A5102954368- Organic Electronics and Photovoltaics
- Conducting polymers and applications
- Molecular Junctions and Nanostructures
- Photochemistry and Electron Transfer Studies
- Organic Light-Emitting Diodes Research
- Machine Learning in Materials Science
- Liquid Crystal Research Advancements
- Domain Adaptation and Few-Shot Learning
- Synthesis and Properties of Aromatic Compounds
- Perovskite Materials and Applications
- Reinforcement Learning in Robotics
- Spectroscopy and Quantum Chemical Studies
- Advanced Chemical Physics Studies
- TiO2 Photocatalysis and Solar Cells
- Protein Structure and Dynamics
- Molecular spectroscopy and chirality
- Computational Drug Discovery Methods
- Neural Networks and Reservoir Computing
- Enzyme Structure and Function
- Fullerene Chemistry and Applications
- Quantum Dots Synthesis And Properties
- Model Reduction and Neural Networks
- Force Microscopy Techniques and Applications
- Semiconductor materials and interfaces
- Neural Networks and Applications
DeepMind (United Kingdom)
2015-2023
University of Oxford
2011-2023
Google (United Kingdom)
2018
Imperial College London
2005-2017
Google (United States)
2015-2016
Guangxi University for Nationalities
2009-2010
Commonwealth Scientific and Industrial Research Organisation
2009-2010
Stanford University
2010
Kyung Hee University
2009
Delft University of Technology
2009
The ability to learn tasks in a sequential fashion is crucial the development of artificial intelligence. Until now neural networks have not been capable this and it has widely thought that catastrophic forgetting an inevitable feature connectionist models. We show possible overcome limitation train can maintain expertise on they experienced for long time. Our approach remembers old by selectively slowing down learning weights important those tasks. demonstrate our scalable effective solving...
Learning to solve complex sequences of tasks--while both leveraging transfer and avoiding catastrophic forgetting--remains a key obstacle achieving human-level intelligence. The progressive networks approach represents step forward in this direction: they are immune forgetting can leverage prior knowledge via lateral connections previously learned features. We evaluate architecture extensively on wide variety reinforcement learning tasks (Atari 3D maze games), show that it outperforms common...
High-performance, solution-processed transistors fabricated from semiconducting polymers containing indacenodithiohene repeat units are described. The bridging functions on the backbone contribute to suppressing large-scale crystallization in thin films. However, charge carrier mobilities of up 1 cm2/(V s) for a benzothiadiazole copolymer were reported and, coupled with both ambient stability and long-wavelength absorption, make this family particularly attractive application next-generation...
Charge carrier dynamics in an organic semiconductor can often be described terms of charge hopping between localized states. The rates depend on electronic coupling elements, reorganization energies, and driving forces, which vary as a function position orientation the molecules. exact evaluation these contributions molecular assembly is computationally prohibitive. Various, semiempirical, approximations are employed instead. In this work, we review some approaches introduce software toolkit...
State-of-the-art models used for drift-diffusion simulations of organic bulk heterojunction solar cells based on band transport are not capable reproducing the voltage dependence dark current density and carrier concentration such devices, as determined by current-voltage charge-extraction measurements. Here, we show how to correctly reproduce this experimental data including an exponential tail localized states into both electrons holes, allowing recombination occur between free charge...
Abstract We describe AlphaFold, the protein structure prediction system that was entered by group A7D in CASP13. Submissions were made three free‐modeling (FM) methods which combine predictions of neural networks. All systems guided distances between pairs residues produced a network. Two assembled fragments generative network, one using scores from network trained to regress GDT_TS. The third shows simple gradient descent on properly constructed potential is able perform par with more...
Density functional theory describes matter at the quantum level, but all popular approximations suffer from systematic errors that arise violation of mathematical properties exact functional. We overcame this fundamental limitation by training a neural network on molecular data and fictitious systems with fractional charge spin. The resulting functional, DM21 (DeepMind 21), correctly typical examples artificial delocalization strong correlation performs better than traditional functionals...
Theoretical studies of charge transport in organic conducting systems pose a unique challenge since they must describe both extremely short-ranged and fast processes (charge tunneling) long-ranged slow ones (molecular ordering). The description the mobility electrons holes hopping regime relies on determination intermolecular rates large-scale morphologies. Using Marcus theory these can be calculated from transfer integrals on-site energies. Here we present detailed computational study...
