2. The partial density of states of Ti 3d, Al 3s, 3p, and Nb 4d orbitals of (a) the B2-phase bulk and B2-phase sub-interface, and (b) O-phase bulk and O-phase sub-interface (see Figure 1c). (c) The charge density difference (e/Å3) of the B2(110)–O(001) interface (left panel) and the corresponding planar-averaged charge density difference plot (Δρ) along the c direction (right panel). The zero energy in (a,b) indicates the Fermi level.
Figure 2. The partial density of states of Ti 3d, Al 3s, 3p, and Nb 4d orbitals of (a) the B2-phase bulk and B2-phase sub-interface, and (b) O-phase bulk and O-phase sub-interface (see Figure 1c). (c) The charge density difference (e/Å3) of the B2(110)–O(001) interface (left panel) and the corresponding planar-averaged charge density difference plot (Δρ) along the c direction (right panel). The zero energy in (a,b) indicates the Fermi level.
Metals 14 00316 g002
Figure 3. Minimum energy paths for the oxygen diffusion of (a) O8a (4Ti–2Al) → O9a (4Ti–2Nb) and (d) O3b (2Ti–3Al–1Nb) → O5a (4Ti–1Al–1Nb), and the partial density of states of the oxygen atom and the nearest coordinated atoms of Ti, Al, and Nb for oxygen adsorption at (b) O8a, (c) O9a, (e) O3b, and (f) O5a, where zero energy indicates the Fermi level.
Figure 3. Minimum energy paths for the oxygen diffusion of (a) O8a (4Ti–2Al) → O9a (4Ti–2Nb) and (d) O3b (2Ti–3Al–1Nb) → O5a (4Ti–1Al–1Nb), and the partial density of states of the oxygen atom and the nearest coordinated atoms of Ti, Al, and Nb for oxygen adsorption at (b) O8a, (c) O9a, (e) O3b, and (f) O5a, where zero energy indicates the Fermi level.
Metals 14 00316 g003
Figure 4. Crystal orbital Hamilton population (COHP) of the bonds between the oxygen atom and the nearest coordinated atoms of Ti, Al, and Nb for oxygen adsorption at the (a) O8a, (b) O9a, (c) O3b, and (d) O5a sites, where zero energy indicates the Fermi level, red areas represent anti-bonding components, and blue areas represent bonding components.
Figure 4. Crystal orbital Hamilton population (COHP) of the bonds between the oxygen atom and the nearest coordinated atoms of Ti, Al, and Nb for oxygen adsorption at the (a) O8a, (b) O9a, (c) O3b, and (d) O5a sites, where zero energy indicates the Fermi level, red areas represent anti-bonding components, and blue areas represent bonding components.
Metals 14 00316 g004
Table 1. The calculated total energies, E(O–system), and oxygen adsorption energies (Eads) of the B2(110)/O(001) interface corresponding to the coordination environment of oxygen adsorption sites (Coord.).
Table 1. The calculated total energies, E(O–system), and oxygen adsorption energies (Eads) of the B2(110)/O(001) interface corresponding to the coordination environment of oxygen adsorption sites (Coord.).
LocationSite (Coord.)E(O-system) (eV)Eads (eV)
B2-phase sub-interfaceO1a (2Ti-4Al)−548.205-
O1b (2Ti-4Al)−548.377-
O2a (4Ti-2Al)−548.090−3.460
O2b (4Ti-2Al)−548.380−3.750
O2c (4Ti-2Al)−548.394−3.764
InterfaceO3a (2Ti-3Al-1Nb)−548.698-
O3b (2Ti-3Al-1Nb)−547.340−2.710
O4a (4Ti-2Al)−548.195−3.565
O4b (4Ti-2Al)−548.596−3.966
O5a (4Ti-1Al-1Nb)−548.379−3.749
O5b (4Ti-1Al-1Nb)−548.701−4.071
O-phase sub-interfaceO6a (2Ti-2Al-2Nb)−548.915-
O6b (2Ti-2Al-2Nb)−547.820−3.190
O7a (4Ti-1Al-1Nb)−548.306−3.676
O7b (4Ti-1Al-1Nb)−548.377−3.747
O8a (4Ti-2Al)−548.747−4.117
O9a (4Ti-2Nb)−548.917−4.287
Table 2. The calculated energy barriers (eV) of oxygen atom diffusion from one site (Initial) to another site (Final) in the B2-phase sub-interface (B2-phase sub-int.) and O-phase sub-interface (O-phase sub-int.), at the interface, and between the interface and sub-interface in the B2(110)–O(001) interface system, corresponding to the coordination environment of oxygen adsorption sites (Coord.).
Table 2. The calculated energy barriers (eV) of oxygen atom diffusion from one site (Initial) to another site (Final) in the B2-phase sub-interface (B2-phase sub-int.) and O-phase sub-interface (O-phase sub-int.), at the interface, and between the interface and sub-interface in the B2(110)–O(001) interface system, corresponding to the coordination environment of oxygen adsorption sites (Coord.).
LocationInitial (Coord.)Final (Coord.)Energy Barrier (eV)
B2-phase sub-int.O2a (4Ti–2Al)O2b (4Ti–2Al)1.580
O2b (4Ti–2Al)O2c (4Ti–2Al)1.844
InterfaceO3b (2Ti–3Al–1Nb)O5a (4Ti–1Al–1Nb)0.199
O4a (4Ti–2Al)O4b (4Ti–2Al)1.698
O5a (4Ti–1Al–1Nb)O5b (4Ti–1Al–1Nb)1.518
O-phase sub-int.O6b (2Ti–2Al–2Nb)O7a (4Ti–1Al–1Nb)0.568
O7a (4Ti–1Al–1Nb)O8a (4Ti–2Al)1.427
O8a (4Ti–2Al)O9a (4Ti–2Nb)1.667
B2-phase sub-int.—InterfaceO2c (4Ti–2Al)O5b (4Ti–1Al–1Nb)1.632
O2b (4Ti–2Al)O4b (4Ti–2Al)1.825
Interface—O-phase sub-int.O5b (4Ti–1Al–1Nb)O9a (4Ti–2Nb)1.773
O4b (4Ti–2Al)O9a (4Ti–2Nb)1.584
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhang, M.; Xiang, H.; Xu, L.; Feng, A.; Qu, S.; Chen, D. First-Principles Investigation on the Adsorption and Diffusion of Oxygen at the B2(110)–O(001) Interface in Ti2AlNb Alloys. Metals 2024, 14, 316. https://doi.org/10.3390/met14030316

AMA Style

Zhang M, Xiang H, Xu L, Feng A, Qu S, Chen D. First-Principles Investigation on the Adsorption and Diffusion of Oxygen at the B2(110)–O(001) Interface in Ti2AlNb Alloys. Metals. 2024; 14(3):316. https://doi.org/10.3390/met14030316

Chicago/Turabian Style

Zhang, Ming, Hongping Xiang, Lin Xu, Aihan Feng, Shoujiang Qu, and Daolun Chen. 2024. "First-Principles Investigation on the Adsorption and Diffusion of Oxygen at the B2(110)–O(001) Interface in Ti2AlNb Alloys" Metals 14, no. 3: 316. https://doi.org/10.3390/met14030316

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop