WorksPhysics and Astronomy
Phase Transitions and Superconductivity in Ternary Hydride LiSiH at High Pressures
Abstract
We predicted a new ternary hydride LiSiH at high pressures. A systematic structure search in LiSiH compound reveals novel stable phases with intriguing electronic and phonon properties. It is found that LiSiH is dynamically stable from ambient pressure up to 400 GPa with three novel phases: P312, P, and P2m. The calculation of electron-phonon coupling combined with Bardeen-Cooper-Schrieffer's argument indicates that this compound may be a candidate for high superconductors under high pressures. In particular, the maximum of -LiSiH at 400 GPa reaches 56 K. These findings may pave the way for obtaining room temperature superconductors in dense hydrogen-rich compounds.
Introduction
The search for high-temperature superconducting materials has always been a hot topic in the field of condensed matter physics. According to the traditional Bardeen-Cooper-Schrieffer (BCS) theory[1], the superconducting transition temperature is directly proportional to the Debye temperature, and the Debye temperature is inversely proportional to the mass, so lighter elements may have higher superconducting transition temperatures, such as hydrogen. However, solid hydrogen is an insulating molecular crystal under ambient pressure, and there are strong covalent bonds in hydrogen molecules, so it is hard to obtain superconducting hydrogen under ambient pressure. In order to achieve superconductivity, conditions such as external pressure are needed. As a basic thermodynamic variable, pressure can change the distance between atoms of matter, so that atoms can be rearranged, and the crystal and electronic structure can be modulated. High pressure can very effectively shorten the distance between atoms, increase the overlap of adjacent electron orbits, and then change the interaction and electronic structure between atoms/molecules, forming high-pressure new phases with new structures and properties that are difficult to form under conventional conditions. Generally, in a sufficiently high pressure environment, the band gap will be narrowed, and the energy bands will overlap, which can transform the non-metallic state into the metallic state. Wigner and Huntington proposed that insulating hydrogen molecules can be transformed into metallic hydrogen under high pressure, showing a metallic state[2]. Ashcroft proposed chemical preloading[3], that is, non-hydrogen elements in hydrogen-rich compounds have an interaction effect on hydrogen elements in the lattice. Hydrogen-rich compounds can be metallized at a lower pressure than pure hydrogen, and such hydrogen-rich compounds may be potential room temperature superconductors. Therefore, hydrogen-dominated metal alloys may be potential high-temperature superconducting materials under high pressure, because of the high Debye temperature. In recent years, hydrogen sulfide (), discovered under high pressure, set a high-temperature superconductivity record due to its superconducting transition temperature of 203 K [4, 5], and subsequent theoretical studies and experimental demonstrations found that lanthanum hydride compounds are at a high pressure of 170 GPa, with a superconducting transition temperature of 250 K [6, 7]. This is the highest superconducting transition temperature confirmed by experiments on binary hydrides so far, which has further promoted the research on the superconductivity of hydrogen-rich compounds.
In the ”hydride rush” of searching high-temperature superconductors, the theoretical research of hydrides has stepped into the forefront. First-principles calculations based on density functional theory have given a series of superconductivity predictions about hydrogen-rich compounds, and proposed a new family of superconducting hydrides with a unique cage structure[8, 9, 10]. The metal atoms such as calcium or lanthanum are located in the center of the hydrogen cage. In these high superconducting hydrides, doping elements metalize hydrogen at relatively lower pressures, through ”chemical pre-compression”. Based on the characteristics of ”chemical pre-compression”, some ternary hydrogen storage materials are considered as candidates for high-temperature superconductors. Ternary metal hydrides are a convenient and valuable system for studying the metallization and superconductivity of metal hydrides because they can be synthesized under mild conditions and recovered under ambient pressure[11]. Some ternary hydrides have been found in theoretical and experimental work, e. g. ( = 0.3 K, 173 GPa)[12], ( = 20 K, 200 GPa)[13] and ( = 0.7 K, 300 GPa)[14], ( = 7 K above 100 GPa)[15]. A recent experimental work also reported a room-temperature superconductivity with of 288 K in the C-S-H system at 267 GPa[16].
