Houlong Zhuang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Houlong Zhuang has a total of 104 co-authors. Here are the top co-authors by number of shared publications:
Lei Liu
Arizona State Universty, Arizona State University, PO BOX 876106, TEMPE, TEMPE, Arizona, 85287-6106, UNITED STATES
22 shared publications- Yang Jiao
Arizona State University
11 shared publications - Duyu Chen
University of California
9 shared publications - Mohan Chen
Peking University
7 shared publications - Lei Liu
School for Engineering of Matter, Transport and Energy, Arizona State University , Tempe, Arizona 85287, USA
6 shared publications Yu Zheng
Department of Physics, Arizona State University, Tempe, Arizona 85287, USA
6 shared publications
Latest Publications
- Open Access
- 24 October 2025
- Published by Springer Nature in npj Computational Materials
- Vol. 11 (1) , 1-13
Interpretable ensemble learning for materials property prediction with classical interatomic potentials
Machine learning (ML) is widely used to explore crystal materials and predict their properties. However, the training is time-consuming for deep-learning models, and the regression process is a black box that is hard to interpret. Also, the preprocess to transfer a crystal structure into the input of ML, called descriptor, needs to be designed carefully. To efficiently predict important properties of materials, we propose an approach based on ensemble learning consisting of regression trees to predict formation energy and elastic constants based on small-size datasets of carbon allotropes as an example. Without using any descriptor, the inputs are the properties calculated by molecular dynamics with nine different classical interatomic potentials. Overall, the results from ensemble learning are more accurate than those from classical interatomic potentials, and ensemble learning can capture the relatively accurate properties from the nine classical potentials as criteria for predicting the final properties.
- 24 October 2025
- Published by Springer Nature in npj Computational Materials
- Vol. 11 (1) , 1-13
Open AccessInterpretable ensemble learning for materials property prediction with classical interatomic potentials
Machine learning (ML) is widely used to explore crystal materials and predict their properties. However, the training is time-consuming for deep-learning models, and the regression process is a black box that is hard to interpret. Also, the preprocess to transfer a crystal structure into the input of ML, called descriptor, needs to be designed carefully. To efficiently predict important properties of materials, we propose an approach based on ensemble learning consisting of regression trees to predict formation energy and elastic constants based on small-size datasets of carbon allotropes as an example. Without using any descriptor, the inputs are the properties calculated by molecular dynamics with nine different classical interatomic potentials. Overall, the results from ensemble learning are more accurate than those from classical interatomic potentials, and ensemble learning can capture the relatively accurate properties from the nine classical potentials as criteria for predicting the final properties.
- Open Access
- 9 September 2025
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 17 (38) , 53547-53562
H2O and CO2 Sorption in Ion-Exchange Sorbents: Distinct Interactions in Amine Versus Quaternary Ammonium Materials
This study investigates the H2O and CO2 sorption behavior of two chemically distinct polystyrene-divinylbenzene-based ion exchange sorbents: a primary amine and a permanently charged strong base quaternary ammonium (QA+) group with (bi)carbonate counter anions. We compare their distinct interactions with H2O and CO2 through simultaneous thermal gravimetric, calorimetric, gas analysis, and molecular modeling approaches to evaluate their performance for dilute CO2 separations like direct air capture. Thermal and hybrid (heat + low-temperature hydration) desorption experiments demonstrate that the QA+-based sorbent binds both water and CO2 more strongly than the amine counterparts but undergoes degradation at moderate temperatures, limiting its compatibility with thermal swing regeneration. However, a low-temperature moisture-driven regeneration pathway is uniquely effective for the QA+-based sorbent. To inform the energetics of a moisture-based CO2 separation (i.e., a moisture swing), we compare calorimetric water sorption enthalpies to Clausius–Clapeyron-derived total isosteric enthalpies. To our knowledge, this includes the first direct calorimetric measurement of water sorption enthalpy in a QA+-based sorbent. Both methods reveal monolayer-multilayer sorption behavior for both sorbents, with the QA+-based material having slightly higher water sorption enthalpies at the initially occupied strongest sorption sites. Molecular modeling supports this observation, showing higher water sorption energies and denser charge distributions in the QA+-based sorbent at λH2O = 1 mmol/mmolsite. Mixed gas experiments in the QA+-based sorbent show that not only does water influence CO2 binding, but CO2 influences water uptake through counterion-dependent hydration states, and that moisture swing responsiveness in this material causes hydration-induced CO2 release and drying-induced CO2 uptake, an important feature for low-energy CO2 separation under ambient conditions. Overall, the two classes of sorbents offer distinct pathways for the CO2 separation.
