High-pressure phase and pressure-induced phase transition of Ag₃YCl₆

Abstract

Hexagonal Ag₃YCl₆ was found to transform into a new phase by applying pressure above ∼1.5 GPa via a diamond anvil cell. In situ XRD analysis revealed that the phase is monoclinic Ag₃YCl₆. The bulk modulus of the hexagonal and monoclinic phases was 37 GPa and 49 GPa, respectively. Computational investigations also supported this phase transition and bulk modulus. After releasing pressure from 2 GPa, the diffraction peaks of hexagonal Ag₃YCl₆ were detected, indicating the reversibility of this phase transformation.

---

Introduction

High-pressure chemistry in compressive matter is expanded by X-ray diffraction (XRD) techniques with a diamond anvil cell (DAC) and computational chemistry.¹ The phase transition of simple chlorides, such as NaCl and AgCl, has been studied for model systems, which transformed above 30 GPa (ref. 2) and 6.6 GPa, respectively.³ Higher pressure of the Na–Cl system brings metal-rich subchlorides, such as Na₂Cl and Na₃Cl.⁴ Moreover, the physical properties of simple chlorides under high pressure have been recently explored, such as the thermal conductivity of NaCl (ref. 5) and the ionic conductivity of AgCl.⁶

On the other hand, the high-pressure chemistry of ternary or more complex chlorides has yet to be explored, lacking the fundamental understanding primarily because of the experimental difficulty owing to the hygroscopic nature of chlorides. By changing the charge balance with cations with valence, complicated and defective structures afford structural variety and exciting properties.

Among ternary chlorides, alkali-metal yttrium chlorides construct rather complicated structures because of the rather large trivalent Y³⁺ (0.90 Å (ref. 7)) producing YCl₆ octahedra similar to NaCl₆ ones but cation vacancies and/or distortion are essential to maintain charge balance. Na₃YCl₆ presents two polymorphs at ambient pressure, namely, monoclinic and hexagonal. The hexagonal structure is a low-temperature phase with low density, while the monoclinic one is a high-temperature phase with high density.⁸ As a result, the phase transition by increasing temperature decreases the volume.⁸ Zirconium-doped Na₃YCl₆ analogous to the monoclinic Na₃YCl₆ phase has also been studied as an ionic conductor,¹⁰ and Li₃YCl₆ with a variety of defective rock-salt structures shows high ionic conductivity.^11,12

Given the similar ionic radii of Ag⁺ (1.00 Å) and Na⁺ (0.99 Å),⁷ it is not surprising that the known structure of Ag₃YCl₆ is isostructural with the hexagonal form of Na₃YCl₆.¹³ However, no observation has been made for Ag₃YCl₆ being isostructural with the monoclinic Na₃YCl₆, which could potentially function as an Ag ion conductor, akin to the Na₃YCl₆ counterpart. While it is known that heating the hexagonal Ag₃YCl₆ leads to phase transformation at 350 °C, the resulting phase remains unsolved.¹⁴

In this study, we showed, for the first time, that high-pressure phase transformation converts hexagonal Ag₃YCl₆ to the novel, monoclinic Ag₃YCl₆ with the aid of synchrotron XRD and atomistic calculations.

Experimental and computational methods

All operations in the experiment were performed under non-atmospheric conditions without exposure to moisture using an Ar-filled glovebox. After mixing AgCl and YCl₃, hexagonal Ag₃YCl₆ was synthesized by heating at 800 °C for 3 hours, cooling to 300 °C at 250 °C h⁻¹, and annealing at 300 °C for 10 hours.

For in situ XRD measurement under pressure, the synthesized hexagonal Ag₃YCl₆ was pressurized in a diamond anvil cell using helium gas as the pressure medium. The pressure-induced phase transition was investigated on BL10XU for the diamond anvil cell (DAC) at SPring-8 (approved numbers: 2022A1323 and 2023A1385).¹⁵ The wavelength was 0.4127389 Å, and the size of the X-ray beam was 6.5 micrometers vertically and 9.5 micrometers horizontally. Three pieces of rubies were installed in the DAC. The applied pressure was calculated from the average of the pressures obtained from three ruby fluorescence measurements placed in the sample chamber of the DAC. The pressure medium was He gas, and a hydrostatic pressure of approximately 0.1 to 6 GPa was applied.

For ex situ XRD measurement, hydrostatic pressure was applied to the synthesized Ag₃YCl₆ at 2 GPa using a multi-anvil cell. The measurement was performed under ambient pressure after the release of the pressure.