Silicon and carbon belong to the same group IV elements. It is controversial that silicon atoms will bond with four H atoms to form molecular units in the Mg-Si-H system similar to . However, since the nucleus of silicon is larger than that of carbon, the strength of the Si-H bond is weaker than that of the C-H bond. This may promote the dissociation of the Si-H bond, resulting in a completely different crystal structure, and there will be new changes under high pressure. A typical example is the (M = Si[17], Ge[18], and Sn[19]) system, where the difference in bonding results in greatly different structures and properties. In addition, recent studies have found that lithium atom doping helps stabilize the structure of the Au-H system [20], because under high pressure, the Au and H systems are unstable, but the doping of lithium has a stabilizing effect, which may originate from the low electron negative of lithium. Based on first-principles calculations, by adding lithium to the poor superconducting , it is found that the superconducting transition temperature of reaches 472 K under 250GPa [21], indicating that it can be doped with lithium atoms, carrying ultra-high superconducting transition temperature. Zhang et al. performed structure searching simulations on the Li-Si-H system at pressure range of 50-350 GPa[22]. and are predicted to become good phonon-mediated superconductors with estimated of 54 and 77 K at 172 and 250 GPa, respectively. compound is predicted with relatively low symmetry ( and Cmcm) and insulating. While another study on predicted a potential high-temperature superconductor with a of 150-167 K at 200 GPa[23]. Boeri et.al also revealed a metallic phase which is thermodynamically stable down to 100 GPa with a critical temperature of K[24]. Thus, the stable phases and superconductivity of needed further exploration. In this paper, we focus on compound, searching for structures and superconductivity under high pressures. We found a series of phase transitions and a candidate high superconductors - with the maximum of 56 K. Our results will further explore the superconductivity in the Li-Si-H system and may help to guide future experiments.
Methods
To determine the high-pressure structures, we used our machine-learning-based crystal prediction code CRYSTREE [25]. CRYSTREE can learn the structural optimization procedure and output optimized structure using multi regression methods including random forest regressor, ridge regressor and ExtraTrees regressor, et al. [26]. The results of structures are checked by evolutionary crystal structure prediction method USPEX [27]. The electronic structure calculations with high accuracy for the stable structures were performed using the full-potential linearized augmented plane wave (FP−LAPW) method implemented in the WIEN2K code. [28] The Wu-Cohen generalized gradient approximation (WC-GGA) [29] was applied to the exchange-correlation potential calculation. The plane-wave cutoff was defined by , where is the minimum LAPW sphere radius and is the plane-wave vector cutoff. The phonon calculations were carried out by using a density functional perturbation theory (DFPT) approach through the Quantum-ESPRESSO program [30]. Ultrasoft pseudopotentials are used in the calculation with the cutoffs chosen as 60 Ry for the wave functions and 600 Ry for the charge density. The electronic self-consistent calculation was performed over a -point mesh. Dynamical matrices and the electron-phonon couplings were calculated on a -point grid. A dense grid was used for evaluating an accurate electron-phonon interaction matrix.
Results
The crystal structures of LiSiH are predicted at 0, 100, 200, 300 and 400 GPa, respectively, with system sizes containing up to 4 formula units (f.u.) per simulation cell. Three energy-competitive phases P312, P̄3 and P̄62m were found through our structure prediction. As shown in Figure 1, from ambient pressure up to 20 GPa, the more stable phase is P312. phase (space group NO.149), then phase (space group NO.147) becomes a more stable phase until 91 GPa, following a phase transition to phase (space group NO.189) up to the highest pressure we studied of 400 GPa. undergoes a series of phase transitions from trigonal symmetry ( and phases) to hexagonal symmetry ( phase).
Figure 1. The enthalpies per formula unit of P312 and P̄3 phases as a function of pressure, referenced to P̄62m phase of Li₂SiH₆.
All of the crystal structures of the stable compounds at 0 GPa, 50 GPa, and 300 GPa are shown in Figure 2. The low-pressure phases and contain units in the cell, with lithium atoms inserted between these units. For high-pressure phase, there are units with discrete lithium atoms. With the increasing pressure, more hydrogen atoms trend to condensate around silicon atom. This is a prerequisite for a higher superconducting transition temperature in hydrides. In and phases, units construct regular octahedrons, in which Si-H bond in phase is 1.599 Å and the length of Si-H bonds of phase is 1.472 Å. When the pressure further increases, the units change from the original octahedrons to 14-sided polyhedrons, and more hydrogen atoms condensate and bonded with central Si atom. Each Si atom in structure is connected to nine H atoms with Si-H bonds ranging from 1.417 Å to 1.480 Å. A Si-Si bond with a bond length of 2.014 Å is also formed in this phase at 300 GPa.
