Docente
LNNano-CNPEM
Gabriel R. Schleder
Publicações
2026
Zanineli, Pedro H. M.; Focassio, Bruno; Schleder, Gabriel R.
2026.
@misc{zanineli2026crossgeometrytransferabilityassessmentuniversal,
title = {Cross-Geometry Transferability Assessment of Universal Machine Learning Interatomic Potentials: From Bulk Materials to Atomic Nanowires},
author = {Pedro H. M. Zanineli and Bruno Focassio and Gabriel R. Schleder},
url = {https://arxiv.org/abs/2608.06662},
year = {2026},
date = {2026-08-07},
urldate = {2026-01-01},
abstract = {Foundation machine-learning interatomic potentials (MLIPs) enable atomistic simulations at substantially lower computational cost than first-principles methods, but their reliability across structural geometries remains insufficiently understood. Here, we construct a density-functional-theory dataset of ZrO2 configurations spanning bulk, slab, particle, neck, and atomically thin wire environments motivated by an experimentally observed ZrO2 desintering process involving neck thinning and atomic wire formation. We first benchmark 26 pretrained MLIPs and observe pronounced geometry-dependent degradation in zero-shot predictions. Without any training, after only reference-energy alignment, the best zero-shot model (ORB-V3) reaches energy and force root-mean-square errors of 6 meV/atom and 197.3 meV/Å, respectively, with the largest force errors in neck and wire configurations. We then compare zero-shot inference, fine-tuning, and training from scratch strategies. Fine-tuning yields lower energy and force errors than training from scratch, while both require comparable wall-clock time. Geometry-specific fine-tuning improves in-domain accuracy but frequently produces negative transfer to other structural classes, whereas mixed-geometry fine-tuning reduces cross-geometry errors. Evaluations of elastic and vibrational properties, surface energies, and neck dynamics further show that rankings based on average energy and force errors do not universally predict property-level behavior. These results demonstrate that geometry-diverse target data and independent physical validations are necessary when adapting foundation MLIPs to low-coordination (ionic) nanostructures.},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Foundation machine-learning interatomic potentials (MLIPs) enable atomistic simulations at substantially lower computational cost than first-principles methods, but their reliability across structural geometries remains insufficiently understood. Here, we construct a density-functional-theory dataset of ZrO2 configurations spanning bulk, slab, particle, neck, and atomically thin wire environments motivated by an experimentally observed ZrO2 desintering process involving neck thinning and atomic wire formation. We first benchmark 26 pretrained MLIPs and observe pronounced geometry-dependent degradation in zero-shot predictions. Without any training, after only reference-energy alignment, the best zero-shot model (ORB-V3) reaches energy and force root-mean-square errors of 6 meV/atom and 197.3 meV/Å, respectively, with the largest force errors in neck and wire configurations. We then compare zero-shot inference, fine-tuning, and training from scratch strategies. Fine-tuning yields lower energy and force errors than training from scratch, while both require comparable wall-clock time. Geometry-specific fine-tuning improves in-domain accuracy but frequently produces negative transfer to other structural classes, whereas mixed-geometry fine-tuning reduces cross-geometry errors. Evaluations of elastic and vibrational properties, surface energies, and neck dynamics further show that rankings based on average energy and force errors do not universally predict property-level behavior. These results demonstrate that geometry-diverse target data and independent physical validations are necessary when adapting foundation MLIPs to low-coordination (ionic) nanostructures.
Lopes, Luiz E. F. C.; Zanineli, Pedro H. M.; Focassio, Bruno; Schleder, Gabriel R.
Beyond Retrieval: Compounding Scientific Extelligence with Artificial Intelligence Wikis Não publicado
2026, (working paper or preprint).
@unpublished{efclopes:hal-05635223,
title = {Beyond Retrieval: Compounding Scientific Extelligence with Artificial Intelligence Wikis},
author = {Luiz E. F. C. Lopes and Pedro H. M. Zanineli and Bruno Focassio and Gabriel R. Schleder},
url = {https://hal.science/hal-05635223},
year = {2026},
date = {2026-05-27},
urldate = {2026-05-01},
abstract = {Information overload severely bottlenecks scientific synthesis and continuity. We argue that AI-assisted science requires more than retrieval: persistent memory systems that accumulate structured knowledge across interactions must become a foundational layer of future research infrastructure.},
note = {working paper or preprint},
keywords = {},
pubstate = {published},
tppubtype = {unpublished}
}
Information overload severely bottlenecks scientific synthesis and continuity. We argue that AI-assisted science requires more than retrieval: persistent memory systems that accumulate structured knowledge across interactions must become a foundational layer of future research infrastructure.
Zanineli, P.; Lopes, E. V. C.; Schleder, G. R.; Lemos, L. N.; Lima, F. Crasto; Fazzio, A.
