Jun Wang
Jun Wang has a total of 67 co-authors. Here are the top co-authors by number of shared publications:
- Guodong Xia
MOE Key Laboratory of Enhanced Heat Transfer and Energy Conservation, Beijing Key Laboratory of Heat Transfer and Energy Conversion, Beijing University of Technology, Beijing 100124, China
16 shared publications Guodong Xia
15 shared publications- Zhigang Li
Department of Biostatistics, University of Florida, Gainesville, USA
11 shared publications - Kai Qi
Beijing University of Technology
7 shared publications Haiyang Li
School of Mechanical and Power Engineering
6 shared publications- Shuang Luo
Department of Chemical Engineering
5 shared publications
Latest Publications
- Open Access
- 20 September 2026
- Published by Elsevier in SSRN Electronic Journal
Passive Desalination by Solar Photothermal Conversion and Janus Membrane
Solar driven desalination offers potential for decentralized freshwater, but thermal losses and slow water production limit its practical use. In this work, a passive desalination strategy is developed by integrating solar photothermal heating with an asymmetric wettability Janus membrane. A millimeter sized confined seawater layer localizes solar energy, minimizes bulk heat dissipation, and sustains the thermal gradient required for continuous vapor transport. Prepared by gas phase deposition of carboxyl functionalized carbon nanotubes onto a hydrophobic polytetrafluoroethylene substrate, the Janus membrane couples a capillary-assisted water supply on its hydrophilic side with vapor transport through its hydrophobic layer. This asymmetry creates an in-pore evaporation pathway that promotes thermo-osmotic transport, and it can contribute to the decrease in evaporation enthalpy based on molecular dynamics simulations. Indoor and outdoor experiments demonstrate stable freshwater production and effective salt rejection without external energy input, while solar concentration can substantially enhance water productivity. By unifying localized solar heating, capillary-driven in-pore evaporation, and directional thermo-osmotic transport, this work establishes a robust membrane-thermal coupling strategy for efficient, passive, and off-grid desalination. It is believed that this work could provide a theoretical guide for engineering applications of solar seawater desalination systems utilizing low-grade energy.
- 20 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessPassive Desalination by Solar Photothermal Conversion and Janus Membrane
Solar driven desalination offers potential for decentralized freshwater, but thermal losses and slow water production limit its practical use. In this work, a passive desalination strategy is developed by integrating solar photothermal heating with an asymmetric wettability Janus membrane. A millimeter sized confined seawater layer localizes solar energy, minimizes bulk heat dissipation, and sustains the thermal gradient required for continuous vapor transport. Prepared by gas phase deposition of carboxyl functionalized carbon nanotubes onto a hydrophobic polytetrafluoroethylene substrate, the Janus membrane couples a capillary-assisted water supply on its hydrophilic side with vapor transport through its hydrophobic layer. This asymmetry creates an in-pore evaporation pathway that promotes thermo-osmotic transport, and it can contribute to the decrease in evaporation enthalpy based on molecular dynamics simulations. Indoor and outdoor experiments demonstrate stable freshwater production and effective salt rejection without external energy input, while solar concentration can substantially enhance water productivity. By unifying localized solar heating, capillary-driven in-pore evaporation, and directional thermo-osmotic transport, this work establishes a robust membrane-thermal coupling strategy for efficient, passive, and off-grid desalination. It is believed that this work could provide a theoretical guide for engineering applications of solar seawater desalination systems utilizing low-grade energy.
- 15 May 2026
- Published by Elsevier in International Journal of Thermal Sciences
- Vol. 228
The direction prediction of thermo-osmosis based on potential of mean force and machine learning
- 15 May 2026
- Published by Elsevier in International Journal of Thermal Sciences
- Vol. 228
The direction prediction of thermo-osmosis based on potential of mean force and machine learning
- 3 March 2026
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry C
- Vol. 130 (10) , 3846-3854
Simulating the Self-Assembly of MOF-5 Using Reactive Force Field (ReaxFF)
- 3 March 2026
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry C
- Vol. 130 (10) , 3846-3854
Simulating the Self-Assembly of MOF-5 Using Reactive Force Field (ReaxFF)
- Open Access
- 3 February 2026
- Published by Elsevier in SSRN Electronic Journal
The Direction Prediction of Thermo-osmosis Based on Potential of Mean Force and Machine Learning
Thermo-osmosis describes a fluid flow phenomenon in nanochannels. However, how to accurately predict its flow direction and magnitude remains a challenge. Conventional methods depend on complex calculations of the thermo-osmotic coefficient, which needs high computational cost. This work introduces a thermo-osmotic direction prediction method. The method employed molecular dynamics (MD) simulations and a Backpropagation (BP) neural network, which is optimized by the Grey Wolf Optimization (GWO) algorithm. It takes operating parameters as input and directly outputs the potential of mean force (PMF) distribution. This generates a criterion for predicting the direction of thermo-osmosis. The optimal neural network architecture is determined to be an 8-6 double-hidden layer configuration through a systematic structural search. Results show that the GWO-BP neural network significantly outperforms the BP, which achieves a coefficient of determination (R²) above 0.976 on the testing set. The residual analysis indicates no overfitting. Crucially, it accurately predicts both the PMF distribution and captures the direction reversal of thermo-osmosis under unseen operating conditions. It is believed that this work can provide an efficient and reliable method for the prediction of nanoscale thermo-osmosis transport.
