Jiaming Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jiaming Zhu has a total of 137 co-authors. Here are the top co-authors by number of shared publications:
- Hong-Hui Wu
Institute for Carbon Neutrality, University of Science and Technology Beijing, Beijing 100083, China
17 shared publications Dong Wang
Center of Microstructure Science, Frontier Institute of Science and Technology, Xi’an Jiaotong University, Xi’an 710049, China
5 shared publicationsJunheng Gao
Institute for Carbon Neutrality, University of Science and Technology Beijing, Beijing, China
5 shared publicationsTong-Yi Zhang
5 shared publications- Tong-Yi Zhang
Shanghai University
5 shared publications Xinping Mao
Institute for Carbon Neutrality, University of Science and Technology Beijing, Beijing, China
5 shared publications
Latest Publications
- Open Access
- 1 November 2026
- Published by IOP Publishing in Smart Materials and Structures
- Vol. 35 (11) , 115025
Crystalline–amorphous architectures modulate martensitic transformation and mechanical response in NiTi
NiTi shape memory alloys (SMAs) exhibit excellent superelasticity, but their first-order martensitic transformation (MT) produces large hysteresis that limits functional performance. Recent experiments have shown that crystalline–amorphous (C + A) NiTi composites can significantly reduce hysteresis, yet the underlying microstructural mechanisms remain unclear. Here, phase-field simulations are performed to investigate stress-induced MT in C + A NiTi with bi-continuous and nanolaminate architectures. The effects of crystalline channel width, amorphous morphology, and crystalline/amorphous (C/A) interface orientation are systematically quantified. Reducing the crystalline channel width suppresses martensitic avalanches, transforming the stress–strain response from plateau-like pseudoelasticity to nearly linear superelasticity with vanishing hysteresis. The critical transformation stress can be tuned from 125 MPa to above 800 MPa, while the hysteresis decreases from approximately 500 MPa to nearly zero. Bi-continuous architectures exhibit robust low-hysteresis behavior over a wide range of channel widths, whereas in nanolaminates the C/A interface orientation relative to the habit plane between the parent austenite and the twinned martensite provides an additional mechanism for controlling elastic confinement. The microstructural effects are quantitatively rationalized through the elastic interaction energy, establishing a direct link between microstructural architecture and MT thermodynamics. These findings provide design principles for programmable, low-hysteresis NiTi SMAs.
- 1 November 2026
- Published by IOP Publishing in Smart Materials and Structures
- Vol. 35 (11) , 115025
Open AccessCrystalline–amorphous architectures modulate martensitic transformation and mechanical response in NiTi
NiTi shape memory alloys (SMAs) exhibit excellent superelasticity, but their first-order martensitic transformation (MT) produces large hysteresis that limits functional performance. Recent experiments have shown that crystalline–amorphous (C + A) NiTi composites can significantly reduce hysteresis, yet the underlying microstructural mechanisms remain unclear. Here, phase-field simulations are performed to investigate stress-induced MT in C + A NiTi with bi-continuous and nanolaminate architectures. The effects of crystalline channel width, amorphous morphology, and crystalline/amorphous (C/A) interface orientation are systematically quantified. Reducing the crystalline channel width suppresses martensitic avalanches, transforming the stress–strain response from plateau-like pseudoelasticity to nearly linear superelasticity with vanishing hysteresis. The critical transformation stress can be tuned from 125 MPa to above 800 MPa, while the hysteresis decreases from approximately 500 MPa to nearly zero. Bi-continuous architectures exhibit robust low-hysteresis behavior over a wide range of channel widths, whereas in nanolaminates the C/A interface orientation relative to the habit plane between the parent austenite and the twinned martensite provides an additional mechanism for controlling elastic confinement. The microstructural effects are quantitatively rationalized through the elastic interaction energy, establishing a direct link between microstructural architecture and MT thermodynamics. These findings provide design principles for programmable, low-hysteresis NiTi SMAs.
