research papers
Effect of swap disorder on the physical properties of the quaternary Heusler alloy PdMnTiAl: a firstprinciples study
^{a}Physics Department and International Centre for Quantum and Molecular Structures, Shanghai University, Shanghai 200444, People's Republic of China, ^{b}Materials Genome Institute and Shanghai Key Laboratory of HighTemperature Superconductors, Shanghai University, Shanghai 200444, People's Republic of China, and ^{c}State Key Laboratory of Advanced Special Steel, Shanghai University, Shanghai 200072, People's Republic of China
^{*}Correspondence email: renwei@shu.edu.cn
Heusler alloys crystallize in a closepacked cubic structure, having a fouratom basis forming a facecentred cubic lattice. By selecting different composite elements, Heusler alloys provide a large family of members for frontier research of spintronics and magnetic materials and devices. In this paper, the structural, electronic and magnetic properties of a novel quaternary Heusler alloy, PdMnTiAl, have been investigated using a firstprinciples computational materials calculation. It was found that the stable ordered structure is a nonmagnetic Ytype1, in good agreement with the Slater–Pauling rule. From the band structure and the it is predicted that this Ytype1 configuration is a new gapless semimetal material. Furthermore, it was discovered that the Pd–Mn swapdisordered structure is more stable than the Ytype1 structure. The present work provides a guide for experiments to synthesize and characterize this Heusler alloy.
1. Introduction
In recent years, spintronic materials and devices have been intensively investigated because of their great potential for information technology applications (Prinz, 1998). At the same time, the development of computational modelling techniques in materials science has triggered the study of a huge variety of magnetic materials such as Heusler compounds (Özdoğan et al., 2013). Since 1903 (Heusler, 1903), when the first Heusler compound was synthesized, their intriguing physical properties, e.g. spintronic, optoelectronic and thermoelectric effects, have attracted the attention of numerous researchers (Zhang et al., 2004; Nikolaev et al., 2009; Abid et al., 2016; Zhang et al., 2015). Many novel materials have been synthesized experimentally and investigated theoretically (Jamer et al., 2015; Fang et al., 2014; Gao et al., 2015; Ouardi et al., 2013).
In regular Heusler X_{2}YZ compounds, X and Y are transition metals or rare earth elements and Z belongs to the main group of elements. There are generally two types of structure, Cu_{2}MnAl with (225) and Hg_{2}CuTi with (216) (Graf et al., 2011; Mohanta et al., 2017). HalfHeusler alloys (XYZ) represent another large class of this family of materials and may crystallize in a noncentrosymmetric cubic (Rogl et al., 2016). The quaternary Heusler compounds XX′YZ, the structural prototype of which is the alloy LiMgPdSn, denoted as a Y structure, have recently been intensively investigated (Eberz et al., 1980; Alijani et al., 2011). The valence of X′ is lower than that of the X atom, and the valence of the Y element is lower than the valence of either X or X′. The sequence of atoms along the facecentred cube's diagonal is X–Y–X′–Z and this was found to be the most energetically stable (Alijani et al., 2011), called a Ytype1 phase, with fractional coordinates X (0, 0, 0), X′ , Y and Z . There are a further two types of atomic arrangement in the quaternary Heusler XX′YZ compound, namely X–X′–Y–Z Ytype2: X (0, 0, 0), X′ , Y and Z , and X′–Y–X–Z Ytype3: X , X′ (0, 0, 0), Y and Z (Benkaddour et al., 2016).