Heterogeneous Molecular Signatures of Human Odor Perception Miscellaneous
2026.
@misc{zanineli2026heterogeneousmolecularsignatureshuman,
title = {Heterogeneous Molecular Signatures of Human Odor Perception},
author = {P. Zanineli and E. V. C. Lopes and G. R. Schleder and L. N. Lemos and F. Crasto Lima and A. Fazzio},
url = {https://arxiv.org/abs/2604.09758},
year = {2026},
date = {2026-04-10},
urldate = {2026-01-01},
abstract = {Understanding how molecular structure gives rise to odor perception remains a long-standing challenge, with ongoing debate over whether olfaction is primarily governed by molecular shape, vibrational properties, or their interplay at the level of olfactory receptors. Here, we ask whether different odors rely on common molecular determinants or instead emerge from distinct physicochemical regimes. Using interpretable machine-learning models trained on molecular descriptors derived from first-principles calculations that span electronic, vibrational, and structural properties, we analyze feature contributions for odor categories and their associated receptors. We find that no single descriptor class universally dominates odor prediction; instead, different odors exhibit strongly odor-specific patterns of feature importance, with substantial variability across physicochemical domains. This heterogeneity is consistent across different models, suggesting that a universal encoding scheme does not capture odor perception but reflects receptor- and odor-dependent structure-odor relationships. Our results provide statistical constraints on competing olfactory theories and offer a data-driven framework for organizing odor space.},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Understanding how molecular structure gives rise to odor perception remains a long-standing challenge, with ongoing debate over whether olfaction is primarily governed by molecular shape, vibrational properties, or their interplay at the level of olfactory receptors. Here, we ask whether different odors rely on common molecular determinants or instead emerge from distinct physicochemical regimes. Using interpretable machine-learning models trained on molecular descriptors derived from first-principles calculations that span electronic, vibrational, and structural properties, we analyze feature contributions for odor categories and their associated receptors. We find that no single descriptor class universally dominates odor prediction; instead, different odors exhibit strongly odor-specific patterns of feature importance, with substantial variability across physicochemical domains. This heterogeneity is consistent across different models, suggesting that a universal encoding scheme does not capture odor perception but reflects receptor- and odor-dependent structure-odor relationships. Our results provide statistical constraints on competing olfactory theories and offer a data-driven framework for organizing odor space.
Zanineli, P.; Lopes, E. V. C.; Schleder, G. R.; Lemos, L. N.; Lima, F. Crasto; Fazzio, A.
Heterogeneous Molecular Signatures of Human Odor Perception Miscellaneous
2026.
@misc{zanineli2026heterogeneousmolecularsignatureshumanb,
title = {Heterogeneous Molecular Signatures of Human Odor Perception},
author = {P. Zanineli and E. V. C. Lopes and G. R. Schleder and L. N. Lemos and F. Crasto Lima and A. Fazzio},
url = {https://arxiv.org/abs/2604.09758},
year = {2026},
date = {2026-04-10},
urldate = {2026-01-01},
abstract = {Understanding how molecular structure gives rise to odor perception remains a long-standing challenge, with ongoing debate over whether olfaction is primarily governed by molecular shape, vibrational properties, or their interplay at the level of olfactory receptors. Here, we ask whether different odors rely on common molecular determinants or instead emerge from distinct physicochemical regimes. Using interpretable machine-learning models trained on molecular descriptors derived from first-principles calculations that span electronic, vibrational, and structural properties, we analyze feature contributions for odor categories and their associated receptors. We find that no single descriptor class universally dominates odor prediction; instead, different odors exhibit strongly odor-specific patterns of feature importance, with substantial variability across physicochemical domains. This heterogeneity is consistent across different models, suggesting that a universal encoding scheme does not capture odor perception but reflects receptor- and odor-dependent structure-odor relationships. Our results provide statistical constraints on competing olfactory theories and offer a data-driven framework for organizing odor space.},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Understanding how molecular structure gives rise to odor perception remains a long-standing challenge, with ongoing debate over whether olfaction is primarily governed by molecular shape, vibrational properties, or their interplay at the level of olfactory receptors. Here, we ask whether different odors rely on common molecular determinants or instead emerge from distinct physicochemical regimes. Using interpretable machine-learning models trained on molecular descriptors derived from first-principles calculations that span electronic, vibrational, and structural properties, we analyze feature contributions for odor categories and their associated receptors. We find that no single descriptor class universally dominates odor prediction; instead, different odors exhibit strongly odor-specific patterns of feature importance, with substantial variability across physicochemical domains. This heterogeneity is consistent across different models, suggesting that a universal encoding scheme does not capture odor perception but reflects receptor- and odor-dependent structure-odor relationships. Our results provide statistical constraints on competing olfactory theories and offer a data-driven framework for organizing odor space.