Figure 2. Predicted crystal structure of phases. Lithium atoms are represented by large green balls, silicon atoms are represented by blue balls, and hydrogen atoms are represented by small pink balls. The crystal structure of the predicted phase is generated using VESTA software. [31]
Figure 3 shows the calculated electronic band structure of for , and phases. Band structure for phase at 0 GPa are shown in Figure 3a, there is no energy band that crosses the Fermi level (), which demonstrates phase an insulator at ambient pressure. As shown in Figure 3b, with the increasing pressure, the most stable phase changes from to . A valence band slightly crossing at 50 GPa which shows a weak metal characteristic. becomes the most stable phase at 91 GPa. As shown in Figure 3 A hole-type band cross through the Fermi level between and at 300 GPa. When pressure further increase to 400 GPa, there are one electron-type band and two hole-type bands crossing the Fermi energy. This may help enhance the electron-phonon interaction. The total and projected density of states (DOS) are shown in Figure 4. In low-pressure phases P312 at ambient pressure and P at 50 GPa, the band gap are around 3 eV and 1.2 eV, respectively. While in high-pressure phase P2m, the DOS at the Fermi level show remarkable mentality with 0.17 eV/atom at 300 GPa and 0.15 eV/atom at 400 GPa. The contribution to the DOS of each element shows that the orbitals are highly hybridized.
Figure 3. Electronic band structure for each phase. (a) P312 phase at 0 GPa, (b) P phase at 50 GPa, (c) P2m at 300 GPa, (d) P2m at 400 GPa. The Fermi level is set to zero. The red, green, yellow, and cyan color represent the contributions of H-s, Si-p, Si-s and Li-s orbitals to the density of states.
Figure 4. The density of states of P̄62m, P̄3 and P312 phases at different pressures.
Figure 5 show the phonon spectra and partial phonon density of states (PDOS) of LiSiH under (a) P312 at 0 GPa, (b) P at 50 GPa, (c) P2m at 300 GPa and (d) P2m at 400 GPa, respectively. No soft phonon mode can be observed in all phases at the pressures we studied, indicating that these structures are dynamically stable. We also preformed symmetry analysis on the phonon modes. For P312 phase, its point group is . The irreducible representation can be written as , with $3A''_2 \oplus 6E'$ infrared active modes and Raman active modes. For P̄3 phase, its point group is , The irreducible representation can be written as , with infrared active modes and Raman active modes. For P̄62m phase, its point group is . The irreducible representation can be written as , with infrared active modes and Raman active modes. The PDOS show that, the vibrations of hydrogen atoms dominate the high-frequency region, while vibrations of lithium and silicon mainly distribute in the medium and low-frequency regions, respectively. It is consistent with the ordering of elements’ masses, the heavier atoms prefer vibrations with lower frequencies. We also noted that, with the increasing pressure, the maximum phonon frequency changes from at 0 GPa, to $\sim 2300
cm at 50 GPa, then to cm at 300 GPa and cm at 400 GPa. It will help to enhance the Debye temperature and logarithmic average frequency , which is directly proportional to the superconducting critical temperature.
Figure 5. The phonon dispersion curves and partial phonon density of states of P312, P and P2m phases, under different pressures.
| Structure | P (GPa) | (K) | (K) () | |
| P312-LiSiH | 0 | 784.45 | 0.003 | 0.00 |
| P-LiSiH | 50 | 880.68 | 0.23 | 0.01 |
| P2m-LiSiH | 100 | 1692.10 | 0.04 | 0.00 |
| P2m-LiSiH | 200 | 2059.12 | 0.07 | 0.00 |
| P2m-LiSiH | 300 | 1134.88 | 0.73 | 43.60 |
| P2m-LiSiH | 400 | 1143.95 | 0.82 | 56.33 |
Table 1. The logarithmical averaged phonon frequency , electron-phonon coupling (), and superconducting in LiSiH at different pressures.
TABLE I collects predicted superconducting properties of LiSiH phases. The calculated logarithmic average phonon frequency , electron-phonon coupling parameters , and superconducting critical temperatures at various pressures are listed. in the range of 100 and 400 GPa rise upon compression. We use the Allen and Dynes modified McMillan equation [32, 33] to estimate the superconducting temperature at different pressures.