- Open Access
- 14 August 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
A superelastic titanium alloy showing nearly temperature-independent high strength and elastic modulus
Lightweight superelastic alloys with temperature-insensitive properties are critical for deep-space exploration devices, yet their development is hindered by the challenge of simultaneously balancing strength, recoverable strain, and temperature sensitivity. Here, we overcome these limitations in a Ti50.5Zr40Nb6Sn2O1.5 alloy through Zr/O co-doping to induce local chemical orders (LCOs). This strategy yields an unprecedented combination of large recovery strain (εr > 4.5%), high superelastic stress (σc > 700 MPa), nearly temperature-independent strength (dσc/dT ≈ 0.59 MPa·K⁻¹), and stable elastic modulus over a wide temperature range from 123 K to 298 K. These properties arise from the elastic confinement imposed on the transformable matrix induced by the LCOs, which enhances resistance to dislocation slip and suppresses martensitic transformation, while the modulus hardening caused by LCOs due to anharmonic atomic vibrations compensates for matrix softening. This work provides a feasible and effective route for developing high-performance superelastic alloys with wide-temperature stability through LCO engineering.
- 14 August 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessA superelastic titanium alloy showing nearly temperature-independent high strength and elastic modulus
Lightweight superelastic alloys with temperature-insensitive properties are critical for deep-space exploration devices, yet their development is hindered by the challenge of simultaneously balancing strength, recoverable strain, and temperature sensitivity. Here, we overcome these limitations in a Ti50.5Zr40Nb6Sn2O1.5 alloy through Zr/O co-doping to induce local chemical orders (LCOs). This strategy yields an unprecedented combination of large recovery strain (εr > 4.5%), high superelastic stress (σc > 700 MPa), nearly temperature-independent strength (dσc/dT ≈ 0.59 MPa·K⁻¹), and stable elastic modulus over a wide temperature range from 123 K to 298 K. These properties arise from the elastic confinement imposed on the transformable matrix induced by the LCOs, which enhances resistance to dislocation slip and suppresses martensitic transformation, while the modulus hardening caused by LCOs due to anharmonic atomic vibrations compensates for matrix softening. This work provides a feasible and effective route for developing high-performance superelastic alloys with wide-temperature stability through LCO engineering.
- 6 July 2026
- Published by IOP Publishing in Modelling and Simulation in Materials Science and Engineering
- Vol. 34 (5) , 055012
A highly precise surrogate model for predicting microstructure evolution via integrated data augmentation technologies
Phase field modeling is powerful in predicting microstructures and their spatiotemporal evolutions. However, performing highly precise phase field calculations usually incurs considerable computational time cost and substantial computational resources, which is especially true for large space and time scale simulations. Recently, surrogate models based on the machine learning method have emerged as a promising tool for accelerating phase field simulations, but their accuracy and efficiency are often constrained by the lack of sufficient training data and reliable data augmentation techniques. In this study, we propose a highly precise surrogate model that incorporates the U-Net architecture and four data augmentation technologies, including periodic cropping, periodic padding, flipping and rotation, for periodic problems. After training the surrogate model, the test error is reduced to the order of 10 −5 , which is around three orders lower than that of the state-of-the-art models. The high precision of this model enables us to make long time prediction based purely on the surrogate model, in sharp contrast to previous surrogate models that rely on the aid of phase field calculations to reduce error accumulation. In autoregressive prediction, this advance reduces the computational time from 9 mins for phase field simulation to 1 second for surrogate model predictions. This study provides not only a general framework for developing highly precise surrogate models but also a powerful tool for predicting microstructures of materials.
- 6 July 2026
- Published by IOP Publishing in Modelling and Simulation in Materials Science and Engineering
- Vol. 34 (5) , 055012
A highly precise surrogate model for predicting microstructure evolution via integrated data augmentation technologies
Phase field modeling is powerful in predicting microstructures and their spatiotemporal evolutions. However, performing highly precise phase field calculations usually incurs considerable computational time cost and substantial computational resources, which is especially true for large space and time scale simulations. Recently, surrogate models based on the machine learning method have emerged as a promising tool for accelerating phase field simulations, but their accuracy and efficiency are often constrained by the lack of sufficient training data and reliable data augmentation techniques. In this study, we propose a highly precise surrogate model that incorporates the U-Net architecture and four data augmentation technologies, including periodic cropping, periodic padding, flipping and rotation, for periodic problems. After training the surrogate model, the test error is reduced to the order of 10 −5 , which is around three orders lower than that of the state-of-the-art models. The high precision of this model enables us to make long time prediction based purely on the surrogate model, in sharp contrast to previous surrogate models that rely on the aid of phase field calculations to reduce error accumulation. In autoregressive prediction, this advance reduces the computational time from 9 mins for phase field simulation to 1 second for surrogate model predictions. This study provides not only a general framework for developing highly precise surrogate models but also a powerful tool for predicting microstructures of materials.