- 3 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Direction Prediction of Thermo-osmosis Based on Potential of Mean Force and Machine Learning
Thermo-osmosis describes a fluid flow phenomenon in nanochannels. However, how to accurately predict its flow direction and magnitude remains a challenge. Conventional methods depend on complex calculations of the thermo-osmotic coefficient, which needs high computational cost. This work introduces a thermo-osmotic direction prediction method. The method employed molecular dynamics (MD) simulations and a Backpropagation (BP) neural network, which is optimized by the Grey Wolf Optimization (GWO) algorithm. It takes operating parameters as input and directly outputs the potential of mean force (PMF) distribution. This generates a criterion for predicting the direction of thermo-osmosis. The optimal neural network architecture is determined to be an 8-6 double-hidden layer configuration through a systematic structural search. Results show that the GWO-BP neural network significantly outperforms the BP, which achieves a coefficient of determination (R²) above 0.976 on the testing set. The residual analysis indicates no overfitting. Crucially, it accurately predicts both the PMF distribution and captures the direction reversal of thermo-osmosis under unseen operating conditions. It is believed that this work can provide an efficient and reliable method for the prediction of nanoscale thermo-osmosis transport.
- Open Access
- 1 January 2026
- Published by Acta Physica Sinica, Chinese Physical Society and Institute of Physics, Chinese Academy of Sciences in Acta Physica Sinica
- Vol. 75 (11)
Predicting the thermal conductivity of Janus nanofluids based on molecular dynamics simulation and machine learning
在悬浮液中添加纳米尺寸的金属、金属氧化物或非金属颗粒,可制备具有较高导热性能的纳米流体。Janus纳米流体是将表面性质经不对称修饰的Janus纳米颗粒添加到基础流体中所形成的纳米流体。相较于普通纳米流体,Janus纳米流体具有更高的热导率。本文将平衡分子动力学模拟方法与机器学习方法相结合,计算并分析了银、铜或铁的Janus纳米流体热导率。基于分子动力学模拟计算所得Janus纳米流体热导率数据,分别采用反向传播神经网络、支持向量回归、数据分组处理方法以及随机森林四种机器学习模型,用于预测Janus纳米流体的热导率。机器学习模型的输入参数包括纳米颗粒粒径、体积分数、Janus纳米颗粒差异性参数、纳米流体密度、纳米颗粒质量以及颗粒材料的热导率。通过对比各机器学习模型的预测结果与误差,验证了机器学习预测Janus纳米流体热导率的可行性。综合分析表明,数据分组处理方法在预测Janus纳米流体的热导率中表现最优,均方根误差为0.0058 W/(m·K),决定系数为0.9844,具有较高的预测精度。分析表明,各参数对Janus纳米流体热导率的正向影响规律与已有研究结论相符合,并得到了各影响参数的重要性排序。
- 1 January 2026
- Published by Acta Physica Sinica, Chinese Physical Society and Institute of Physics, Chinese Academy of Sciences in Acta Physica Sinica
- Vol. 75 (11)
Open AccessPredicting the thermal conductivity of Janus nanofluids based on molecular dynamics simulation and machine learning
在悬浮液中添加纳米尺寸的金属、金属氧化物或非金属颗粒,可制备具有较高导热性能的纳米流体。Janus纳米流体是将表面性质经不对称修饰的Janus纳米颗粒添加到基础流体中所形成的纳米流体。相较于普通纳米流体,Janus纳米流体具有更高的热导率。本文将平衡分子动力学模拟方法与机器学习方法相结合,计算并分析了银、铜或铁的Janus纳米流体热导率。基于分子动力学模拟计算所得Janus纳米流体热导率数据,分别采用反向传播神经网络、支持向量回归、数据分组处理方法以及随机森林四种机器学习模型,用于预测Janus纳米流体的热导率。机器学习模型的输入参数包括纳米颗粒粒径、体积分数、Janus纳米颗粒差异性参数、纳米流体密度、纳米颗粒质量以及颗粒材料的热导率。通过对比各机器学习模型的预测结果与误差,验证了机器学习预测Janus纳米流体热导率的可行性。综合分析表明,数据分组处理方法在预测Janus纳米流体的热导率中表现最优,均方根误差为0.0058 W/(m·K),决定系数为0.9844,具有较高的预测精度。分析表明,各参数对Janus纳米流体热导率的正向影响规律与已有研究结论相符合,并得到了各影响参数的重要性排序。
- 23 July 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
Self-Sustained Flow in Janus Nanochannels Based on Thermo-Osmosis
Thermo-osmosis refers to the fluid flow that occurs in response to thermal gradients. Typically, thermo-osmosis is fluid flow from high to low temperatures. However, in certain cases, the direction of the thermo-osmosis may be from low to high temperatures, i.e., negative thermo-osmosis. By combining the positive and negative thermo-osmosis in Janus nanochannels, a self-sustained flow system is proposed in this work, wherein one side of the Janus nanochannel has a high surface energy and the other side has a relatively low surface energy. Molecular dynamics simulations show that the fluid within the Janus nanochannel can be driven by isotropic thermo-osmotic flow without any external pump. Additionally, the conditions required to generate self-sustained flow are presented, and the parameters affecting self-sustained flow are also discussed. In conclusion, the Janus nanochannel is believed to drive the fluid flow only using externally generated cold or heat without the need for an additional pump. Consequently, a deeper understanding of self-sustained flow at the nanoscale is critical for improving the theoretical framework of thermally driven mass flow and various emerging industrial applications such as efficient and energy-saving water purification and efficient energy conversion.