Deleigo, Ana Vitória Ferreira; Lelis, Gabrielle Coelho; Braunger, Maria Luisa; Casalini, Stefano; Watanabe, Yasmin; Schleder, Gabriel Ravanhani; Fonseca, Wilson Tiago; Oliveira, Rafael Furlan
Ultrasensitive MicroRNA Detection Combining Reduced Graphene Oxide Electrolyte‐Gated Transistors and Machine Learning Journal Article
Em: Small, 2026, ISSN: 1613-6829.
@article{Deleigo2026,
title = {Ultrasensitive MicroRNA Detection Combining Reduced Graphene Oxide Electrolyte‐Gated Transistors and Machine Learning},
author = {Ana Vitória Ferreira Deleigo and Gabrielle Coelho Lelis and Maria Luisa Braunger and Stefano Casalini and Yasmin Watanabe and Gabriel Ravanhani Schleder and Wilson Tiago Fonseca and Rafael Furlan Oliveira},
doi = {10.1002/smll.202512199},
issn = {1613-6829},
year = {2026},
date = {2026-01-28},
urldate = {2026-01-28},
journal = {Small},
publisher = {Wiley},
abstract = {<jats:title>ABSTRACT</jats:title>
<jats:p>
MicroRNAs (miRNAs) are promising biomarkers for disease diagnosis, but conventional detection methods such as reverse transcription polymerase chain reaction (RT‐PCR) require complex instrumentation and reagents, limiting their suitability for portable diagnostics. Here, we report an ultrasensitive and selective biosensor that integrates DNA‐functionalized reduced graphene oxide (rGO), electrolyte‐gated transistors (EGTs), and machine learning (ML) for miRNA detection. The platform targets the miR‐34 family (miR‐34a, miR‐34b, and miR‐34c), which is associated with cancer and neurological disorders. The biosensor discriminates perfectly matched from mismatched sequences over a wide dynamic range (0.1–1000 amol L
<jats:sup>−1</jats:sup>
) with an ultralow limit of detection of 0.098 amol L
<jats:sup>−1</jats:sup>
. ML enables multidimensional analysis of EGT transfer curves and extraction of physically meaningful features from high‐dimensional data. This approach advances point‐of‐care technologies for highly sensitive and selective miRNA detection, with strong potential for portable molecular diagnostics.
</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:title>ABSTRACT</jats:title>
<jats:p>
MicroRNAs (miRNAs) are promising biomarkers for disease diagnosis, but conventional detection methods such as reverse transcription polymerase chain reaction (RT‐PCR) require complex instrumentation and reagents, limiting their suitability for portable diagnostics. Here, we report an ultrasensitive and selective biosensor that integrates DNA‐functionalized reduced graphene oxide (rGO), electrolyte‐gated transistors (EGTs), and machine learning (ML) for miRNA detection. The platform targets the miR‐34 family (miR‐34a, miR‐34b, and miR‐34c), which is associated with cancer and neurological disorders. The biosensor discriminates perfectly matched from mismatched sequences over a wide dynamic range (0.1–1000 amol L
<jats:sup>−1</jats:sup>
) with an ultralow limit of detection of 0.098 amol L
<jats:sup>−1</jats:sup>
. ML enables multidimensional analysis of EGT transfer curves and extraction of physically meaningful features from high‐dimensional data. This approach advances point‐of‐care technologies for highly sensitive and selective miRNA detection, with strong potential for portable molecular diagnostics.
</jats:p>
<jats:p>
MicroRNAs (miRNAs) are promising biomarkers for disease diagnosis, but conventional detection methods such as reverse transcription polymerase chain reaction (RT‐PCR) require complex instrumentation and reagents, limiting their suitability for portable diagnostics. Here, we report an ultrasensitive and selective biosensor that integrates DNA‐functionalized reduced graphene oxide (rGO), electrolyte‐gated transistors (EGTs), and machine learning (ML) for miRNA detection. The platform targets the miR‐34 family (miR‐34a, miR‐34b, and miR‐34c), which is associated with cancer and neurological disorders. The biosensor discriminates perfectly matched from mismatched sequences over a wide dynamic range (0.1–1000 amol L
<jats:sup>−1</jats:sup>
) with an ultralow limit of detection of 0.098 amol L
<jats:sup>−1</jats:sup>
. ML enables multidimensional analysis of EGT transfer curves and extraction of physically meaningful features from high‐dimensional data. This approach advances point‐of‐care technologies for highly sensitive and selective miRNA detection, with strong potential for portable molecular diagnostics.
</jats:p>
Pereira, Gabriel X.; Leite, Marcos C.; Zanineli, Pedro H. M.; Costa, Juliana N. Y.; Cunha, Débora M.; Serra, Sophia P.; Sophia, Pedro H.; Shimizu, Flávio M.; Moura, Yasmin W.; Oliveira, Célio C.; Schleder, Gabriel R.; Persinoti, Gabriela F.; Merces, Leandro; Lima, Renato S.