- 9 September 2025
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 17 (38) , 53547-53562
Open AccessH2O and CO2 Sorption in Ion-Exchange Sorbents: Distinct Interactions in Amine Versus Quaternary Ammonium Materials
This study investigates the H2O and CO2 sorption behavior of two chemically distinct polystyrene-divinylbenzene-based ion exchange sorbents: a primary amine and a permanently charged strong base quaternary ammonium (QA+) group with (bi)carbonate counter anions. We compare their distinct interactions with H2O and CO2 through simultaneous thermal gravimetric, calorimetric, gas analysis, and molecular modeling approaches to evaluate their performance for dilute CO2 separations like direct air capture. Thermal and hybrid (heat + low-temperature hydration) desorption experiments demonstrate that the QA+-based sorbent binds both water and CO2 more strongly than the amine counterparts but undergoes degradation at moderate temperatures, limiting its compatibility with thermal swing regeneration. However, a low-temperature moisture-driven regeneration pathway is uniquely effective for the QA+-based sorbent. To inform the energetics of a moisture-based CO2 separation (i.e., a moisture swing), we compare calorimetric water sorption enthalpies to Clausius–Clapeyron-derived total isosteric enthalpies. To our knowledge, this includes the first direct calorimetric measurement of water sorption enthalpy in a QA+-based sorbent. Both methods reveal monolayer-multilayer sorption behavior for both sorbents, with the QA+-based material having slightly higher water sorption enthalpies at the initially occupied strongest sorption sites. Molecular modeling supports this observation, showing higher water sorption energies and denser charge distributions in the QA+-based sorbent at λH2O = 1 mmol/mmolsite. Mixed gas experiments in the QA+-based sorbent show that not only does water influence CO2 binding, but CO2 influences water uptake through counterion-dependent hydration states, and that moisture swing responsiveness in this material causes hydration-induced CO2 release and drying-induced CO2 uptake, an important feature for low-energy CO2 separation under ambient conditions. Overall, the two classes of sorbents offer distinct pathways for the CO2 separation.
- 11 March 2025
- Published by IOP Publishing in Modelling and Simulation in Materials Science and Engineering
- Vol. 33 (3) , 035005
Kolmogorov–Arnold neural networks for high-entropy alloys design
A wide range of deep learning-based machine learning (ML) techniques are extensively applied to the design of high-entropy alloys (HEAs), yielding numerous valuable insights. Kolmogorov–Arnold networks (KAN) is a recently developed architecture that aims to improve both the accuracy and interpretability of input features. In this work, we explore three different datasets for HEA design and demonstrate the application of KAN for both classification and regression models. In the first example, we use a KAN classification model to predict the probability of single-phase formation in high-entropy carbide ceramics based on various properties such as mixing enthalpy and valence electron concentration. In the second example, we employ a KAN regression model to predict the yield strength and ultimate tensile strength of HEAs based on their chemical composition and process conditions including annealing time, cold rolling percentage, and homogenization temperature. The third example involves a KAN classification model to determine whether a certain composition is an HEA or non-HEA, followed by a KAN regressor model to predict the bulk modulus of the identified HEA, aiming to identify HEAs with high bulk modulus. In all three examples, KAN either outperform or match the performance in terms of accuracy such as F1 score for classification and mean square error, and coefficient of determination (R 2) for regression of the multilayer perceptron by demonstrating the efficacy of KAN in handling both classification and regression tasks. We provide a promising direction for future research to explore advanced ML techniques, which lead to more accurate predictions and better interpretability of complex materials, ultimately accelerating the discovery and optimization of HEAs with desirable properties.
