James William Ryan
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
Research Subjects
James William Ryan has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Emilio Palomares
Institute of Chemical Research of Catalonia (ICIQ-CERCA), Avda. Països Catalans, 16, 43007 Tarragona, Spain
13 shared publications - Eugenia Martínez-Ferrero
Institute of Chemical Research of Catalonia (ICIQ-CERCA), Avda. Països Catalans, 16, 43007 Tarragona, Spain
3 shared publications - Francesc Gispert-Guirado
Scientific Resources Service, Rovira i Virgili University, 43007 Tarragona, Spain
3 shared publications - Francisco J. Martin-Martinez
Department of Chemistry Swansea University Swansea UK
3 shared publications - Rafael Gómez-Bombarelli
Massachusetts Institute of Technology
3 shared publications - Yutaka Matsuo
Department of Chemical Systems Engineering, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8603, Japan
3 shared publications
Latest Publications
- Open Access
- 15 October 2026
- Published by Wiley in Advanced Intelligent Discovery
Embedded Morgan Fingerprints for More Efficient Molecular Property Predictions With Machine Learning
High‐dimensional molecular fingerprints have become a quiet bottleneck for chemical machine learning, where bit collisions in large, chemically diverse datasets and escalating computational costs constrain model discovery. This work introduces the embedded Morgan Fingerprint (eMFP), a postprocessing scheme that compresses standard Morgan Fingerprints into continuously valued, lower‐dimensional representations governed by a compression factor. Applied to three molecular datasets of increasing size (RedDB, Non‐Fullerene Acceptors, and QM9), five regression models, and four encoding configurations under scaffold‐aware nested cross‐validation, eMFP maintains the predictive power of full‐length fingerprints for tree‐based methods at moderate compression (), while exhibiting more model‐ and setting‐dependent behavior elsewhere. Its continuous outputs pair favorably with Fourier Feature Neural Networks, where they interact more naturally with sinusoidal feature mappings than the original binary fingerprints, and consistently shorten training times per trial. Together, these results position eMFP as a simple, drop‐in route to more scalable molecular representations that retain accuracy while easing the computational burden of model development.
- 15 October 2026
- Published by Wiley in Advanced Intelligent Discovery
Open AccessEmbedded Morgan Fingerprints for More Efficient Molecular Property Predictions With Machine Learning
High‐dimensional molecular fingerprints have become a quiet bottleneck for chemical machine learning, where bit collisions in large, chemically diverse datasets and escalating computational costs constrain model discovery. This work introduces the embedded Morgan Fingerprint (eMFP), a postprocessing scheme that compresses standard Morgan Fingerprints into continuously valued, lower‐dimensional representations governed by a compression factor. Applied to three molecular datasets of increasing size (RedDB, Non‐Fullerene Acceptors, and QM9), five regression models, and four encoding configurations under scaffold‐aware nested cross‐validation, eMFP maintains the predictive power of full‐length fingerprints for tree‐based methods at moderate compression (), while exhibiting more model‐ and setting‐dependent behavior elsewhere. Its continuous outputs pair favorably with Fourier Feature Neural Networks, where they interact more naturally with sinusoidal feature mappings than the original binary fingerprints, and consistently shorten training times per trial. Together, these results position eMFP as a simple, drop‐in route to more scalable molecular representations that retain accuracy while easing the computational burden of model development.
- Open Access
- 25 March 2026
- Published by American Chemical Society (ACS) in ACS Applied Engineering Materials
- Vol. 4 (4) , 1841-1851
Hyper-Cross-Linked Microporous Polymers as Cheap and Efficient Catalysts for the Synthesis of Biodiesel
In this work, we report a series of hyper-cross-linked sulfonated polymers, structurally related to PIMs, as highly efficient and versatile catalysts for biodiesel production. By tuning monomer composition to increase aromatic content, we generate porous polymers with tailored surface areas and porosity, which are thoroughly characterized and assessed for CO2/N2 separation. These catalysts promote both Fischer esterification of free fatty acids and transesterification of triglyceride oils derived from the same fatty acids, achieving over 95% conversion to fatty acid methyl esters (FAME) in 24 h or less under diverse conditions. Remarkably, by using a variety of oils, high activity is maintained even with reduced catalyst and methanol loadings, demonstrating intrinsic efficiency and robustness. The polymers prove to be fully scalable and recyclable, retaining performance over multiple cycles and efficiently converting waste cooking sunflower oil with comparable yields to pure edible oils. This work establishes a direct structure–property–performance relationship, linking polymer architecture and porosity to catalytic activity, and provides a versatile, sustainable platform for next-generation porous catalysts in biodiesel production and broader chemical transformations.