Accelerating Biosensor Discovery: A Computationally‐Driven Pipeline for Microplastics Monitoring Journal Article
Em: adv. intell. discov., 2026, ISSN: 2943-9981.
@article{Pereira2026,
title = {Accelerating Biosensor Discovery: A Computationally‐Driven Pipeline for Microplastics Monitoring},
author = {Gabriel X. Pereira and Marcos C. Leite and Pedro H. M. Zanineli and Juliana N. Y. Costa and Débora M. Cunha and Sophia P. Serra and Pedro H. Sophia and Flávio M. Shimizu and Yasmin W. Moura and Célio C. Oliveira and Gabriel R. Schleder and Gabriela F. Persinoti and Leandro Merces and Renato S. Lima},
doi = {10.1002/aidi.202500156},
issn = {2943-9981},
year = {2026},
date = {2026-01-22},
urldate = {2026-01-22},
journal = {adv. intell. discov.},
publisher = {Wiley},
abstract = {<jats:p>
The intelligent discovery of novel biosensors is often hampered by slow, trial‐and‐error experimental cycles. To overcome this gap, we introduce and validate a computationally‐guided discovery pipeline that synergizes molecular simulation, synthetic biology, electrochemical engineering, and machine learning guided analysis for the rational design of high‐performance sensors. We demonstrate the power of this approach by tackling the urgent challenge of detecting micro‐ and nanoplastics (MNPs), which are potential harmful agents, causing respiratory, cardiovascular, and oncological disorders. Our pipeline begins with computational screening, using molecular dynamics simulations to evaluate the thermal stability and binding affinity of a candidate protein for recognition element, the carbohydrate‐binding module of
<jats:italic>Bacillus anthracis</jats:italic>
(
<jats:italic>Ba</jats:italic>
CBM2). Building upon the positive computational results, the protein was synthesized and integrated onto a custom‐fabricated electrochemical platform. The computationally‐informed protein was experimentally verified to have a superior analytical performance when covalently tethered to gold surfaces of on‐chip electrodes, which acted as label‐free electrochemical biosensors. To date, this system implied reductions in the root mean square error in MNP quantification and measurement standard deviation by 92.24% and 24.83%, respectively, compared to unmodified devices, and its integration with multivariate analysis (Sure Independence Screening and Sparsifying Operator) overcame device‐to‐device variability. This work not only delivers a promising biosensor for MNP monitoring, but, more importantly, establishes a validated workflow for the intelligent discovery of functional and specific proteins. This pipeline provides a direct pathway for future integration with high‐throughput virtual screening and machine learning models, enabling the inverse design of next‐generation environmental and diagnostic sensor technologies.
</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:p>
The intelligent discovery of novel biosensors is often hampered by slow, trial‐and‐error experimental cycles. To overcome this gap, we introduce and validate a computationally‐guided discovery pipeline that synergizes molecular simulation, synthetic biology, electrochemical engineering, and machine learning guided analysis for the rational design of high‐performance sensors. We demonstrate the power of this approach by tackling the urgent challenge of detecting micro‐ and nanoplastics (MNPs), which are potential harmful agents, causing respiratory, cardiovascular, and oncological disorders. Our pipeline begins with computational screening, using molecular dynamics simulations to evaluate the thermal stability and binding affinity of a candidate protein for recognition element, the carbohydrate‐binding module of
<jats:italic>Bacillus anthracis</jats:italic>
(
<jats:italic>Ba</jats:italic>
CBM2). Building upon the positive computational results, the protein was synthesized and integrated onto a custom‐fabricated electrochemical platform. The computationally‐informed protein was experimentally verified to have a superior analytical performance when covalently tethered to gold surfaces of on‐chip electrodes, which acted as label‐free electrochemical biosensors. To date, this system implied reductions in the root mean square error in MNP quantification and measurement standard deviation by 92.24% and 24.83%, respectively, compared to unmodified devices, and its integration with multivariate analysis (Sure Independence Screening and Sparsifying Operator) overcame device‐to‐device variability. This work not only delivers a promising biosensor for MNP monitoring, but, more importantly, establishes a validated workflow for the intelligent discovery of functional and specific proteins. This pipeline provides a direct pathway for future integration with high‐throughput virtual screening and machine learning models, enabling the inverse design of next‐generation environmental and diagnostic sensor technologies.