- Open Access
- 20 November 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 10224
Achieving high strain hardening and strength in an additively manufactured titanium alloy
Strain hardening is a crucial property of metals and alloys that directly affects their mechanical processability, safe usage, and durability throughout their service life. However, titanium alloys traditionally used in structural applications often exhibit limited strain hardening, restricting their broader use. In this work, we demonstrate that by employing additive manufacturing (AM), strong strain hardening with high strength can be simultaneously achieved in a commercially available titanium alloy. These remarkable properties arise from a martensitic microstructure originated from the AM process. The microstructure is characterized by nanosized martensite plates with extremely fine triple-twinned substructures. During tensile deformation, detwinning rather than dislocation slip gradually transforms this microstructure into single-twinned lamellae with ~10 nm twin boundary spacing and internal stacking faults, necessitating progressively higher stresses and resulting in significant strain hardening.
- 20 November 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 10224
Open AccessAchieving high strain hardening and strength in an additively manufactured titanium alloy
Strain hardening is a crucial property of metals and alloys that directly affects their mechanical processability, safe usage, and durability throughout their service life. However, titanium alloys traditionally used in structural applications often exhibit limited strain hardening, restricting their broader use. In this work, we demonstrate that by employing additive manufacturing (AM), strong strain hardening with high strength can be simultaneously achieved in a commercially available titanium alloy. These remarkable properties arise from a martensitic microstructure originated from the AM process. The microstructure is characterized by nanosized martensite plates with extremely fine triple-twinned substructures. During tensile deformation, detwinning rather than dislocation slip gradually transforms this microstructure into single-twinned lamellae with ~10 nm twin boundary spacing and internal stacking faults, necessitating progressively higher stresses and resulting in significant strain hardening.
- Open Access
- 24 October 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (43)
Universal exact solutions for multiphysical inhomogeneities and inclusions in Fourier space
Physical inhomogeneities—spanning atomic-scale crystal defects and quantum dots to macroscopic features like bone cavities and geological faults—are fundamental to both engineered and natural systems. The study of how these inclusions influence material fields has driven critical advances across multiple disciplines, from composite materials and solid-state physics to biomechanics and geophysics. Eshelby revolutionized this field with the equivalent inclusion method (EIM), transforming inhomogeneous problems into homogeneous ones by introducing an equivalent eigen-strain. However, determining Eshelby’s equivalent eigen-strain for arbitrary inclusions has remained a long-standing challenge. Here, we resolve this challenge and applied the EIM to multiphysics problems, deriving three-dimensional universal exact solutions in Fourier space for the generalized equivalent eigenfield, encompassing eigen-strain, eigen-electric, and eigen-magnetic fields. These solutions apply to both single and multiple inclusions of arbitrary shape, inhomogeneity, and anisotropy. Our results yield exact expressions for the effective properties of magnetoelectroelastic composites. We demonstrate the framework’s versatility through complex multiphase applications, including a general expression for stress intensity factors of arbitrarily shaped cracks, a criterion for designing auxetic materials via engineered elastic inhomogeneities and an optimized pathway to enhance magnetoelectric coupling. This work establishes a universal approach to multiphysics inhomogeneous inclusion problems, rooted in multiple disciplines and spanning multiple length scales. It offers both profound fundamental insights and practical utility, paving the way for advancements in diverse fields.
- 24 October 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (43)
Open AccessUniversal exact solutions for multiphysical inhomogeneities and inclusions in Fourier space
Physical inhomogeneities—spanning atomic-scale crystal defects and quantum dots to macroscopic features like bone cavities and geological faults—are fundamental to both engineered and natural systems. The study of how these inclusions influence material fields has driven critical advances across multiple disciplines, from composite materials and solid-state physics to biomechanics and geophysics. Eshelby revolutionized this field with the equivalent inclusion method (EIM), transforming inhomogeneous problems into homogeneous ones by introducing an equivalent eigen-strain. However, determining Eshelby’s equivalent eigen-strain for arbitrary inclusions has remained a long-standing challenge. Here, we resolve this challenge and applied the EIM to multiphysics problems, deriving three-dimensional universal exact solutions in Fourier space for the generalized equivalent eigenfield, encompassing eigen-strain, eigen-electric, and eigen-magnetic fields. These solutions apply to both single and multiple inclusions of arbitrary shape, inhomogeneity, and anisotropy. Our results yield exact expressions for the effective properties of magnetoelectroelastic composites. We demonstrate the framework’s versatility through complex multiphase applications, including a general expression for stress intensity factors of arbitrarily shaped cracks, a criterion for designing auxetic materials via engineered elastic inhomogeneities and an optimized pathway to enhance magnetoelectric coupling. This work establishes a universal approach to multiphysics inhomogeneous inclusion problems, rooted in multiple disciplines and spanning multiple length scales. It offers both profound fundamental insights and practical utility, paving the way for advancements in diverse fields.