- 11 March 2025
- Published by IOP Publishing in Modelling and Simulation in Materials Science and Engineering
- Vol. 33 (3) , 035005
Kolmogorov–Arnold neural networks for high-entropy alloys design
A wide range of deep learning-based machine learning (ML) techniques are extensively applied to the design of high-entropy alloys (HEAs), yielding numerous valuable insights. Kolmogorov–Arnold networks (KAN) is a recently developed architecture that aims to improve both the accuracy and interpretability of input features. In this work, we explore three different datasets for HEA design and demonstrate the application of KAN for both classification and regression models. In the first example, we use a KAN classification model to predict the probability of single-phase formation in high-entropy carbide ceramics based on various properties such as mixing enthalpy and valence electron concentration. In the second example, we employ a KAN regression model to predict the yield strength and ultimate tensile strength of HEAs based on their chemical composition and process conditions including annealing time, cold rolling percentage, and homogenization temperature. The third example involves a KAN classification model to determine whether a certain composition is an HEA or non-HEA, followed by a KAN regressor model to predict the bulk modulus of the identified HEA, aiming to identify HEAs with high bulk modulus. In all three examples, KAN either outperform or match the performance in terms of accuracy such as F1 score for classification and mean square error, and coefficient of determination (R 2) for regression of the multilayer perceptron by demonstrating the efficacy of KAN in handling both classification and regression tasks. We provide a promising direction for future research to explore advanced ML techniques, which lead to more accurate predictions and better interpretability of complex materials, ultimately accelerating the discovery and optimization of HEAs with desirable properties.
- Open Access
Combining Reinforcement Learning with Graph Convolutional Neural Networks for Efficient Design of TiAl/TiAlN Atomic-Scale Interfaces
Ti/TiN coatings are utilized in a wide variety of engineering applications due to their superior properties such as high hardness and toughness. Doping Al into Ti/TiN can also enhance properties and lead to even higher performance. Therefore, studying the atomic-level behavior of the TiAl/TiAlN interface is important. However, due to the large number of possible combinations for the 50 mol% Al-doped Ti/TiN system, it is time-consuming to use the DFT-based Monte Carlo method to find the optimal TiAl/TiAlN system with a high work of adhesion. In this study, we use a graph convolutional neural network as an interatomic potential, combined with reinforcement learning, to improve the efficiency of finding optimal structures with a high work of adhesion. By inspecting the features of structures in neural networks, we found that the optimal structures follow a certain pattern of doping Al near the interface. The electronic structure and bonding analysis indicate that the optimal TiAl/TiAlN structures have higher bonding strength. We expect our approach to significantly accelerate the design of advanced ceramic coatings, which can lead to more durable and efficient materials for engineering applications.
Open AccessCombining Reinforcement Learning with Graph Convolutional Neural Networks for Efficient Design of TiAl/TiAlN Atomic-Scale Interfaces
Ti/TiN coatings are utilized in a wide variety of engineering applications due to their superior properties such as high hardness and toughness. Doping Al into Ti/TiN can also enhance properties and lead to even higher performance. Therefore, studying the atomic-level behavior of the TiAl/TiAlN interface is important. However, due to the large number of possible combinations for the 50 mol% Al-doped Ti/TiN system, it is time-consuming to use the DFT-based Monte Carlo method to find the optimal TiAl/TiAlN system with a high work of adhesion. In this study, we use a graph convolutional neural network as an interatomic potential, combined with reinforcement learning, to improve the efficiency of finding optimal structures with a high work of adhesion. By inspecting the features of structures in neural networks, we found that the optimal structures follow a certain pattern of doping Al near the interface. The electronic structure and bonding analysis indicate that the optimal TiAl/TiAlN structures have higher bonding strength. We expect our approach to significantly accelerate the design of advanced ceramic coatings, which can lead to more durable and efficient materials for engineering applications.
