University of Tokyo Study Reveals Mechanism of Pressure-Induced Amorphous Transitions in Silicon

University of Tokyo

Key points of the presentation

• Although there is usually only one amorphous state for each substance obtained by rapid cooling of a liquid state, it is known that a plurality of amorphous states exist in substances such as silicon, water, and silica. However, the mechanism by which the transition between different amorphous states (amorphous-amorphous transition) occurs has not been elucidated.
• By studying the dynamic process of the amorphous-amorphous transition of silicon induced by the application of pressure using molecular simulations, we will clarify the role of thermodynamics and mechanics in the solid state transformation without diffusion. succeeded in.
• The results of this research shed light on the microscopic mechanism of material transformation in the solid state, and contribute to the design of new amorphous materials and the advancement of the earth science field, where amorphous physical properties under high pressure are important keys. It is expected that

• Typical structural snapshot of low-density amorphous/high-density amorphous transition process

Presentation summary

Senior Program Advisor Hajime Tanaka (Specially Appointed Researcher/Professor Emeritus of the University of Tokyo) in the Highly Functional Materials Field of the Research Center for Advanced Science and Technology, the University of Tokyo, and Fang Zhao, a collaborating researcher (at the time of the research), have advanced machine learning potential andLocal structure analysis (Note 1)of silicon due to rapid pressure changes.Amorphous/Amorphous transition (Note 2)We investigated in detail the temporal changes in the structure through molecular dynamics simulations.
This study identified three amorphous forms that exist at room temperature (low-density amorphous: LDA, high-density amorphous: HDA, and ultra-high-density amorphous: VHDA), and revealed the structure of these phases. We succeeded in identifying the characteristic structural order variables. Furthermore, we revealed the dynamic pathways and mechanisms of the transition between these phases and the transition to denser crystals.
Specifically, in the transition from LDA to HDA, a nucleus of HDA with a non-spherical interface is formed within LDA and grows over time.Nucleation/growth type transition (Note 3)was observed (Figure 1). It turns out that this process is strongly influenced by mechanical factors specific to solid-state phase transitions.On the other hand, the reverse transition from HDA to LDA upon rapid pressure decrease results in the continuous development of fluctuations in the structural order.Spinodal decomposition type transition (Note 4)It has become clear that this process shows a characteristic process. It was also revealed that when further pressure is applied, LDA transitions to VHDA, but passes through HDA as an intermediate state. The ultimately formed VHDA state was found to be inherently unstable and transforms into even denser crystals.
These discoveries reveal the importance of the concerted thermodynamic and mechanical factors driving amorphous-amorphous transitions and crystallization, and improve the dynamics of structural elements in diffusionless solid-state transformations. The importance of physical destabilization was demonstrated. The results of this research will shed light on the conditions under which amorphous/amorphous transitions are induced by applying pressure to the amorphous solid state of various materials, as well as the microscopic mechanisms of the transitions, as well as new amorphous solid states. It is expected that the development of high quality functional materials and drugs, amorphous physical properties under high pressure, and phase transition between solids will provide useful basic knowledge in the fields of earth and planetary science.

This result was published in the online version of “Nature Communications” on January 16, 2024 (British time).

-A word from a researcher-

Silicon is a familiar substance, but it is not well known that it exists in multiple amorphous states. This study revealed that when an amorphous material changes its structure under pressure, it effectively utilizes a shape close to the structure to which it is transitioning. We hope that similar mechanisms universally exist in other materials such as silica, which is an important material in earth science. (Tanaka Hajime Senior Program Advisor)

Presentation content

Even materials with a single composition of atoms and molecules, such as carbon graphite and diamond, can have multiple crystalline forms. This phenomenon is known as crystal polymorphism, in which different forms exist even in the amorphous state, and the transition between these different amorphous states is called the amorphous-amorphous transition (Reference Reference 1).

Until now, it has been experimentally known that amorphous polymorphs exist in various materials widely used in daily life and industry, such as water, silicon, oxide glasses, chalcogenide glasses, and metallic glasses. Therefore, a basic understanding of amorphous and non-crystalline transitions is important not only in fundamental physics but also in the field of materials science. Furthermore, it is important to understand the relationship between the amorphous-amorphous transition in the solid state and the liquid-liquid transition in the liquid state. It is expected that this will lead to a fundamental understanding of the state.

However, the nature of the amorphous/non-crystalline transition is still shrouded in mystery due to its low structural order and the transition between states far from equilibrium (Reference 1). There is still debate as to whether the amorphous-amorphous transition is a continuous transition or whether it has a first-order nature similar to the phase transition between thermodynamic equilibrium states such as a crystal and a liquid. I am. This controversy is thought to be due in part to the lack of well-defined local structural order parameters that can describe the amorphous-amorphous transition. Also, it is important to know whether the amorphous/non-crystalline transition takes a transition mode unique to first-order transitions, such as the nucleation/growth type or the spinodal decomposition type, and whether the unique properties of solids, such as mechanical factors and disappearance of thermal diffusion, are involved. It has not been clarified whether they will be involved in the same way.