Previous studies of quaternary Heusler alloys (Galanakis et al., 2016) showed interesting phenomena such as high spin polarization, halfmetallic spingapless semiconductors and zerogap material behaviours, which provide potential applications in spintronics, electronics and sensors (Wang et al., 2010; Wang, 2008; Wang, Cheng, Wang, Wang & Liu, 2016; Wang, Cheng, Wang & Liu et al., 2016). It has been found that the physical properties of Heusler alloys are highly dependent on the ordering of their structures (Graf et al., 2011; Liu et al., 2015). From experimental work, an unknown second phase precipitates in some Heusler alloys (Umetsu et al., 2012). An interesting question is how the disordered structure influences the magnetic properties of a Heusler alloy. To answer this question, we systematically investigated disordered configurations and magnetic structures. Ti_{2}MnZ compounds were found to be halfmetallic ferrimagnets with potential application in spintronic devices (Skaftouros et al., 2013; Lukashev et al., 2016; Fang et al., 2014). In the present work, a new quaternary Heusler alloy, PdMnTiAl, has been designed on the basis of firstprinciples calculations. We studied both ordered and disordered structures of this material, and found this compound to be most stable when ordered in a Ytype1 structure. By using VASP and AkaiKKR calculations (see section 2 for more details), we found the ordered PdMnTiAl alloy to be a nonmagnetic material in a Ytype1 ground state, in good agreement with the Slater–Pauling rule. Interestingly, we revealed that the Pd–Mn swapdisordered structure (Picozzi et al., 2004; Zhang et al., 2013) is more stable than the Ytype1 structure, and may present different magnetic moments that can be tuned by the degree of swap disorder.
2. Computational methods
The firstprinciples calculations were performed using the Vienna Ab initio Simulation Package (VASP) (Kresse & Furthmüller, 1996) within density functional theory (DFT). The ideal ordered quaternary Heusler alloy PdMnTiAl was built into a 16atom (four formula units). The geometry was then optimized to attain minimal energy structures. The energy cutoff was set to 500 eV. The convergences of energy and force were set to 10^{−6} eV and 0.01 eV Å^{−1}, respectively. In our work, the exchangecorrelation interaction is described by the Perdew–Burke–Ernzerhof (PBE) generalized gradient approximation (GGA) (Perdew et al., 1996). The was sampled using a set of 7 × 7 × 7 kpoint mesh. We tested the VASP calculations by including the spin orbital coupling (SOC), and the results show that it has negligible influence on subsequent results in the present work, including the calculated total energy and of the studied compounds.
To explore the influence of atomistic disordering on the properties of the alloy, we used the program package AkaiKKR (Akai, 1992; Durham et al., 1980), which utilizes the Korringa–Kohn–Rostoker (KKR) Green's function method and has a high speed and high precision for systems with random distributions of atoms at given sites. For disordered systems, the coherent potential approximation (CPA) is one of the most efficient solution methods for averaged properties. Here, the configuration average of the Green's function is expressed in terms of the effective mean Both ordered and disordered structures of the alloy were simulated by AkaiKKR calculations, and the ordered results were compared with the VASP predictions. The exchangecorrelation effects were included by using a GGA91 scheme in the AkaiKKR calculations.
3. Results and discussion
First, the ordered quaternary Heusler alloy PdMnTiAl is considered. Fig. 1(a) shows the three PdMnTiAl structures with their different atomic arrangements. The Ytype1 structure with (No. 216) has an optimized lattice parameter of 6.05 Å. From Fig. 1(b), the Ytype2 and Ytype3 configurations are found from our calculations to have relatively higher total energies. Both VASP and AkaiKKR were used to optimize the structures of these three types of alloy to obtain the equilibrium lattice parameters. Similar trends were observed, such that the Ytype1 configuration had the smallest lattice parameter and Ytype3 the largest. A comparison of the two different calculation methods is shown in Table 1, together with the VASP and AkaiKKR predictions for the magnetic moments. The Ytype1 configuration of PdMnTiAl was found to have zero magnetization by both methods. To confirm the DFT calculation results, we checked the magnetic moments with the Slater–Pauling (SP) rule (Galanakis et al., 2002) given by
where M_{tot} is the total and Z_{tot} is the total number of valence electrons in the compound. The elements studied here have the following valence electron configurations: Pd (s^{2}d^{8}), Ti (s^{2}d^{2}), Mn (s^{2}d^{5}) and Al (s^{2}p^{1}). Thus the structure of our Ytype1 configuration complies perfectly with the Slater–Pauling rule.