- 23 July 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
Self-Sustained Flow in Janus Nanochannels Based on Thermo-Osmosis
Thermo-osmosis refers to the fluid flow that occurs in response to thermal gradients. Typically, thermo-osmosis is fluid flow from high to low temperatures. However, in certain cases, the direction of the thermo-osmosis may be from low to high temperatures, i.e., negative thermo-osmosis. By combining the positive and negative thermo-osmosis in Janus nanochannels, a self-sustained flow system is proposed in this work, wherein one side of the Janus nanochannel has a high surface energy and the other side has a relatively low surface energy. Molecular dynamics simulations show that the fluid within the Janus nanochannel can be driven by isotropic thermo-osmotic flow without any external pump. Additionally, the conditions required to generate self-sustained flow are presented, and the parameters affecting self-sustained flow are also discussed. In conclusion, the Janus nanochannel is believed to drive the fluid flow only using externally generated cold or heat without the need for an additional pump. Consequently, a deeper understanding of self-sustained flow at the nanoscale is critical for improving the theoretical framework of thermally driven mass flow and various emerging industrial applications such as efficient and energy-saving water purification and efficient energy conversion.
- 9 July 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (1) , 015403
Thermophoresis of Janus nanoparticles in liquids
In this work, we investigate the thermophoresis of Janus nanoparticles in liquids through molecular dynamics simulations. Thermophoretic forces acting on cylindrical Janus nanoparticles are calculated through MD simulations by varying the particle size, the heterogenous surface energies, and the mean temperature of the fluid. It is found that a transition from negative thermophoresis to positive thermophoresis occurs when these parameters are varied. The surface energies and mean temperature affect both the magnitude of the thermophoretic force and the transition from negative to positive thermophoresis. However, the particle size only influences the magnitude of the thermophoretic force, not the transition from negative to positive thermophoresis. Contributions of the total energy and pressure of fluid molecules are studied by evaluating the excess enthalpy density around the particle. A general scaling relationship for the thermophoretic force is developed using dimensionless numbers. This work advances the understanding of the thermophoresis of Janus nanoparticles and offers theoretical support for manipulating Janus nanoparticles in liquids by temperature gradients.
- 9 July 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (1) , 015403
Thermophoresis of Janus nanoparticles in liquids
In this work, we investigate the thermophoresis of Janus nanoparticles in liquids through molecular dynamics simulations. Thermophoretic forces acting on cylindrical Janus nanoparticles are calculated through MD simulations by varying the particle size, the heterogenous surface energies, and the mean temperature of the fluid. It is found that a transition from negative thermophoresis to positive thermophoresis occurs when these parameters are varied. The surface energies and mean temperature affect both the magnitude of the thermophoretic force and the transition from negative to positive thermophoresis. However, the particle size only influences the magnitude of the thermophoretic force, not the transition from negative to positive thermophoresis. Contributions of the total energy and pressure of fluid molecules are studied by evaluating the excess enthalpy density around the particle. A general scaling relationship for the thermophoretic force is developed using dimensionless numbers. This work advances the understanding of the thermophoresis of Janus nanoparticles and offers theoretical support for manipulating Janus nanoparticles in liquids by temperature gradients.