The Coulomb pseudopotential is assumed to be 0.1 for all materials. Along with the calculated logarithmic average frequency , the resultant of P2-LiSiH reaches maximum value of 56 K at 400 GPa, which is the maximum pressure we studied.
Conclusions
We searched stable structures of the LiSiH compound in the pressure range from 0 to 400 GPa using the machine-learning-based crystal structure prediction algorithm, and predicted new stable phases of P312, P and P2m at different pressures. The maximum value of LiSiH phases obtained by first-principles calculations are 56 K in high pressure P2m phase. This study complements the investigation of a series of high- superconducting ternary hydrides.
Acknowledgments
This work is sponsored by the National Key R&D Program of China (Grant No. 2018YFA0704300), the National Natural Science Foundation of China (Grants No. U1932217, 12175107), open program from the National Lab of Solid State Microstructures of Nanjing University (Grant No. M32025), and NUPTSF (Grant No. NY219087, NY220038).
Author contributions: S.X.W wrote the paper with input from all the co-authors.
Competing financial interests: The authors declare no competing financial interests.
References
- [2]E. Wigner and H. B. Huntington, On the possibility of a metallic modification of hydrogen, The Journal of Chemical Physics 3, 764 (1935).
- [3]N. W. Ashcroft, Hydrogen dominant metallic alloys: High temperature superconductors?, Phys. Rev. Lett. 92, 187002 (2004).pubmed.ncbi.nlm.nih.gov/15169525
- [4]D. Duan, Y. Liu, F. Tian, D. Li, X. Huang, Z. Zhao, H. Yu, B. Liu, W. Tian, and T. Cui, Pressure-induced metallization of dense (h2s)2h2 with high-tc superconductivity, Scientific reports 4, 1 (2014).DOI
- [5]A. Drozdov, M. Eremets, I. Troyan, V. Ksenofontov, and S. I. Shylin, Conventional super-conductivity at 203 kelvin at high pressures in the sulfur hydride system, Nature 525, 73 (2015).DOI
- [6]M. Somayazulu, M. Ahart, A. K. Mishra, Z. M. Geballe, M. Baldini, Y. Meng, V. V. Struzhkin, and R. J. Hemley, Evidence for superconductivity above 260 k in lanthanum superhydride at megabar pressures, Phys. Rev. Lett. 122, 027001 (2019).DOI
- [7]A. Drozdov, P. Kong, V. Minkov, S. Besedin, M. Kuzovnikov, S. Mozaffari, L. Balicas, F. Bal-akirev, D. Graf, V. Prakapenka, et al., Superconductivity at 250 k in lanthanum hydride under high pressures, Nature 569, 528 (2019).
- [8]H. Wang, J. S. Tse, K. Tanaka, T. Iitaka, and Y. Ma, Superconductive sodalite-like clathrate calcium hydride at high pressures, Proceedings of the National Academy of Sciences 109, 6463 (2012).DOI
- [9]F. Peng, Y. Sun, C. J. Pickard, R. J. Needs, Q. Wu, and Y. Ma, Hydrogen clathrate structures in rare earth hydrides at high pressures: Possible route to room-temperature superconductiv-ity, Phys. Rev. Lett. 119, 107001 (2017).DOI
- [10]H. Liu, I. I. Naumov, Z. M. Geballe, M. Somayazulu, J. S. Tse, and R. J. Hemley, Dynamics and superconductivity in compressed lanthanum superhydride, Phys. Rev. B 98, 100102 (2018).DOI
- [11]D. Meng, M. Sakata, K. Shimizu, Y. Iijima, H. Saitoh, T. Sato, S. Takagi, and S.-i. Orimo, Superconductivity of the hydrogen-rich metal hydride Li5Moh11 under high pressure, Phys. Rev. B 99, 024508 (2019).