- 25 March 2026
- Published by American Chemical Society (ACS) in ACS Applied Engineering Materials
- Vol. 4 (4) , 1841-1851
Open AccessHyper-Cross-Linked Microporous Polymers as Cheap and Efficient Catalysts for the Synthesis of Biodiesel
In this work, we report a series of hyper-cross-linked sulfonated polymers, structurally related to PIMs, as highly efficient and versatile catalysts for biodiesel production. By tuning monomer composition to increase aromatic content, we generate porous polymers with tailored surface areas and porosity, which are thoroughly characterized and assessed for CO2/N2 separation. These catalysts promote both Fischer esterification of free fatty acids and transesterification of triglyceride oils derived from the same fatty acids, achieving over 95% conversion to fatty acid methyl esters (FAME) in 24 h or less under diverse conditions. Remarkably, by using a variety of oils, high activity is maintained even with reduced catalyst and methanol loadings, demonstrating intrinsic efficiency and robustness. The polymers prove to be fully scalable and recyclable, retaining performance over multiple cycles and efficiently converting waste cooking sunflower oil with comparable yields to pure edible oils. This work establishes a direct structure–property–performance relationship, linking polymer architecture and porosity to catalytic activity, and provides a versatile, sustainable platform for next-generation porous catalysts in biodiesel production and broader chemical transformations.
- Open Access
- 3 February 2025
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 4 (3) , 776-789
Embedded machine-readable molecular representation for resource-efficient deep learning applications
The implementation of an embedded encoding of molecular representations reduces memory and disk usage while maintaining performance in deep learning models.
- 3 February 2025
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 4 (3) , 776-789
Open AccessEmbedded machine-readable molecular representation for resource-efficient deep learning applications
The implementation of an embedded encoding of molecular representations reduces memory and disk usage while maintaining performance in deep learning models.
- Open Access
- 7 August 2024
- Published by American Chemical Society (ACS) in ChemRxiv
Embedded machine-readable molecular representationfor more resource-efficient deep learning applications
The practical implementation of Deep Learning methods for chemistry applications relies on encoding chemical structures into machine-readable formats that can be efficiently processed by computational tools. To this end, One Hot Encoding (OHE) is an established representation of alphanumeric categorical data into expanded numerical matrices. We have developed an embedded alternative to OHE that encodes discrete alphanumeric tokens of an N-sized alphabet into a few real numbers that constitute a simpler matrix representation of chemical structures. The implementation of this embedded One Hot Encoding (eOHE) in training machine learning models achieves comparable results to OHE in model accuracy and robustness, while significantly reducing the use of computational resources. Our benchmarks across three molecular representations (SMILES, DeepSMILES, and SELFIES) and three different molecular databases (ZINC, QM9, and GDB-13) for Variational Autoencoders (VAE) and Recurrent Neural Networks (RNN) show that using eOHE reduces RAM memory usage by up to 50%, while increasing disk Memory Reduction Efficiency (MRE) to 80% average. This encoding method opens new avenues for data representation into embedded formats that promote energy efficiency, and scalable computing in resource-constrained devices, or in scenarios with limited computing resources. The application of eOHE impacts not only the chemistry field, but other disciplines that rely on the use of OHE.
- 7 August 2024
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessEmbedded machine-readable molecular representationfor more resource-efficient deep learning applications
The practical implementation of Deep Learning methods for chemistry applications relies on encoding chemical structures into machine-readable formats that can be efficiently processed by computational tools. To this end, One Hot Encoding (OHE) is an established representation of alphanumeric categorical data into expanded numerical matrices. We have developed an embedded alternative to OHE that encodes discrete alphanumeric tokens of an N-sized alphabet into a few real numbers that constitute a simpler matrix representation of chemical structures. The implementation of this embedded One Hot Encoding (eOHE) in training machine learning models achieves comparable results to OHE in model accuracy and robustness, while significantly reducing the use of computational resources. Our benchmarks across three molecular representations (SMILES, DeepSMILES, and SELFIES) and three different molecular databases (ZINC, QM9, and GDB-13) for Variational Autoencoders (VAE) and Recurrent Neural Networks (RNN) show that using eOHE reduces RAM memory usage by up to 50%, while increasing disk Memory Reduction Efficiency (MRE) to 80% average. This encoding method opens new avenues for data representation into embedded formats that promote energy efficiency, and scalable computing in resource-constrained devices, or in scenarios with limited computing resources. The application of eOHE impacts not only the chemistry field, but other disciplines that rely on the use of OHE.