- Open Access
Kolmogorov-Arnold Neural Networks for High-Entropy Alloys Design
A wide range of deep learning-based machine learning techniques are extensively applied to the design of high-entropy alloys (HEAs), yielding numerous valuable insights. Kolmogorov-Arnold Networks (KAN) is a recently developed architecture that aims to improve both the accuracy and interpretability of input features. In this work, we explore three different datasets for HEA design and demonstrate the application of KAN for both classification and regression models. In the first example, we use a KAN classification model to predict the probability of single-phase formation in high-entropy carbide ceramics based on various properties such as mixing enthalpy and valence electron concentration. In the second example, we employ a KAN regression model to predict the yield strength and ultimate tensile strength of HEAs based on their chemical composition and process conditions including annealing time, cold rolling percentage, and homogenization temperature. The third example involves a KAN classification model to determine whether a certain composition is an HEA or non-HEA, followed by a KAN regressor model to predict the bulk modulus of the identified HEA, aiming to identify HEAs with high bulk modulus. In all three examples, KAN either outperform or match the performance in terms of accuracy such as F1 score for classification and Mean Square Error (MSE), and coefficient of determination (R2) for regression of the multilayer perceptron (MLP) by demonstrating the efficacy of KAN in handling both classification and regression tasks. We provide a promising direction for future research to explore advanced machine learning techniques, which lead to more accurate predictions and better interpretability of complex materials, ultimately accelerating the discovery and optimization of HEAs with desirable properties.
Open AccessKolmogorov-Arnold Neural Networks for High-Entropy Alloys Design
A wide range of deep learning-based machine learning techniques are extensively applied to the design of high-entropy alloys (HEAs), yielding numerous valuable insights. Kolmogorov-Arnold Networks (KAN) is a recently developed architecture that aims to improve both the accuracy and interpretability of input features. In this work, we explore three different datasets for HEA design and demonstrate the application of KAN for both classification and regression models. In the first example, we use a KAN classification model to predict the probability of single-phase formation in high-entropy carbide ceramics based on various properties such as mixing enthalpy and valence electron concentration. In the second example, we employ a KAN regression model to predict the yield strength and ultimate tensile strength of HEAs based on their chemical composition and process conditions including annealing time, cold rolling percentage, and homogenization temperature. The third example involves a KAN classification model to determine whether a certain composition is an HEA or non-HEA, followed by a KAN regressor model to predict the bulk modulus of the identified HEA, aiming to identify HEAs with high bulk modulus. In all three examples, KAN either outperform or match the performance in terms of accuracy such as F1 score for classification and Mean Square Error (MSE), and coefficient of determination (R2) for regression of the multilayer perceptron (MLP) by demonstrating the efficacy of KAN in handling both classification and regression tasks. We provide a promising direction for future research to explore advanced machine learning techniques, which lead to more accurate predictions and better interpretability of complex materials, ultimately accelerating the discovery and optimization of HEAs with desirable properties.