</jats:p>
The intelligent discovery of novel biosensors is often hampered by slow, trial‐and‐error experimental cycles. To overcome this gap, we introduce and validate a computationally‐guided discovery pipeline that synergizes molecular simulation, synthetic biology, electrochemical engineering, and machine learning guided analysis for the rational design of high‐performance sensors. We demonstrate the power of this approach by tackling the urgent challenge of detecting micro‐ and nanoplastics (MNPs), which are potential harmful agents, causing respiratory, cardiovascular, and oncological disorders. Our pipeline begins with computational screening, using molecular dynamics simulations to evaluate the thermal stability and binding affinity of a candidate protein for recognition element, the carbohydrate‐binding module of
<jats:italic>Bacillus anthracis</jats:italic>
(
<jats:italic>Ba</jats:italic>
CBM2). Building upon the positive computational results, the protein was synthesized and integrated onto a custom‐fabricated electrochemical platform. The computationally‐informed protein was experimentally verified to have a superior analytical performance when covalently tethered to gold surfaces of on‐chip electrodes, which acted as label‐free electrochemical biosensors. To date, this system implied reductions in the root mean square error in MNP quantification and measurement standard deviation by 92.24% and 24.83%, respectively, compared to unmodified devices, and its integration with multivariate analysis (Sure Independence Screening and Sparsifying Operator) overcame device‐to‐device variability. This work not only delivers a promising biosensor for MNP monitoring, but, more importantly, establishes a validated workflow for the intelligent discovery of functional and specific proteins. This pipeline provides a direct pathway for future integration with high‐throughput virtual screening and machine learning models, enabling the inverse design of next‐generation environmental and diagnostic sensor technologies.
</jats:p>
2025
Zanineli, Pedro H. M.; Monteiro, Matheus Zaia; Wasques, Vinicius Francisco; Simões, Francielle Santo Pedro; Schleder, Gabriel R.
Fuzzy Neural Network Performance and Interpretability of Quantum Wavefunction Probability Predictions Miscellaneous
2025.
@misc{zanineli2025fuzzyneuralnetworkperformance,
title = {Fuzzy Neural Network Performance and Interpretability of Quantum Wavefunction Probability Predictions},
author = {Pedro H. M. Zanineli and Matheus Zaia Monteiro and Vinicius Francisco Wasques and Francielle Santo Pedro Simões and Gabriel R. Schleder},
url = {https://arxiv.org/abs/2511.05261},
year = {2025},
date = {2025-11-07},
urldate = {2025-01-01},
abstract = {Predicting quantum wavefunction probability distributions is crucial for computational chemistry and materials science, yet machine learning (ML) models often face a trade-off between accuracy and interpretability. This study compares Artificial Neural Networks (ANNs) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) in modeling quantum probability distributions for the H ion, leveraging data generated via Physics-Informed Neural Networks (PINNs). While ANN achieved superior accuracy (R = 0.99 vs ANFIS's 0.95 with Gaussian membership functions), it required over 50x more parameters (2,305 vs 39-45). ANFIS, however, provided unique interpretability: its Gaussian membership functions encoded spatial electron localization near proton positions (), mirroring Born probability densities, while fuzzy rules reflected quantum superposition principles. Rules prioritizing the internuclear direction revealed the system's 1D symmetry, aligning with Linear Combination of Atomic Orbitals theory–a novel data-driven perspective on orbital hybridization. Membership function variances () further quantified electron delocalization trends, and peak prediction errors highlighted unresolved quantum cusps. The choice of functions critically impacted performance: Gaussian/Generalized Bell outperformed Sigmoid, with errors improving as training data increased, showing scalability. This study underscores the context-dependent value of ML: ANN for precision and ANFIS for interpretable, parameter-efficient approximations that link inputs to physical behavior. These findings advocate hybrid approaches in quantum simulations, balancing accuracy with explainability to accelerate discovery. Future work should extend ANFIS to multi-electron systems and integrate domain-specific constraints (e.g., kinetic energy terms), bridging data-driven models and fundamental physics.},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Predicting quantum wavefunction probability distributions is crucial for computational chemistry and materials science, yet machine learning (ML) models often face a trade-off between accuracy and interpretability. This study compares Artificial Neural Networks (ANNs) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) in modeling quantum probability distributions for the H ion, leveraging data generated via Physics-Informed Neural Networks (PINNs). While ANN achieved superior accuracy (R = 0.99 vs ANFIS's 0.95 with Gaussian membership functions), it required over 50x more parameters (2,305 vs 39-45). ANFIS, however, provided unique interpretability: its Gaussian membership functions encoded spatial electron localization near proton positions (), mirroring Born probability densities, while fuzzy rules reflected quantum superposition principles. Rules prioritizing the internuclear direction revealed the system's 1D symmetry, aligning with Linear Combination of Atomic Orbitals theory–a novel data-driven perspective on orbital hybridization. Membership function variances () further quantified electron delocalization trends, and peak prediction errors highlighted unresolved quantum cusps. The choice of functions critically impacted performance: Gaussian/Generalized Bell outperformed Sigmoid, with errors improving as training data increased, showing scalability. This study underscores the context-dependent value of ML: ANN for precision and ANFIS for interpretable, parameter-efficient approximations that link inputs to physical behavior. These findings advocate hybrid approaches in quantum simulations, balancing accuracy with explainability to accelerate discovery. Future work should extend ANFIS to multi-electron systems and integrate domain-specific constraints (e.g., kinetic energy terms), bridging data-driven models and fundamental physics.