From the optimized structures, we calculated their VASP and AkaiKKR. Figs. 2 and 3 show that we obtain the same resulting electronic structures using AkaiKKR as with VASP. Fig. 2 shows that all three Ytype structures have bands which cross the thus indicating metallic behaviour. The nonmagnetic Ytype1 structure has a pseudogaplike DOS at the indicating a semimetal, whereas the magnetic structures Ytype2 and Ytype3 have Fermi levels located near DOS peaks. This may help to explain the energetic instability of the Ytype2 and Ytype3 configurations. In Fig. 3, we present the band structures which correspond to the DOS results.
(DOS) and band structures usingFrom the above calculations the ordered Ytype structures have different total energies and magnetic moments, and PdMnTiAl is more likely to have the Ytype1 structure. The Ytype2 and Ytype3 structures have higher energy, larger lattice parameters and greater magnetic moments than the Ytype1 structure. To investigate the atomistic disordered configurations of these quaternary Heusler alloys further, we carried out advanced calculations using AkaiKKR. Based on the above ordered structures, we considered a number of possible swap disorder types by intermixing between any two of the Pd, Ti, Mn and Al atoms. In the following, we use the numerals I, II, III, IV, V and VI to represent interchanges, or swaps, between Ti–Mn, Pd–Mn, Ti–Al, Pd–Al, Pd–Ti and Mn–Al, respectively, and VII, VIII, IX, X, XI and XII to represent Pd–Mn, Ti–Al, Pd–Ti, Mn–Al, Ti–Mn and Pd–Al swaps, respectively. For example, the M point in Fig. 4 indicates a disordered configuration of (Pd_{0.7}Mn_{0.3})Ti(Mn_{0.7}Pd_{0.3})Al with a Pd–Mn swap. Fig. 4 shows the calculated total energy and the of all these different disordered structures. Surprisingly, the Pd–Mn swapdisordered structure is energetically more stable than the ordered Ytype1. Moreover, the Pd–Mn swap not only lowers the total energy, but also gives rise to significant magnetization. The new groundstate disordered structure (Pd_{0.7}Mn_{0.3})Ti(Mn_{0.7}Pd_{0.3})Al has an energy decrease of 0.092 eV per formula unit compared with the ordered Ytype1 PdMnTiAl, with its total enhanced to 0.964 µ_{B}. The halfandhalf Pd–Mn randomly disordered configuration (Pd_{0.5}Mn_{0.5})Ti(Mn_{0.5}Pd_{0.5})Al is 0.047 eV lower in energy per formula unit and has the maximum of 1.132 µ_{B}. In the other cases of disorder, the total energies are increased by the disordering effect and the swap disorders tend to introduce finite total magnetization. We also calculated the DOS for all the different disordered structures, as presented in Fig. S1 in the supporting information.
We attempted to construct larger supercells and to use VASP to verify the validity of the corresponding AkaiKKR results. In a doublesized with 32 atoms (eight formula units), 25% swap disorders were simulated by exchanging two of the eight Pd atom positions and two of the eight Mn atom positions. Similarly, 50% swap disorders were simulated by exchanging four of the eight Pd atoms and four of the eight Mn atom positions. Four possible disordered structures for the 50% configurational swap disorder are shown in Fig. S2 in the supporting information. We found the swapdisordered structures to be more stable than the ordered PdMnTiAl Ytype1 structure and have also verified qualitatively the correctness of the magnetic moments. The results, shown in Table S1 in the supporting information, suggest that much larger supercells might be necessary to achieve a better quantitative comparison between AkaiKKR disorder calculations and VASP calculations. This comparison is beyond our current computational resources. However, experimental work is expected to synthesize and characterize the proposed PdMnTiAl quaternary Heusler alloy for eventual confirmation of our prediction.