- [12]S. Zhang, L. Zhu, H. Liu, and G. Yang, Structure and electronic properties of fe2sh3 compound under high pressure, Inorganic Chemistry 55, 11434 (2016), pMID: 27783489.pubmed.ncbi.nlm.nih.gov/27783489
- [13]K. Grishakov, N. Degtyarenko, and E. Mazur, Electron, phonon, and superconducting prop- oretical Physics 128, 105 (2019).DOI
- [14]J. Chen, W. Cui, J. Shi, M. Xu, J. Hao, A. P. Durajski, and Y. Li, Computational design of novel hydrogen-rich ysCh compounds, ACS Omega 4, 14317 (2019), pMID: 31508557.DOI
- [15]T. Muramatsu, W. K. Wanene, M. Somayazulu, E. Vinitsky, D. Chandra, T. A. Strobel, V. V. Struzhkin, and R. J. Hemley, Metalization and superconductivity in the hydrogen-rich ionic salt bareh9, The Journal of Physical Chemistry C 119, 18007 (2015).DOI
- [16]E. Snider, N. Dasenbrock-Gammon, R. McBride, M. Debessai, H. Vindana, K. Vencatasamy, K. V. Lawler, A. Salamat, and R. P. Dias, Room-temperature superconductivity in a carbonaceous sulfur hydride, Nature 586, 373 (2020).DOI
- [17]M. I. Eremets, I. A. Trojan, S. A. Medvedev, J. S. Tse, and Y. Yao, Superconductivity in hydrogen dominant materials: Silane, Science 319, 1506 (2008), https://www.science.org/doi/pdf/10.1126/science.1153282.DOI
- [18]G. Gao, A. R. Oganov, A. Bergara, M. Martinez-Canales, T. Cui, T. Iitaka, Y. Ma, and G. Zou, Superconducting high pressure phase of germane, Phys. Rev. Lett. 101, 107002 (2008).pubmed.ncbi.nlm.nih.gov/18851245
- [19]J. S. Tse, Y. Yao, and K. Tanaka, Novel superconductivity in metallic snh4 under high pressure, Phys. Rev. Lett. 98, 117004 (2007).pubmed.ncbi.nlm.nih.gov/17501079
- [20]M. Rahm, R. Hoffmann, and N. W. Ashcroft, Ternary gold hydrides: Routes to stable and potentially superconducting compounds, Journal of the American Chemical Society 139, 8740 (2017), pMID: 28627173.DOI
- [21]Y. Sun, J. Lv, Y. Xie, H. Liu, and Y. Ma, Route to a superconducting phase above room temperature in electron-doped hydride compounds under high pressure, Phys. Rev. Lett. 123, 097001 (2019).DOI
- [22]P. Zhang, Y. Sun, X. Li, J. Lv, and H. Liu, Structure and superconductivity in compressed li-si-h compounds: Density functional theory calculations, Phys. Rev. B 102, 184103 (2020).DOI
- [23]Z. Shao, D. Duan, Y. Ma, H. Yu, H. Song, H. Xie, D. Li, F. Tian, B. Liu, and T. Cui, Ternary superconducting cophosphorus hydrides stabilized via lithium, npj Computational Materials 5, 1 (2019).DOI
- [24]C. Kokail, W. von der Linden, and L. Boeri, Prediction of high-Tc conventional superconductivity in the ternary lithium borohydride system, Phys. Rev. Materials 1, 074803 (2017).DOI
- [25]B. Li, Crystree is a crystal structure predictor based on machine learning (2020).
- [26]F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, Scikit-learn: Machine learning in Python, Journal of Machine Learning Research 12, 2825 (2011).arxiv.org/abs/1201.0490
- [27]P. V. Bushlanov, V. A. Blatov, and A. R. Oganov, Topology-based crystal structure generator, Computer Physics Communications 236, 1 (2019).DOI
- [28]K. Schwarz, P. Blaha, and G. Madsen, Electronic structure calculations of solids using the wien2k package for material sciences, Computer Physics Communications 147, 71 (2002), proceedings of the Europhysics Conference on Computational Physics Computational Modeling and Simulation of Complex Systems.DOI
- [29]Z. Wu and R. E. Cohen, More accurate generalized gradient approximation for solids, Phys. Rev. B 73, 235116 (2006).DOI
- [30]P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. D. Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer, U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. P. Seitsonen, A. Smogunov, P. Umari, and R. M. Wentzcovitch, QUANTUM ESPRESSO: a modular and open-source software project for quantum simulations of materials, Journal of Physics: Condensed Matter 21, 395502 (2009).DOI
- [31]K. Momma and F. Izumi, VESTA3 for three-dimensional visualization of crystal, volumetric and morphology data, Journal of Applied Crystallography 44, 1272 (2011).DOI
- [32]P. B. Allen and R. C. Dynes, Transition temperature of strong-coupled superconductors re-analyzed, Phys. Rev. B 12, 905 (1975).
- [33]W. L. McMillan, Transition temperature of strong-coupled superconductors, Phys. Rev. 167, 331 (1968).DOI