Brito, Ana Carolina Ferreira; Pinto, Alysson Alves; Plutnar, Jan; Sofer, Zdenek; Schleder, Gabriel Ravanhani; Capaz, Rodrigo B.; Barcelos, Ingrid David; Neves, Bernardo Ruegger Almeida
Unveiling Composition–Properties Relationships in Mo1−xWxSe2 Alloys: A Theoretical and Experimental Study Journal Article
Em: Nanotechnology, 2025, ISSN: 1361-6528.
@article{Brito2025,
title = {Unveiling Composition–Properties Relationships in Mo1−xWxSe2 Alloys: A Theoretical and Experimental Study},
author = {Ana Carolina Ferreira Brito and Alysson Alves Pinto and Jan Plutnar and Zdenek Sofer and Gabriel Ravanhani Schleder and Rodrigo B. Capaz and Ingrid David Barcelos and Bernardo Ruegger Almeida Neves},
doi = {10.1088/1361-6528/ae00cd},
issn = {1361-6528},
year = {2025},
date = {2025-08-29},
urldate = {2025-08-29},
journal = {Nanotechnology},
publisher = {IOP Publishing},
abstract = {Two-dimensional transition metal dichalcogenide (TMD) alloys have emerged as a versatile platform for electronic, optoelectronic, and quantum applications due to their tunable crystal structure and unique electronic properties. In this study, we investigate the influence of atomic composition on the structural, electronic, and optical properties of the Mo1−xWxSe2 alloy, combining experimental and theoretical approaches. Samples with different Mo and W ratios were synthesized and characterized using Raman and photoluminescence (PL) spectroscopies, and atomic force microscopy (AFM). Local anodic oxidation (LAO) was employed to manipulate monolayers within the alloy flakes, revealing significant luminescence enhancement in the engineered islands,
suggesting structural and electronic modifications. Additionally, density functional theory (DFT) calculations indicated that oxidation stability strongly depends on atomic composition, with the Mo0.5W0.5Se2 and Mo0.75W0.25Se2 alloys exhibiting the highest
resistance to vacancy formation. These findings highlight the potential for structural and electronic engineering of Mo1−xWxSe2 alloys, paving the way for advanced applications in nanotechnology and quantum computing.</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Two-dimensional transition metal dichalcogenide (TMD) alloys have emerged as a versatile platform for electronic, optoelectronic, and quantum applications due to their tunable crystal structure and unique electronic properties. In this study, we investigate the influence of atomic composition on the structural, electronic, and optical properties of the Mo1−xWxSe2 alloy, combining experimental and theoretical approaches. Samples with different Mo and W ratios were synthesized and characterized using Raman and photoluminescence (PL) spectroscopies, and atomic force microscopy (AFM). Local anodic oxidation (LAO) was employed to manipulate monolayers within the alloy flakes, revealing significant luminescence enhancement in the engineered islands,
suggesting structural and electronic modifications. Additionally, density functional theory (DFT) calculations indicated that oxidation stability strongly depends on atomic composition, with the Mo0.5W0.5Se2 and Mo0.75W0.25Se2 alloys exhibiting the highest
resistance to vacancy formation. These findings highlight the potential for structural and electronic engineering of Mo1−xWxSe2 alloys, paving the way for advanced applications in nanotechnology and quantum computing.</jats:p>
2024
Focassio, Bruno; Freitas, Luis Paulo M.; Schleder, Gabriel R.
Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials’ Surfaces Journal Article
Em: ACS Applied Materials & Interfaces, vol. 0, não 0, pp. null, 2024, (PMID: 38990833).