- 7 August 2025
- Published by Elsevier in Scripta Materialia
- Vol. 269
Design of high yield strength stress-induced martensite titanium through cooling-induced martensite strengthening
- 7 August 2025
- Published by Elsevier in Scripta Materialia
- Vol. 269
Design of high yield strength stress-induced martensite titanium through cooling-induced martensite strengthening
- 26 May 2025
- Published by Springer Nature in International Journal of Minerals, Metallurgy and Materials
- Vol. 32 (6) , 1427-1440
Multiphase field modeling of austenite to pearlite–ferrite transformation in hypoeutectoid steel
- 26 May 2025
- Published by Springer Nature in International Journal of Minerals, Metallurgy and Materials
- Vol. 32 (6) , 1427-1440
Multiphase field modeling of austenite to pearlite–ferrite transformation in hypoeutectoid steel
- Open Access
- 24 March 2025
- Published by Elsevier in Journal of Materials Research and Technology
- Vol. 36, 1683-1689
Microstructure evolution of austenite-to-martensite transformation in low-alloy steel via thermodynamically assisted phase-field method
- 24 March 2025
- Published by Elsevier in Journal of Materials Research and Technology
- Vol. 36, 1683-1689
Open AccessMicrostructure evolution of austenite-to-martensite transformation in low-alloy steel via thermodynamically assisted phase-field method
- Open Access
- 10 February 2025
- Published by OAE Publishing Inc. in Journal of Materials Informatics
- Vol. 5 (1)
An optimized strategy for density prediction of intermetallics across varied crystal structures via graph neural network
Intermetallic compounds are crucial in modern industry due to their exceptional properties, where density is identified as a critical parameter determining their potentiality for lightweight applications. In this study, over 7,000 density data points are collected for binary intermetallic compounds from different crystal structures. A new intermetallics graph neural network (IGNN) model is developed to perform regression and classification tasks for density prediction. Compared to traditional machine learning models, the IGNN model demonstrated superior capability in capturing crystal structure and effectively addressing challenges posed by polymorphism. The interpretability of the IGNN model classification process is enhanced through the t-distributed stochastic neighbor embedding (t-SNE) visualization method. Additionally, the IGNN model exhibited excellent performance in predicting the density of multicomponent complex intermetallic compounds, indicating its robustness and generalizability. This study presents a graph neural network (GNN) method suitable for multi-crystal structure data modeling, providing a novel computational framework for density prediction in intermetallic compounds. This advancement represents a significant contribution to this field, paving the way for more targeted material selection and application in lightweight technologies.
- 10 February 2025
- Published by OAE Publishing Inc. in Journal of Materials Informatics
- Vol. 5 (1)
Open AccessAn optimized strategy for density prediction of intermetallics across varied crystal structures via graph neural network
Intermetallic compounds are crucial in modern industry due to their exceptional properties, where density is identified as a critical parameter determining their potentiality for lightweight applications. In this study, over 7,000 density data points are collected for binary intermetallic compounds from different crystal structures. A new intermetallics graph neural network (IGNN) model is developed to perform regression and classification tasks for density prediction. Compared to traditional machine learning models, the IGNN model demonstrated superior capability in capturing crystal structure and effectively addressing challenges posed by polymorphism. The interpretability of the IGNN model classification process is enhanced through the t-distributed stochastic neighbor embedding (t-SNE) visualization method. Additionally, the IGNN model exhibited excellent performance in predicting the density of multicomponent complex intermetallic compounds, indicating its robustness and generalizability. This study presents a graph neural network (GNN) method suitable for multi-crystal structure data modeling, providing a novel computational framework for density prediction in intermetallic compounds. This advancement represents a significant contribution to this field, paving the way for more targeted material selection and application in lightweight technologies.