- 23 April 2024
- Published by Springer Nature in JOM
- Vol. 76 (6) , 3128-3142
A Phenomenological Model for Interstitial Hydrogen Absorption in Niobium
- 23 April 2024
- Published by Springer Nature in JOM
- Vol. 76 (6) , 3128-3142
A Phenomenological Model for Interstitial Hydrogen Absorption in Niobium
- 15 April 2024
- Published by IOP Publishing in Journal of Physics: Condensed Matter
- Vol. 36 (28) , 285703
Vibrational properties of disordered stealthy hyperuniform 1D atomic chains
Disorder hyperuniformity is a recently discovered exotic state of many-body systems that possess a hidden order in between that of a perfect crystal and a completely disordered system. Recently, this novel disordered state has been observed in a number of quantum materials including amorphous 2D graphene and silica, which are endowed with unexpected electronic transport properties. Here, we numerically investigate 1D atomic chain models, including perfect crystalline, disordered stealthy hyperuniform (SHU) as well as randomly perturbed atom packing configurations to obtain a quantitative understanding of how the unique SHU disorder affects the vibrational properties of these low-dimensional materials. We find that the disordered SHU chains possess lower cohesive energies compared to the randomly perturbed chains, implying their potential reliability in experiments. Our inverse partition ratio (IPR) calculations indicate that the SHU chains can support fully delocalized states just like perfect crystalline chains over a wide range of frequencies, i.e. ω ∈ ( 0 , 100 ) cm−1, suggesting superior phonon transport behaviors within these frequencies, which was traditionally considered impossible in disordered systems. Interestingly, we observe the emergence of a group of highly localized states associated with ω ∼ 200 cm−1, which is characterized by a significant peak in the IPR and a peak in phonon density of states at the corresponding frequency, and is potentially useful for decoupling electron and phonon degrees of freedom. These unique properties of disordered SHU chains have implications in the design and engineering of novel quantum materials for thermal and phononic applications.
- 15 April 2024
- Published by IOP Publishing in Journal of Physics: Condensed Matter
- Vol. 36 (28) , 285703
Vibrational properties of disordered stealthy hyperuniform 1D atomic chains
Disorder hyperuniformity is a recently discovered exotic state of many-body systems that possess a hidden order in between that of a perfect crystal and a completely disordered system. Recently, this novel disordered state has been observed in a number of quantum materials including amorphous 2D graphene and silica, which are endowed with unexpected electronic transport properties. Here, we numerically investigate 1D atomic chain models, including perfect crystalline, disordered stealthy hyperuniform (SHU) as well as randomly perturbed atom packing configurations to obtain a quantitative understanding of how the unique SHU disorder affects the vibrational properties of these low-dimensional materials. We find that the disordered SHU chains possess lower cohesive energies compared to the randomly perturbed chains, implying their potential reliability in experiments. Our inverse partition ratio (IPR) calculations indicate that the SHU chains can support fully delocalized states just like perfect crystalline chains over a wide range of frequencies, i.e. ω ∈ ( 0 , 100 ) cm−1, suggesting superior phonon transport behaviors within these frequencies, which was traditionally considered impossible in disordered systems. Interestingly, we observe the emergence of a group of highly localized states associated with ω ∼ 200 cm−1, which is characterized by a significant peak in the IPR and a peak in phonon density of states at the corresponding frequency, and is potentially useful for decoupling electron and phonon degrees of freedom. These unique properties of disordered SHU chains have implications in the design and engineering of novel quantum materials for thermal and phononic applications.
- Open Access
- 4 January 2024
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 135 (1)
Multi-principal element materials: Structure, property, and processing
Materials with multiple principal elements and under different names, such as high-entropy alloys (HEAs) and complex concentrated alloys (CCAs),1 are attracting
- 4 January 2024
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 135 (1)
Open AccessMulti-principal element materials: Structure, property, and processing
Materials with multiple principal elements and under different names, such as high-entropy alloys (HEAs) and complex concentrated alloys (CCAs),1 are attracting
- Open Access
A phenomenological model for interstitial hydrogen absorption in niobium
A phenomenological model has been developed for hydrogen absorption in niobium. The model has 9 free parameters that have a physical basis. The model provides an excellent fit to the highly accurate isotherm data by Veleckis et al. and has been cross validated by limiting the fitting procedure to a training set. The model makes it possible to extract more information from the data than could be extracted with the computational methods available when the measurements were made. The partial molal enthalpy and partial molal entropy of hydrogen dissolution were calculated using the parameter values corresponding to the best fit to the isotherms. These quantities are consistent with those reported by Veleckis et al. Interpreting the model parameter values reveals insights into the nature of the interaction between hydrogen atoms in niobium. These insights supports our previous analysis of hydrogen interactions in niobium using density functional theory calculations.