We used PreFerred Potential (PFP)¹⁶ version 4.0.0 with D3 correction with the aid of the atomic simulation environment (ASE)¹⁷ to optimize crystal structures and evaluate the bulk modulus. The initial crystal structures of hexagonal and monoclinic Na₃YCl₆ were retrieved from the Inorganic Crystal Structure Database. After the isomorphic substitution of Na with Ag, structure relaxation was performed without external pressure until the maximum force became less than 0.001 eV Å⁻¹. The optimized structures were used as inputs for structure optimization under different hydrostatic pressures until the maximum force became less than 0.02 eV Å⁻¹. The enthalpy H was computed from the volume V and the total energy E of the structures optimized under the pressure P, using the relation H = E + PV. Density functional theory (DFT) calculations were also conducted to assess the density of states using VASP version 6.2.1 at the optimized structures obtained by PFP. DFT calculations were performed by employing the PBE functional along with the D3 correction. A cutoff energy of 520 eV was employed. The k-point sampling was carried out using the Monkhorst–Pack method with a 20 × 20 × 20 grid.

Results and discussion

The XRD patterns of the synthesized products showed the diffraction of hexagonal Ag₃YCl₆ and residual AgCl. The lattice parameters of the synthesized hexagonal Ag₃YCl₆ were determined to be a = 6.8822(10) Å and c = 18.338(3) Å, which are consistent with the previously reported values (a = 6.8669(14) Å and c = 18.3050(50) Å).¹⁴

Upon the application of hydrostatic pressure, a clear phase transition was observed around 1.5 GPa in the in situ XRD pattern (Fig. 1a and S1†). All the peaks except for AgCl can be assigned to the monoclinic phase, which is isostructural with monoclinic Na₃YCl₆ (Fig. 1b). The intensity of the diffraction peaks did not match the simulated intensity. This can be attributed to the focused micro-sized beam that acquired only a specific position of the crystalline powder and could not obtain the information on the averaged structure.

Fig. 1 Phase transition of hexagonal Ag₃YCl₆ under pressure. a) In situ XRD patterns of Ag₃YCl₆ under hydrostatic pressure (λ = 0.412739 Å). b) Comparison of the XRD pattern of hexagonal Ag₃YCl₆ under 2.0 GPa and the simulated pattern of monoclinic Ag₃YCl₆.¹⁸

The application of pressure has led to the broadening of the peak, as shown in Fig. 1a. This can be attributed to 1) more inhomogeneous pressure detected in the differences of the Raman spectrum of ruby crystals (Table S1†), 2) lattice strain-induced broadening, and 3) lowering of the crystallite size by large volume shrinkage during the phase transformation from the hexagonal structure to the monoclinic structure. Above the pressure of 6.5 GPa, no other distinct phases are observed except for AgCl. This can be attributed to the formation of a low-crystalline or amorphous phase. While pressure-induced amorphization has been reported in various metals, oxides, and van der Waals' solids,⁸ amorphization under high pressure in ionic compounds such as chlorides is rare. AgCl and NaCl do not undergo a phase transition into an amorphous phase at pressures below 16.1 GPa. Thus, further study is necessary to clarify this transition.

From the results of in situ XRD measurements, the lattice parameters at each pressure of monoclinic Ag₃YCl₆ and hexagonal Ag₃YCl₆ were determined (Tables S2 and S3†). Fig. 2a illustrates the pressure-dependent change in lattice volume for the monoclinic structure isostructural with Na₃YCl₆ and hexagonal phases. Despite minor differences in lattice parameters, the slope of experimental data aligns with that of computational values. The experimentally determined bulk modulus was 37 GPa for hexagonal Ag₃YCl₆ and 49 GPa for monoclinic Ag₃YCl₆. Meanwhile, atomistic simulations determined the bulk modulus to be 32 GPa for hexagonal Ag₃YCl₆ and 49 GPa for monoclinic Ag₃YCl₆. The agreement between the experimental and simulated values implies the reliability of the bulk modulus. The lower bulk modulus of hexagonal Ag₃YCl₆ can be attributed to the Ag split sites in the low-density hexagonal Ag₃YCl₆.

Fig. 2 a) Pressure-dependent lattice volumes of hexagonal Ag₃YCl₆ and monoclinic Ag₃YCl₆ (experiment: circle, simulation: triangle). b) Pressure-dependent calculated enthalpy of hexagonal Ag₃YCl₆ relative to the monoclinic phase.