@article{doi:10.1021/acsami.4c03815,
title = {Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials’ Surfaces},
author = {Bruno Focassio and Luis Paulo M. Freitas and Gabriel R. Schleder},
url = {https://doi.org/10.1021/acsami.4c03815},
doi = {10.1021/acsami.4c03815},
year = {2024},
date = {2024-07-11},
journal = {ACS Applied Materials & Interfaces},
volume = {0},
number = {0},
pages = {null},
abstract = {Machine learning interatomic potentials (MLIPs) are one of the main techniques in the materials science toolbox, able to bridge ab initio accuracy with the computational efficiency of classical force fields. This allows simulations ranging from atoms, molecules, and biosystems, to solid and bulk materials, surfaces, nanomaterials, and their interfaces and complex interactions. A recent class of advanced MLIPs, which use equivariant representations and deep graph neural networks, is known as universal models. These models are proposed as foundation models suitable for any system, covering most elements from the periodic table. Current universal MLIPs (UIPs) have been trained with the largest consistent data set available nowadays. However, these are composed mostly of bulk materials’ DFT calculations. In this article, we assess the universality of all openly available UIPs, namely MACE, CHGNet, and M3GNet, in a representative task of generalization: calculation of surface energies. We find that the out-of-the-box foundation models have significant shortcomings in this task, with errors correlated to the total energy of surface simulations, having an out-of-domain distance from the training data set. Our results show that while UIPs are an efficient starting point for fine-tuning specialized models, we envision the potential of increasing the coverage of the materials space toward universal training data sets for MLIPs.},
note = {PMID: 38990833},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Machine learning interatomic potentials (MLIPs) are one of the main techniques in the materials science toolbox, able to bridge ab initio accuracy with the computational efficiency of classical force fields. This allows simulations ranging from atoms, molecules, and biosystems, to solid and bulk materials, surfaces, nanomaterials, and their interfaces and complex interactions. A recent class of advanced MLIPs, which use equivariant representations and deep graph neural networks, is known as universal models. These models are proposed as foundation models suitable for any system, covering most elements from the periodic table. Current universal MLIPs (UIPs) have been trained with the largest consistent data set available nowadays. However, these are composed mostly of bulk materials’ DFT calculations. In this article, we assess the universality of all openly available UIPs, namely MACE, CHGNet, and M3GNet, in a representative task of generalization: calculation of surface energies. We find that the out-of-the-box foundation models have significant shortcomings in this task, with errors correlated to the total energy of surface simulations, having an out-of-domain distance from the training data set. Our results show that while UIPs are an efficient starting point for fine-tuning specialized models, we envision the potential of increasing the coverage of the materials space toward universal training data sets for MLIPs.
Fiuza, Tanna E. R.; Focassio, Bruno; Bettini, Jefferson; Schleder, Gabriel R.; Rodrigues, Murillo H. M.; Junior, João B. Souza; Fazzio, Adalberto; Capaz, Rodrigo B.; Leite, Edson R.
Visualization of electron beam-induced desintering of nanostructured ceramics at the atomic scale Journal Article
Em: Cell Reports Physical Science, pp. 101828, 2024, ISSN: 2666-3864.
@article{FIUZA2024101828,
title = {Visualization of electron beam-induced desintering of nanostructured ceramics at the atomic scale},
author = {Tanna E. R. Fiuza and Bruno Focassio and Jefferson Bettini and Gabriel R. Schleder and Murillo H. M. Rodrigues and João B. Souza Junior and Adalberto Fazzio and Rodrigo B. Capaz and Edson R. Leite},
url = {https://www.sciencedirect.com/science/article/pii/S2666386424000535},
doi = {https://doi.org/10.1016/j.xcrp.2024.101828},
issn = {2666-3864},
year = {2024},
date = {2024-02-12},
urldate = {2024-01-01},
journal = {Cell Reports Physical Science},
pages = {101828},
abstract = {Summary
Mass diffusion and local tensile stress associated with electron beam irradiation can favor the desintering process. Here, we report the electron beam-induced desintering of ZrO2 thin films at the atomic scale with unprecedented spatial resolution using high-resolution transmission electron microscopy (HRTEM). Our results confirm earlier works in which desintering is driven by tensile stress acting on the bridge if an external stimulus, such as irradiation, triggers atom mobility. Additionally, we find departures from classical microscopic descriptions: a very stable nanobridge is formed and evolves until rupture with a constant dihedral angle instead of a brittle rupture. An adapted model for desintering at the nanoscale is proposed using the experimental findings. This work provides insights that may improve the knowledge of the rupture of ceramic materials at the nano and atomic scales, contributing to a better knowledge of materials’ behavior.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Summary
Mass diffusion and local tensile stress associated with electron beam irradiation can favor the desintering process. Here, we report the electron beam-induced desintering of ZrO2 thin films at the atomic scale with unprecedented spatial resolution using high-resolution transmission electron microscopy (HRTEM). Our results confirm earlier works in which desintering is driven by tensile stress acting on the bridge if an external stimulus, such as irradiation, triggers atom mobility. Additionally, we find departures from classical microscopic descriptions: a very stable nanobridge is formed and evolves until rupture with a constant dihedral angle instead of a brittle rupture. An adapted model for desintering at the nanoscale is proposed using the experimental findings. This work provides insights that may improve the knowledge of the rupture of ceramic materials at the nano and atomic scales, contributing to a better knowledge of materials’ behavior.
