Yi Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Yi Zhou has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- GuanHua Chen
Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Hong Kong, China; Chinese National Engineering Research Center for Control & Treatment of Heavy Metal Pollution (Hong Kong Branch), The Hong Kong University of Science and Technology, Hong Kong, China; Water Technology Center, The Hong Kong University of Science and Technology, Hong Kong, China
4 shared publications - Antonietta De Sio
Institut für Physik, Carl von Ossietzky Universität, Oldenburg, Germany
2 shared publications - Christoph Lienau
Carl von Ossietzky Universität Oldenburg
2 shared publications Jiang Wu
Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong SAR, China
2 shared publicationsLynn Groß
Bremen Center for Computational Materials Science, University of Bremen, Am Fallturm 1, 28359 Bremen, Germany
2 shared publicationsThomas Frauenheim
Shenzhen JL Computational Science and Applied Research Institute, Shenzhen 518110, China
2 shared publications
Latest Publications
- 2 September 2022
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 126 (36) , 6295-6300
Predicting Experimental Heats of Formation via Deep Learning with Limited Experimental Data
When it comes to predicting experimental values of molecular properties with deep learning, the key problem is the lack of sufficient experimental data for training. We propose a method that consists of pretraining a graph neural network that aims to reproduce first-principles quantum mechanical results, followed by fine-tuning of a fully connected neural network against experimental results. The combined pretraining and fine-tuning model is expected to yield molecular properties close to experimental accuracy. This is made possible because first-principles quantum mechanical methods are often qualitatively correct or semiquantitatively accurate; thus, a calibration of the calculation results against high-precision but limited experiment data can improve accuracy greatly. Moreover, the method is highly efficient, as first-principles quantum mechanical calculation is bypassed. To demonstrate this, we apply the combined model to determine the experimental heats of formation of organic molecules made of H, C, O, N, or F atoms (up to 30 atoms), where mere 405 experimental data are used. The overall mean absolute error is 1.8 kcal/mol for these molecules.
- 2 September 2022
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 126 (36) , 6295-6300
Predicting Experimental Heats of Formation via Deep Learning with Limited Experimental Data
When it comes to predicting experimental values of molecular properties with deep learning, the key problem is the lack of sufficient experimental data for training. We propose a method that consists of pretraining a graph neural network that aims to reproduce first-principles quantum mechanical results, followed by fine-tuning of a fully connected neural network against experimental results. The combined pretraining and fine-tuning model is expected to yield molecular properties close to experimental accuracy. This is made possible because first-principles quantum mechanical methods are often qualitatively correct or semiquantitatively accurate; thus, a calibration of the calculation results against high-precision but limited experiment data can improve accuracy greatly. Moreover, the method is highly efficient, as first-principles quantum mechanical calculation is bypassed. To demonstrate this, we apply the combined model to determine the experimental heats of formation of organic molecules made of H, C, O, N, or F atoms (up to 30 atoms), where mere 405 experimental data are used. The overall mean absolute error is 1.8 kcal/mol for these molecules.
- Open Access
- 18 June 2021
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 7 (25) , eabf7672
Revealing generation, migration, and dissociation of electron-hole pairs and current emergence in an organic photovoltaic cell
Photoexcitation, electron-hole pair migration, charge separation, and current generation are simulated quantum mechanically.
- 18 June 2021
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 7 (25) , eabf7672
Open AccessRevealing generation, migration, and dissociation of electron-hole pairs and current emergence in an organic photovoltaic cell
Photoexcitation, electron-hole pair migration, charge separation, and current generation are simulated quantum mechanically.
- 8 November 2019
- Published by American Chemical Society (ACS) in Nano Letters
- Vol. 19 (12) , 8630-8637
Coherent Real-Space Charge Transport Across a Donor–Acceptor Interface Mediated by Vibronic Couplings
There is growing experimental and theoretical evidence that vibronic couplings, couplings between electronic and nuclear degrees of freedom, play a fundamental role for ultrafast excited state dynamics in organic donor–acceptor hybrids. While vibronic coupling has been shown to support charge separation at donor–acceptor interfaces, so far little is known about its role for the real-space transport of charges in such systems. Here, we theoretically study charge transport in thiophene:fullerene stacks using time-dependent density functional tight-binding theory combined with Ehrenfest molecular dynamics for open systems. Our results reveal coherent oscillations of the charge density between neighboring donor sites, persisting for about 200 fs and promoting charge transport within the polymer stacks. At the donor–acceptor interface, vibronic wave packets are launched, propagating coherently over distances of more than 3 nm into the acceptor region. This supports earlier experimental observations of long-range ballistic charge carrier motion in organic photovoltaic systems and highlights the importance of vibronic coupling engineering as a concept for tailoring the functionality of hybrid organic devices.