The phase transition of hexagonal Ag₃YCl₆ was rationalized by calculating its enthalpy relative to monoclinic Ag₃YCl₆ under varying pressures, using an atomistic simulation technique (Fig. 2b). At lower pressures, the hexagonal structure was more stable than the monoclinic form, aligning with experimental observations that identify the hexagonal structure as the stable phase under ambient conditions. As the pressure increased, the enthalpy difference between the two structures decreased linearly. Fig. 2b reveals that the monoclinic structure surpasses the hexagonal structure in stability beyond a pressure of 1.29 GPa. This indicates that the theoretical phase transition pressure is 1.29 GPa, a value remarkably consistent with the experimentally determined phase transition range of 1.0–1.6 GPa (Fig. 1a).

Here, we discuss the crystal structures and phase transition. The crystal structures of the hexagonal and monoclinic phases are shown in Fig. 3. Because we could not experimentally determine the atomic position of the monoclinic cell, the proposed monoclinic structure is the calculated one (Table S4†), which is supported by the volume change shown in Fig. 2. Monoclinic Ag₃YCl₆ has a cryolite-type structure, isostructural with monoclinic Na₃YCl₆. This monoclinic structure is composed of two silver (Ag) sites, three chlorine (Cl) sites, and one yttrium (Y) site. The Ag at the Ag₁ site has a disordered tetrahedral shape with four-coordination, while the Ag at the Ag₂ site has an octahedral shape with six-coordination of AgCl₆. Y also takes a six-coordinated YCl₆ octahedron. By the calculation, the shortest Ag–Cl bonding in the monoclinic cells is 2.62 Å, which is shorter than that in the hexagonal cells (2.68 Å).

Fig. 3 Crystal structure models of hexagonal and monoclinic Ag₃YCl₆. In the monoclinic phase, the Ag occupancy of Ag1 and Ag₂ sites is 0.5. The crystal structure of the monoclinic phase was optimized by atomic simulation at 0 GPa and was drawn using VESTA.¹⁸

The occupancy of Ag is the primary difference between the crystal structures of the hexagonal and monoclinic phases. In the hexagonal structure, the Ag occupancy of Ag₁ and Ag₂ sites is 0.5, whereas in the monoclinic structure, the Ag occupancy of all the Ag sites is 1. Consequently, the monoclinic structure is denser, supporting the phase transition from hexagonal to monoclinic with increasing pressure.

Because Ag₃YCl₆ and Na₃YCl₆ are isostructures, we can expect that the phase transformation of Ag₃YCl₆ is similar to that of Na₃YCl₆ in a previous report.⁹ During phase transformation, the positions of atoms at the Y and Ag₃ sites in the hexagonal phase and the Y and Ag₂ sites in the monoclinic phase do not change significantly.⁹ On the other hand, the partially occupied Ag₁ and Ag₂ sites in the hexagonal phase are largely different from the Ag₁ sites in the monoclinic phase,⁹ indicating that the phase transition is correlated with the migration from the partially occupied positions to the fully occupied position.

This phase transformation likely resulted in changes in the electronic structure properties such as the band structures. To demonstrate this, we assessed the density of states (DOS) of hexagonal Ag₃YCl₆ and monoclinic Ag₃YCl₆ using DFT calculations, as depicted in Fig. S2.† The calculated band gap of 2.56 eV for the starting hexagonal Ag₃YCl₆ slightly decreased to 2.51 eV for monoclinic Ag₃YCl₆. Given that the band gaps are composed of covalent Ag–Cl bonds, the reduction in the band gap can be attributed to the shorter bonding in the monoclinic cell.

To investigate the reversibility of the phase transformation, we carried out ex situ XRD analysis for samples before and after pressurization of 2 GPa. As shown in Fig. 4, both samples show the patterns for hexagonal Ag₃YCl₆. Because 2 GPa is above the phase transition pressure shown in Fig. 1 and 2, Ag₃YCl₆ should be transformed into the monoclinic phase, but only the hexagonal phase was observed after applying 2 GPa under ambient pressure, suggesting that the transition is reversible. The diffraction peaks of the hexagonal phase significantly broadened compared to the peaks before pressurization. This can be attributed to the crystalline strain and/or reduced crystallite size. The crystal sizes obtained before and after pressurization from the 102̄ peaks using Scherrer's formula were 51.9 and 26.5 nm, respectively.

Fig. 4 Ex situ XRD patterns of Ag₃YCl₆ before and after pressurization using CuKα radiation.

Conclusion

By using in situ and ex situ XRD with different pressures together with computational investigation, we showed that novel monoclinic Ag₃YCl₆ can be obtained by pressure-induced phase transformation of monoclinic Ag₃YCl₆. The low-pressure stable phase of hexagonal Ag₃YCl₆ transformed into the high-pressure stable phase of monoclinic Ag₃YCl₆ above ∼1.5 GPa. The monoclinic Ag₃YCl₆ phase could not be detected under atmospheric conditions, suggesting the backward transition to the hexagonal Ag₃YCl₆ phase. The bulk modulus of hexagonal Ag₃YCl₆ was lower than that of monoclinic Ag₃YCl₆. Further increase in pressure above 6 GPa diminished the diffraction peaks of Ag₃YCl₆.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was partly supported by JST SAKIGAKE JPMJPR21Q8. K. Muraoka acknowledges the support from JSPS KAKENHI (22K14751) and JST, PRESTO (JPMJPR2378).