Most deep reinforcement learning algorithms are data inefficient in complex and rich environments, limiting their applicability to many scenarios. One direction for improving efficiency is multitask with shared neural network parameters, where may be improved through transfer across related tasks. In practice, however, this not usually observed, because gradients from different tasks can interfere negatively, making unstable sometimes even less efficient. Another issue the reward schemes...
The performance of an organic photovoltaic cell depends critically on the mobility charge carriers within constituent molecular semiconductor materials. However, a complex combination phenomena that span range length and time scales control transport in disordered semiconductors. As result, it is difficult to rationalize properties terms material parameters. Until now, efforts improve mobilities semiconductors have proceeded largely by trial error rather than through systematic design....
The electrical performance of organic semiconducting polymers in field-effect transistor devices is now sufficient for initial low complexity circuit applications. To achieve high performance, either operation an inert atmosphere or a hydrophobic surface treatment and annealing step typically required. In this communication we report strategy to prepare fully air stable, amorphous which can charge carrier mobilities the range 0.04 cm2/(V s), remaining over period 3 months ambient conditions
We present a simple model of molecular photovoltaic device consisting two-level system, connected to external contacts by chains one or more charge transporting orbitals. Electrons may be promoted in the system photon absorption, and transported circuit electron transfer between neighboring Photon absorption emission are described generalized Planck equation is nonadiabatic Marcus theory. find steady-state current solving set coupled rate equations for under illumination as function bias...
Abstract In this article, we discuss a method for calculating transfer integrals pairs of molecules based on Zerner's Independent Neglect Differential Overlap Hamiltonian which requires only single self‐consistent field calculation an isolated molecule to be performed determine the integral pair molecules. This is compared with results obtained by projection molecules' molecular orbitals onto vector space defined each molecule. The two methods are found in good agreement using three...
A correlation is established between the molecular structure and charge mobility of discotic mesophases hexabenzocoronene derivatives by combining electronic calculations, dynamics, kinetic Monte Carlo simulations. It demonstrated that this multiscale approach can provide an accurate ab initio description transport in organic materials.
We study the kinetics of lateral hole transfer occurring between dye molecules anchored at surface metal oxide in Dye Sensitized Solar Cells (DSSC). use Marcus' charge rate equation for which we need electronic coupling two (J) and reorganization energy (λtot). In DSSC medium surrounding dyes is highly polar. This means that contribution solvent to cannot be neglected. Here elaborate a method calculate, from first principles, total (i.e., inner- outer-sphere) exchange ruthenium dyes. The...
Abstract The fused, bicyclic molecule, 2,1,3‐Benzothiadiazole (BT), has become a key ingredient in the design of new organic semiconductors for light emission and energy harvesting applications. Here, synthesis is reported series trigonal, star‐shaped compounds comprising truxene core three quater‐dialkylfluorene arms into each which BT unit inserted sequentially at possible position (T4BT‐A to T4BT‐E). Analysis resulting electronic properties shows that as consequence conjugative coupling...
In this report, we have studied the influence of nature end groups (bromine or hydrogen) regioregular poly(3-hexylthiophene) (P3HT) polymers on performance polymer solar cells made with blend films P3HT and 1-(3-methoxycarbonyl)-propyl-1-phenyl-(6,6)C61 (PCBM). Films devices were before after annealing at 140 °C for 2 h. The effects end-group type properties pristine examined using optical absorption emission spectroscopy, transient measurements photovoltaic device performance. It was...
Discotic mesophases are known for their ability to self-assemble into columnar structures and can serve as semiconducting molecular wires. Charge carrier mobility along these wires strongly depends on packing, which is controlled by intermolecular interactions. By combining wide-angle X-ray scattering experiments with dynamics simulations, we elucidate packing motifs of a perylene tetracarboxdiimide derivative, task hard achieve using single experimental or theoretical technique. We then...
By relating inter- and intrachain ordering to charge dynamics a correlation between the morphology mobility of neutral doped states conjugated polymer, in this case polypyrrole, is established. Morphologies are generated using an all-atom force field, while simulated within framework high temperature nonadiabatic Marcus theory. For short oligomers, carrier insensitive orientational molecular determined by threshold transfer integral which connects percolating clusters molecules, forming...