- 16 October 2020
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 2 (10) , 3088-3092
Organic Memristive Devices Based on Squaraine Nanowires
- 16 October 2020
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 2 (10) , 3088-3092
Organic Memristive Devices Based on Squaraine Nanowires
- 26 July 2018
- Published by Springer Nature in Journal of Materials Science: Materials in Electronics
- Vol. 29 (19) , 16410-16415
Organoboron donor-π-acceptor chromophores for small-molecule organic solar cells
- 26 July 2018
- Published by Springer Nature in Journal of Materials Science: Materials in Electronics
- Vol. 29 (19) , 16410-16415
Organoboron donor-π-acceptor chromophores for small-molecule organic solar cells
- 27 October 2017
- Published by Fundacio Scito
Understanding the Voc in Perovskite Solar Cells Using Photo-Induced Transient Optoelectronic Techniques
While rapid progress has been made in lead halide perovskite solar cells (PSCs) since the seminal papers from Snaith & Miyasaka and Park in 2012,1,2 much of the underlying device physics that makes these devices so efficient is not well understood. In order for PSCs to achieve their maximum efficiency it is necessary to clearly understand the key loss processes that govern their performance...
- 27 October 2017
- Published by Fundacio Scito
Understanding the Voc in Perovskite Solar Cells Using Photo-Induced Transient Optoelectronic Techniques
While rapid progress has been made in lead halide perovskite solar cells (PSCs) since the seminal papers from Snaith & Miyasaka and Park in 2012,1,2 much of the underlying device physics that makes these devices so efficient is not well understood. In order for PSCs to achieve their maximum efficiency it is necessary to clearly understand the key loss processes that govern their performance...
- 8 August 2017
- Published by Royal Society of Chemistry (RSC) in Journal of Materials Chemistry C
- Vol. 5 (34) , 8819-8827
Exploring the effects of interfacial carrier transport layers on device performance and optoelectronic properties of planar perovskite solar cells
The performance of perovskite device was found to be influenced by the interface quality and bulk defect activities induced in perovskite grown on HTL during device fabrication.
- 8 August 2017
- Published by Royal Society of Chemistry (RSC) in Journal of Materials Chemistry C
- Vol. 5 (34) , 8819-8827
Exploring the effects of interfacial carrier transport layers on device performance and optoelectronic properties of planar perovskite solar cells
The performance of perovskite device was found to be influenced by the interface quality and bulk defect activities induced in perovskite grown on HTL during device fabrication.
- 24 May 2017
- Published by Wiley in ChemSusChem
- Vol. 10 (15) , 3118-3134
Understanding the Limiting Factors of Solvent‐Annealed Small‐Molecule Bulk‐Heterojunction Organic Solar Cells from a Chemical Perspective
A detailed account of the limiting factors of solvent‐annealed bulk‐heterojunction small‐molecule organic solar cells is given. This account is based on the extensive characterisation of solar cell devices made from a library of five diketopyrolopyrole (DPP) donor dyes. Their chemical structure is designed in such a way as to provide insights into the energetics of solar cell active layer micro‐structure formation. Numerous chemical and physical properties of the active layers are assessed and inter‐related such as light absorption, molecular packing in the solid state, crystal‐forming properties in thin films, charge carrier mobility and charge carrier recombination kinetics. A myriad of characterisation techniques are used such as UV/Vis absorption spectroscopy, photoluminescence spectroscopy, XRD, AFM and photo‐induced transient measurements, which provide information on the optical properties of the active layers, morphology and recombination kinetics. Consequently, a mechanism for the solvent‐vapour‐annealing‐assisted formation of crystalline domains of donor molecules in the active layer is proposed, and the micro‐structural features are related to the J–V characteristics of the devices. According to this model, the crystalline phase in which the donor crystallise in the active layer is the key determinant to direct the formation of the micro‐structure.
- 24 May 2017
- Published by Wiley in ChemSusChem
- Vol. 10 (15) , 3118-3134
Understanding the Limiting Factors of Solvent‐Annealed Small‐Molecule Bulk‐Heterojunction Organic Solar Cells from a Chemical Perspective
A detailed account of the limiting factors of solvent‐annealed bulk‐heterojunction small‐molecule organic solar cells is given. This account is based on the extensive characterisation of solar cell devices made from a library of five diketopyrolopyrole (DPP) donor dyes. Their chemical structure is designed in such a way as to provide insights into the energetics of solar cell active layer micro‐structure formation. Numerous chemical and physical properties of the active layers are assessed and inter‐related such as light absorption, molecular packing in the solid state, crystal‐forming properties in thin films, charge carrier mobility and charge carrier recombination kinetics. A myriad of characterisation techniques are used such as UV/Vis absorption spectroscopy, photoluminescence spectroscopy, XRD, AFM and photo‐induced transient measurements, which provide information on the optical properties of the active layers, morphology and recombination kinetics. Consequently, a mechanism for the solvent‐vapour‐annealing‐assisted formation of crystalline domains of donor molecules in the active layer is proposed, and the micro‐structural features are related to the J–V characteristics of the devices. According to this model, the crystalline phase in which the donor crystallise in the active layer is the key determinant to direct the formation of the micro‐structure.