4. Conclusions
The structure and electronic and magnetic properties of the quaternary Heusler alloy PdMnTiAl have been investigated by firstprinciples calculations. In these compounds, the ordered configuration of Ytype1 is more stable than those of Ytype2 and Ytype3. The semimetallic Ytype1 configuration shows zero in good agreement with the Slater–Pauling rule. Interestingly, we discovered that the Pd–Mn swapdisordered structure is more stable than the ordered Ytype1 configuration, and that the total magnetizations of these disordered (Pd_{1−x}Mn_{x})Ti(Mn_{1−x}Pd_{x})Al compounds are dependent on the degree of Pd–Mn swap, 0 < x < 1. We hope these findings will stimulate further investigation into spintronics materials and devices.
Supporting information
Additional table and figures. DOI: https://doi.org/10.1107/S205225251700745X/ct5004sup1.pdf
Acknowledgements
The Shanghai Supercomputer Center is gratefully acknowledged.
Funding information
Funding for this research was provided by: National Natural Science Foundation of China (award Nos. 51672171, 11274222); State Key Laboratory of Solidification Processing in NWPU (award No. SKLSP201703); National Key Basic Research Program of China (award No. 2015CB921600); Eastern Scholar Program from the Shanghai Municipal Education Commission; Special Program for Applied Research on Super Computation of the NSFC–Guangdong Joint Fund (second phase).
References
Abid, O. M., Menouer, S., Yakoubi, A., Khachai, H., Omran, S. B., Murtaza, G., Prakash, D., Khenata, R. & Verma, K. D. (2016). Superlattice Microstruct. 93, 171–185. CrossRef CAS Google Scholar
Akai, H. (1992). Hyperfine Interact. 68, 3–14. CrossRef Google Scholar
Alijani, V., Winterlik, J., Fecher, G. H., Naghavi, S. S. & Felser, C. (2011). Phys. Rev. B, 83, 184428. CrossRef Google Scholar
Benkaddour, K., Chahed, A., Amar, A., Rozale, H., Lakdja, A., Benhelal, O. & Sayede, A. (2016). J. Alloys Compd. 687, 211–220. CrossRef CAS Google Scholar
Durham, P., Gyorffy, B. & Pindor, A. (1980). J. Phys. F, 10, 661. CrossRef Google Scholar
Eberz, U., Seelentag, W. & Schuster, H.U. (1980). Z. Naturforsch. Teil B, 35, 1341–1343. Google Scholar
Fang, Q.L., Zhang, J.M. & Xu, K.W. (2014). J. Magn. Magn. Mater. 349, 104–108. CrossRef CAS Google Scholar
Galanakis, I., Dederichs, P. H. & Papanikolaou, N. (2002). Phys. Rev. B, 66, 174429. CrossRef Google Scholar
Galanakis, I., Özdoğan, K. & Şaşıoğlu, E. (2016). AIP Adv. 6, 055606. Google Scholar
Gao, Q., Li, L., Lei, G., Deng, J.B. & Hu, X.R. (2015). J. Magn. Magn. Mater. 379, 288–293. CrossRef CAS Google Scholar
Graf, T., Felser, C. & Parkin, S. S. P. (2011). Prog. Solid State Chem. 39, 1–50. Web of Science CrossRef CAS Google Scholar
Heusler, F. (1903). Verh. Dtsch. Phys. Ges. 5, 219–223. CAS Google Scholar