Open AccessA phenomenological model for interstitial hydrogen absorption in niobium
A phenomenological model has been developed for hydrogen absorption in niobium. The model has 9 free parameters that have a physical basis. The model provides an excellent fit to the highly accurate isotherm data by Veleckis et al. and has been cross validated by limiting the fitting procedure to a training set. The model makes it possible to extract more information from the data than could be extracted with the computational methods available when the measurements were made. The partial molal enthalpy and partial molal entropy of hydrogen dissolution were calculated using the parameter values corresponding to the best fit to the isotherms. These quantities are consistent with those reported by Veleckis et al. Interpreting the model parameter values reveals insights into the nature of the interaction between hydrogen atoms in niobium. These insights supports our previous analysis of hydrogen interactions in niobium using density functional theory calculations.
- Open Access
Vibrational Properties of One-Dimensional Disordered Hyperuniform Atomic Chains
Disorder hyperuniformity (DHU) is a recently discovered exotic state of many-body systems that possess a hidden order in between that of a perfect crystal and a completely disordered system. Recently, this novel DHU state has been observed in a number of quantum materials including amorphous 2D graphene and silica, which are endowed with unexpected electronic transport properties. Here, we numerically investigate 1D atomic chain models, including perfect crystalline, disordered hyperuniform as well as randomly perturbed atom packing configurations to obtain a quantitative understanding of how the unique DHU disorder affects the vibrational properties of these low-dimensional materials. We find that the DHU chains possess lower cohesive energies compared to the randomly perturbed chains, implying their potential reliability in experiments. Our inverse partition ratio (IPR) calculations indicate that the DHU chains can support fully delocalized states just like perfect crystalline chains over a wide range of frequencies, i.e., $ω\in (0, 100)$ cm$^{-1}$, suggesting superior phonon transport behaviors within these frequencies, which was traditionally considered impossible in disordered systems. Interestingly, we observe the emergence of a group of highly localized states associated with $ω\sim 200$ cm$^{-1}$, which is characterized by a significant peak in the IPR and a peak in phonon density of states at the corresponding frequency, and is potentially useful for decoupling electron and phonon degrees of freedom. These unique properties of DHU chains have implications in the design and engineering of novel DHU quantum materials for thermal and phononic applications.Open AccessVibrational Properties of One-Dimensional Disordered Hyperuniform Atomic Chains
Disorder hyperuniformity (DHU) is a recently discovered exotic state of many-body systems that possess a hidden order in between that of a perfect crystal and a completely disordered system. Recently, this novel DHU state has been observed in a number of quantum materials including amorphous 2D graphene and silica, which are endowed with unexpected electronic transport properties. Here, we numerically investigate 1D atomic chain models, including perfect crystalline, disordered hyperuniform as well as randomly perturbed atom packing configurations to obtain a quantitative understanding of how the unique DHU disorder affects the vibrational properties of these low-dimensional materials. We find that the DHU chains possess lower cohesive energies compared to the randomly perturbed chains, implying their potential reliability in experiments. Our inverse partition ratio (IPR) calculations indicate that the DHU chains can support fully delocalized states just like perfect crystalline chains over a wide range of frequencies, i.e., $ω\in (0, 100)$ cm$^{-1}$, suggesting superior phonon transport behaviors within these frequencies, which was traditionally considered impossible in disordered systems. Interestingly, we observe the emergence of a group of highly localized states associated with $ω\sim 200$ cm$^{-1}$, which is characterized by a significant peak in the IPR and a peak in phonon density of states at the corresponding frequency, and is potentially useful for decoupling electron and phonon degrees of freedom. These unique properties of DHU chains have implications in the design and engineering of novel DHU quantum materials for thermal and phononic applications.