- 8 November 2019
- Published by American Chemical Society (ACS) in Nano Letters
- Vol. 19 (12) , 8630-8637
Coherent Real-Space Charge Transport Across a Donor–Acceptor Interface Mediated by Vibronic Couplings
There is growing experimental and theoretical evidence that vibronic couplings, couplings between electronic and nuclear degrees of freedom, play a fundamental role for ultrafast excited state dynamics in organic donor–acceptor hybrids. While vibronic coupling has been shown to support charge separation at donor–acceptor interfaces, so far little is known about its role for the real-space transport of charges in such systems. Here, we theoretically study charge transport in thiophene:fullerene stacks using time-dependent density functional tight-binding theory combined with Ehrenfest molecular dynamics for open systems. Our results reveal coherent oscillations of the charge density between neighboring donor sites, persisting for about 200 fs and promoting charge transport within the polymer stacks. At the donor–acceptor interface, vibronic wave packets are launched, propagating coherently over distances of more than 3 nm into the acceptor region. This supports earlier experimental observations of long-range ballistic charge carrier motion in organic photovoltaic systems and highlights the importance of vibronic coupling engineering as a concept for tailoring the functionality of hybrid organic devices.
- 5 November 2019
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 10 (22) , 7264-7269
Toward the Exact Exchange–Correlation Potential: A Three-Dimensional Convolutional Neural Network Construct
A deep neural network is constructed to yield in-principle exact exchange-correlation potential. It requires merely the electron densities of small molecules and ions, and yet, can determine the exchange-correlation potentials of large molecules. We train and validate the neural network based on the data for H2 and HeH+, and subsequently determine the ground state electron density of stretched HeH+, linear H3+ and H–He–He–H2+. Moreover, the deep neural network is proven to model the van der Waals interaction by being trained and validated on dataset containing He2. Comparisons to B3LYP are given to illustrate the accuracy and transferability of the neural network.
- 5 November 2019
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 10 (22) , 7264-7269
Toward the Exact Exchange–Correlation Potential: A Three-Dimensional Convolutional Neural Network Construct
A deep neural network is constructed to yield in-principle exact exchange-correlation potential. It requires merely the electron densities of small molecules and ions, and yet, can determine the exchange-correlation potentials of large molecules. We train and validate the neural network based on the data for H2 and HeH+, and subsequently determine the ground state electron density of stretched HeH+, linear H3+ and H–He–He–H2+. Moreover, the deep neural network is proven to model the van der Waals interaction by being trained and validated on dataset containing He2. Comparisons to B3LYP are given to illustrate the accuracy and transferability of the neural network.
- 5 February 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 66 (3) , 1167-1173
Fundamental Limit to Scaling Si Field-Effect Transistors Due to Source-to-Drain Direct Tunneling
How far can the miniaturization of metal-oxide-semiconductor field-effect transistors (MOSFETs) continue is a recurring question, essential to all aspects of digital technology. Recent claims of well-performing MOSFETs with gate lengths below 4 nm apparently defy the fundamental limit of source-to-drain direct tunneling (SDDT). Here, we investigate that limit by simulating gate-all-around Si nanowire FETs with gate lengths between 8 and 3 nm using the state-of-the-art atomistic quantum transport modeling. We find that at 3-nm gate length, the current is dominated by SDDT, resulting in large source-drain leakage and poor switching performance even if the gate modulates the potential barrier between the source and drain sufficiently well. However, at 6-nm gate-length SDDT barely starts to degrade the subthreshold characteristics at large drain bias, and the ballistic on-/off-current ratio is ~ 10⁶ with a subthreshold swing of 70 mV/decade, on par with contemporary Si technology. This means that in the best case, the technology roadmap could potentially be extended for several generations beyond the currently projected nodes. In addition, the results substantiate that the experimental devices with the claimed gate lengths below 6 nm in fact operate with a longer effective gate lengths.
- 5 February 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 66 (3) , 1167-1173
Fundamental Limit to Scaling Si Field-Effect Transistors Due to Source-to-Drain Direct Tunneling
How far can the miniaturization of metal-oxide-semiconductor field-effect transistors (MOSFETs) continue is a recurring question, essential to all aspects of digital technology. Recent claims of well-performing MOSFETs with gate lengths below 4 nm apparently defy the fundamental limit of source-to-drain direct tunneling (SDDT). Here, we investigate that limit by simulating gate-all-around Si nanowire FETs with gate lengths between 8 and 3 nm using the state-of-the-art atomistic quantum transport modeling. We find that at 3-nm gate length, the current is dominated by SDDT, resulting in large source-drain leakage and poor switching performance even if the gate modulates the potential barrier between the source and drain sufficiently well. However, at 6-nm gate-length SDDT barely starts to degrade the subthreshold characteristics at large drain bias, and the ballistic on-/off-current ratio is ~ 10⁶ with a subthreshold swing of 70 mV/decade, on par with contemporary Si technology. This means that in the best case, the technology roadmap could potentially be extended for several generations beyond the currently projected nodes. In addition, the results substantiate that the experimental devices with the claimed gate lengths below 6 nm in fact operate with a longer effective gate lengths.