Jamer, M. E., Assaf, B. A., Sterbinsky, G. E., Arena, D., Lewis, L. H., Saúl, A. A., Radtke, G. & Heiman, D. (2015). Phys. Rev. B, 91, 094409. CrossRef Google Scholar
Kresse, G. & Furthmüller, J. (1996). Comput. Mater. Sci. 6, 15–50. CrossRef CAS Web of Science Google Scholar
Liu, B., Luo, H., Xin, Y., Zhang, Y., Meng, F., Liu, H., Liu, E., Wang, W. & Wu, G. (2015). Solid State Commun. 222, 23–27. CrossRef CAS Google Scholar
Lukashev, P., Kharel, P., Gilbert, S., Staten, B., Hurley, N., Fuglsby, R., Huh, Y., Valloppilly, S., Zhang, W., Yang, K., Skomski, R. & Sellmyer, D. J. (2016). Appl. Phys. Lett. 108, 141901. CrossRef Google Scholar
Mohanta, S. K., Tao, Y., Yan, X., Qin, G., Chandragiri, V., Li, X., Jing, C., Cao, S., Zhang, J., Qiao, Z., Gu, H. & Ren, W. (2017). J. Magn. Magn. Mater. 430, 65–69. CrossRef CAS Google Scholar
Nikolaev, K., Kolbo, P., Pokhil, T., Peng, X., Chen, Y., Ambrose, T. & Mryasov, O. (2009). Appl. Phys. Lett. 94, 222501. CrossRef Google Scholar
Ouardi, S., Fecher, G. H., Felser, C. & Kübler, J. (2013). Phys. Rev. Lett. 110, 100401. CrossRef PubMed Google Scholar
Özdoğan, K., Şaşıoğlu, E. & Galanakis, I. (2013). J. Appl. Phys. 113, 193903. Google Scholar
Perdew, J. P., Burke, K. & Ernzerhof, M. (1996). Phys. Rev. Lett. 77, 3865. CrossRef PubMed Google Scholar
Picozzi, S., Continenza, A. & Freeman, A. J. (2004). Phys. Rev. B, 69, 094423. CrossRef Google Scholar
Prinz, G. A. (1998). Science, 282, 1660–1663. Web of Science CrossRef CAS PubMed Google Scholar
Rogl, G., Grytsiv, A., Gürth, M., Tavassoli, A., Ebner, C., Wünschek, A., Puchegger, S., Soprunyuk, V., Schranz, W., Bauer, E., Müller, H., Zehetbauer, M. & Rogl, P. (2016). Acta Mater. 107, 178–195. CrossRef CAS Google Scholar
Skaftouros, S., Özdoğan, K., Şaşıoğlu, E. & Galanakis, I. (2013). Appl. Phys. Lett. 102, 022402. CrossRef Google Scholar
Umetsu, R. Y., Morimoto, N., Nagasako, M., Kainuma, R. & Kanomata, T. (2012). J. Alloys Compd. 528, 34–39. CrossRef CAS Google Scholar
Wang, X. L. (2008). Phys. Rev. Lett. 100, 156404. CrossRef PubMed Google Scholar
Wang, X., Cheng, Z., Wang, J. & Liu, G. (2016). J. Mater. Chem. C, 4, 8535–8544. CrossRef CAS Google Scholar
Wang, X., Cheng, Z., Wang, J., Wang, X.L. & Liu, G. (2016). J. Mater. Chem. C, 4, 7176–7192. CrossRef CAS Google Scholar
Wang, X.L., Dou, S. X. & Zhang, C. (2010). NPG Asia Mater. 2, 31–38. CrossRef CAS Google Scholar
Zhang, M., Brück, E., de Boer, F. R. & Wu, G. (2004). J. Magn. Magn. Mater. 283, 409–414. CrossRef CAS Google Scholar
Zhang, J., Li, X. & Yang, J. (2015). J. Mater. Chem. C, 3, 2563–2567. CrossRef CAS Google Scholar
Zhang, H. G., Zhang, C. Z., Zhu, W., Liu, E. K., Wang, W. H., Zhang, H. W., Cheng, J. L., Luo, H. Z. & Wu, G. H. (2013). J. Appl. Phys. 114, 013903. CrossRef Google Scholar
This is an openaccess article distributed under the terms of the Creative Commons Attribution (CCBY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.