- 1 July 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (7)
The direction reversal of thermo-osmotic flow in a nanochannel
Thermo-osmosis describes the flow of a liquid or dense fluid along a surface caused by a temperature gradient. However, predicting the direction of thermo-osmotic flows at the nanoscale is challenging because the direction is usually affected by the solid–liquid coupling strength. Based on molecular dynamics simulations, this work investigates the mechanism of the direction reversal of thermo-osmosis for a liquid in a nanochannel by adjusting the coupling strength parameter (k). The thermo-osmotic coefficient and the thermo-osmotic force are calculated using microscopic and excess enthalpy methods. The results indicate a correlation between the direction reversal of thermo-osmosis and the thermo-osmotic force, which is dominated by the surface forces. For weak couplings (k ≤ 0.6), the driving force for fluid flow is from high to low temperatures (i.e., positive thermo-osmosis). For strong couplings (k > 0.6), the thermo-osmotic force is in the opposite direction (from low to high temperatures, i.e., negative thermo-osmosis). The direction reversal of thermo-osmosis occurs due to the reversal of the thermo-osmotic force. The results of the present paper can provide a deep understanding of the control of flows in nanochannels.
- 1 July 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (7)
The direction reversal of thermo-osmotic flow in a nanochannel
Thermo-osmosis describes the flow of a liquid or dense fluid along a surface caused by a temperature gradient. However, predicting the direction of thermo-osmotic flows at the nanoscale is challenging because the direction is usually affected by the solid–liquid coupling strength. Based on molecular dynamics simulations, this work investigates the mechanism of the direction reversal of thermo-osmosis for a liquid in a nanochannel by adjusting the coupling strength parameter (k). The thermo-osmotic coefficient and the thermo-osmotic force are calculated using microscopic and excess enthalpy methods. The results indicate a correlation between the direction reversal of thermo-osmosis and the thermo-osmotic force, which is dominated by the surface forces. For weak couplings (k ≤ 0.6), the driving force for fluid flow is from high to low temperatures (i.e., positive thermo-osmosis). For strong couplings (k > 0.6), the thermo-osmotic force is in the opposite direction (from low to high temperatures, i.e., negative thermo-osmosis). The direction reversal of thermo-osmosis occurs due to the reversal of the thermo-osmotic force. The results of the present paper can provide a deep understanding of the control of flows in nanochannels.
- 1 March 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (3)
Evaluation of drag and lift forces on nonspherical particles in a gas
The nonspherical particles suspended in a gas flow will experience drag and lift forces, which dominate the transport properties and dynamics behaviors of nonspherical particles. In this paper, forces on nonspherical particles suspended in a gas are evaluated using the direct simulation Monte Carlo method, wherein the Knudsen number is in the range of 0.01 ≤ Kn ≤ 50. The aspect ratios of nonspherical particles and the incident angles (the angle between the polar axis of a nonspherical particle and the direction of gas flow) are key factors that lead to the transport dynamics of nonspherical particles different from spherical particles. It is found that the theoretical formulas for the orientation dependence of the drag and lift forces in the continuum regime can be extended to the transition and free molecular regimes. Considering the random rotating of particles in a gas flow, the orientation-averaged drag force can be employed to describe the particle transports, which is found to be equal to the drag force on spherical particles with the equivalent spherical surface-area approximation.
- 1 March 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (3)
Evaluation of drag and lift forces on nonspherical particles in a gas
The nonspherical particles suspended in a gas flow will experience drag and lift forces, which dominate the transport properties and dynamics behaviors of nonspherical particles. In this paper, forces on nonspherical particles suspended in a gas are evaluated using the direct simulation Monte Carlo method, wherein the Knudsen number is in the range of 0.01 ≤ Kn ≤ 50. The aspect ratios of nonspherical particles and the incident angles (the angle between the polar axis of a nonspherical particle and the direction of gas flow) are key factors that lead to the transport dynamics of nonspherical particles different from spherical particles. It is found that the theoretical formulas for the orientation dependence of the drag and lift forces in the continuum regime can be extended to the transition and free molecular regimes. Considering the random rotating of particles in a gas flow, the orientation-averaged drag force can be employed to describe the particle transports, which is found to be equal to the drag force on spherical particles with the equivalent spherical surface-area approximation.