References

1. X. Dong, A. R. Oganov, A. F. Goncharov, E. Stavrou, S. Lobanov, G. Saleh, G. R. Qian, Q. Zhu, C. Gatti, V. L. Deringer, R. Dronskowski, X. F. Zhou, V. B. Prakapenka, Z. Konôpková, I. A. Popov, A. I. Boldyrev and H. T. Wang, Nat. Chem., 2017, 9, 440–445.
2. W. A. Bassett, T. Takahashi, H. K. Mao and J. S. Weaver, J. Appl. Phys., 1968, 39, 319–325.
3. S. Hull and D. A. Keen, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 750–761.
4. W. Zhang, A. R. Oganov, A. F. Goncharov, Q. Zhu, S. E. Boulfelfel, A. O. Lyakhov, E. Stavrou, M. Somayazulu, V. B. Prakapenka and Z. Konôpková, Science, 2013, 342, 1502–1505.
5. W.-P. Hsieh, Sci. Rep., 2021, 11, 21321.
6. J. Wang, G. Zhang, H. Liu, Q. Wang, W. Shen, Y. Yan, C. Liu, Y. Han and C. Gao, Appl. Phys. Lett., 2017, 111, 031907.
7. R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751–767.
8. M. S. Wickleder and G. Meyer, ZAAC, 1995, 621, 457–463.
9. F. Stenzel and G. Meyer, ZAAC, 1993, 619, 652–660.
10. E. A. Wu, S. Banerjee, H. Tang, P. M. Richardson, J.-M. Doux, J. Qi, Z. Zhu, A. Grenier, Y. Li, E. Zhao, G. Deysher, E. Sebti, H. Nguyen, R. Stephens, G. Verbist, K. W. Chapman, R. J. Clément, A. Banerjee, Y. S. Meng and S. P. Ong, Nat. Commun., 2021, 12, 1256.
11. H. Ito, Y. Nakahira, N. Ishimatsu, Y. Goto, A. Yamashita, Y. Mizuguchi, C. Moriyoshi, T. Toyao, K. Shimizu, H. Oike, M. Enoki, N. C. Rosero-Navarro, A. Miura and K. Tadanaga, Bull. Chem. Soc. Jpn., 2023, 11, 1262–1268.
12. T. Asano, A. Sakai, S. Ouchi, M. Sakaida, A. Miyazaki and S. Hasegawa, Adv. Mater., 2018, 30, 1803075.
13. T. Staffel and G. Meyer, ZAAC, 1988, 557, 40–44.
14. K. Lerch, W. Laqua and G. Meyer, ZAAC, 1990, 582, 143–150.
15. N. Hirao, S. I. Kawaguchi, K. Hirose, K. Shimizu, E. Ohtani and Y. Ohishi, Matter Radiat. Extremes, 2020, 5, 018403.
16. S. Takamoto, C. Shinagawa, D. Motoki, K. Nakago, W. Li, I. Kurata, T. Watanabe, Y. Yayama, H. Iriguchi, Y. Asano, T. Onodera, T. Ishii, T. Kudo, H. Ono, R. Sawada, R. Ishitani, M. Ong, T. Yamaguchi, T. Kataoka, A. Hayashi, N. Charoenphakdee and T. Ibuka, Nat. Commun., 2022, 13, 2991.
17. A. Hjorth Larsen, J. Jørgen Mortensen, J. Blomqvist, I. E. Castelli, R. Christensen, M. Dułak, J. Friis, M. N. Groves, B. Hammer, C. Hargus, E. D. Hermes, P. C. Jennings, P. Bjerre Jensen, J. Kermode, J. R. Kitchin, E. Leonhard Kolsbjerg, J. Kubal, K. Kaasbjerg, S. Lysgaard, J. Bergmann Maronsson, T. Maxson, T. Olsen, L. Pastewka, A. Peterson, C. Rostgaard, J. Schiøtz, O. Schütt, M. Strange, K. S. Thygesen, T. Vegge, L. Vilhelmsen, M. Walter, Z. Zeng and K. W. Jacobsen, J. Phys.: Condens. Matter, 2017, 29, 273002.
18. K. Momma and F. Izumi, J. Appl. Crystallogr., 2008, 41, 653–658.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ce01185b

---