In particular, silicon is an extremely important atom as a semiconductor in its crystalline state at normal pressure, and amorphous silicon also has great potential for applications in the engineering field. Such semiconducting properties in the solid state at normal pressure are lost in the liquid state or at high pressure, and dramatic changes in physical properties are expected due to the amorphous/non-crystalline transition, such as exhibiting metallic properties. . Furthermore, it is known that silicon locally exhibits a regular tetrahedral structure at low pressure, and how this changes due to transition is an interesting issue.

Molecular dynamics simulations are the most promising and powerful tool to approach these questions regarding amorphous-amorphous transitions. In silicon, research using a simplified interatomic potential has focused on liquid/liquid transitions, but since it is not possible to accurately describe material behavior at high pressures, Research on crystalline transitions was lagging behind. Therefore, the research group took advantage of recently developed machine learning potential to study in detail the dynamics of the amorphous-amorphous transition in silicon induced by pressure jumps.

The group used advanced machine learning and local structure analysis to study the microscopic dynamics of the amorphous-to-amorphous transition in silicon caused by rapid pressure changes. As a result, three amorphous forms (low-density amorphous: LDA, high-density amorphous: HDA, and ultra-high-density amorphous: VHDA) were identified at room temperature, and structures that characterize the structure of each phase were identified. We have successfully identified the order variable. This revealed the dynamic pathways and mechanisms of the transition between these states and even to denser crystals. In the transition from LDA to HDA, a nucleation/growth-type transition is observed in which a HDA nucleus with a non-spherical interface is formed in LDA and grows over time, and the process, such as the roughness of the interface, is characteristic of solid phase transition. It has become clear that this is strongly influenced by mechanical factors. On the other hand, it has been revealed that the reverse transition from HDA to LDA during a rapid pressure decrease shows a spinodal decomposition-type transition process in which fluctuations in the structural order develop continuously. Furthermore, when further pressure is applied, LDA transitions to VHDA, but it has become clear that there is no direct transition path between the two phases, and that the transition occurs via HDA as an intermediate state. We also found that the ultimately formed VHDA state is inherently unstable and transforms into even denser crystals.This crystallization process ultimately results in stableSimple hexagonal crystal (sh crystal) (Note 5)The formation ofβ-Sn crystal (Note 6)It turns out that it goes through a two-step process that includes an intermediate state known as . It has also been revealed that in all of these processes, the precursor structural ordering of the amorphous state triggers the transition, and is important in reducing the thermodynamic and mechanical barriers associated with the transformation. was shown to play a role.

These discoveries reveal the importance of the concerted thermodynamic and mechanical factors driving amorphous-amorphous transitions and crystallization, and improve the dynamics of structural elements in diffusionless solid-state transformations. The importance of physical destabilization was demonstrated.

The results of this study further elucidate the fundamental principles governing structural transformations in diverse materials and provide valuable insight into how these factors can be manipulated and controlled. It is also useful for the development of new amorphous functional materials such as silicon and drugs in the amorphous state, as well as the field of earth and planetary science, where amorphous physical properties under high pressure and solid-state phase transitions play important roles. It is expected that this research will provide basic knowledge.

• Figure 2: Structural changes during HDA-VHDA transition process
  Blue, green, red, yellow, and magenta spheres represent atoms with low-density amorphous, high-density amorphous, ultra-dense amorphous, β-Sn, and simple hexagonal (sh)-like structures, respectively. . Atom sizes have been adjusted for each structure type for clarity.

references

Presenter

University of Tokyo Advanced Science and Technology Research Center Highly Functional Materials Field

• Fang Zhao (cooperating researcher at the time of research)
• Hajime Tanaka (Senior Program Advisor: Specially Appointed Researcher/Professor Emeritus, University of Tokyo)

Paper information

Magazine: Nature Communications (January 16) Title: Microscopic mechanisms of pressure-induced amorphous-amorphous transitions and crystallization in silicon Authors: Zhao Fan and Hajime Tanaka＊
＊Responsible author DOI:
10.1038/s41467-023-44332-6

research grants

This research was supported by the Ministry of Education, Culture, Sports, Science and Technology Grants-in-Aid for Scientific Research (Project Number: JP20H05619).

Glossary

contact information

Professor Emeritus, University of Tokyo
University of Tokyo Advanced Science and Technology Research Center Highly Functional Materials Field
Senior Program Advisor (Specially Appointed Researcher) Hajime Tanaka