Mass diffusion and local tensile stress associated with electron beam irradiation can favor the desintering process. Here, we report the electron beam-induced desintering of ZrO2 thin films at the atomic scale with unprecedented spatial resolution using high-resolution transmission electron microscopy (HRTEM). Our results confirm earlier works in which desintering is driven by tensile stress acting on the bridge if an external stimulus, such as irradiation, triggers atom mobility. Additionally, we find departures from classical microscopic descriptions: a very stable nanobridge is formed and evolves until rupture with a constant dihedral angle instead of a brittle rupture. An adapted model for desintering at the nanoscale is proposed using the experimental findings. This work provides insights that may improve the knowledge of the rupture of ceramic materials at the nano and atomic scales, contributing to a better knowledge of materials’ behavior.
Focassio, Bruno; Schleder, Gabriel R.; Fazzio, Adalberto; Capaz, Rodrigo B.; Lopes, Pedro V.; Ferreira, Jaime; Enderlein, Carsten; Neto, Marcello B. Silva
Magnetic control of Weyl nodes and wave packets in three-dimensional warped semimetals Working paper
2024.
@workingpaper{focassio2024magnetic,
title = {Magnetic control of Weyl nodes and wave packets in three-dimensional warped semimetals},
author = {Bruno Focassio and Gabriel R. Schleder and Adalberto Fazzio and Rodrigo B. Capaz and Pedro V. Lopes and Jaime Ferreira and Carsten Enderlein and Marcello B. Silva Neto},
url = {https://arxiv.org/abs/2401.06282},
doi = {https://doi.org/10.48550/arXiv.2401.06282},
year = {2024},
date = {2024-01-11},
urldate = {2024-01-01},
abstract = {We investigate the topological phase transitions driven by band warping and a transverse magnetic field, for three-dimensional Weyl semimetals. First, we use the Chern number as a mathematical tool to derive the topological phase diagram. Next, we associate each of the topological sectors to a given angular momentum state of a rotating wave packet. Then we show how the position of the Weyl nodes can be manipulated by a transverse external magnetic field that ultimately quenches the wave packet rotation, first partially and then completely, thus resulting in a sequence of field-induced topological phase transitions. Finally, we calculate the current-induced magnetization and the anomalous Hall conductivity of a prototypical warped Weyl material. Both observables reflect the topological transitions associated with the wave packet rotation and can help to identify the elusive 3D quantum anomalous Hall effect in three-dimensional, warped Weyl materials.},
keywords = {},
pubstate = {published},
tppubtype = {workingpaper}
}
We investigate the topological phase transitions driven by band warping and a transverse magnetic field, for three-dimensional Weyl semimetals. First, we use the Chern number as a mathematical tool to derive the topological phase diagram. Next, we associate each of the topological sectors to a given angular momentum state of a rotating wave packet. Then we show how the position of the Weyl nodes can be manipulated by a transverse external magnetic field that ultimately quenches the wave packet rotation, first partially and then completely, thus resulting in a sequence of field-induced topological phase transitions. Finally, we calculate the current-induced magnetization and the anomalous Hall conductivity of a prototypical warped Weyl material. Both observables reflect the topological transitions associated with the wave packet rotation and can help to identify the elusive 3D quantum anomalous Hall effect in three-dimensional, warped Weyl materials.
Focassio, Bruno; Freitas, Luis Paulo Mezzina; Schleder, Gabriel R.
Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials' Surfaces Working paper
2024.
@workingpaper{focassio2024performance,
title = {Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials' Surfaces},
author = {Bruno Focassio and Luis Paulo Mezzina Freitas and Gabriel R. Schleder},
url = {https://arxiv.org/abs/2403.04217},
doi = {10.48550/arXiv.2403.04217},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
abstract = {Machine learning interatomic potentials (MLIPs) are one of the main techniques in the materials science toolbox, able to bridge ab initio accuracy with the computational efficiency of classical force fields. This allows simulations ranging from atoms, molecules, and biosystems, to solid and bulk materials, surfaces, nanomaterials, and their interfaces and complex interactions. A recent class of advanced MLIPs, which use equivariant representations and deep graph neural networks, is known as universal models. These models are proposed as foundational models suitable for any system, covering most elements from the periodic table. Current universal MLIPs (UIPs) have been trained with the largest consistent dataset available nowadays. However, these are composed mostly of bulk materials' DFT calculations. In this article, we assess the universality of all openly available UIPs, namely MACE, CHGNet, and M3GNet, in a representative task of generalization: calculation of surface energies. We find that the out-of-the-box foundational models have significant shortcomings in this task, with errors correlated to the total energy of surface simulations, having an out-of-domain distance from the training dataset. Our results show that while UIPs are an efficient starting point for fine-tuning specialized models, we envision the potential of increasing the coverage of the materials space towards universal training datasets for MLIPs.},
keywords = {},
pubstate = {published},
tppubtype = {workingpaper}
}
Machine learning interatomic potentials (MLIPs) are one of the main techniques in the materials science toolbox, able to bridge ab initio accuracy with the computational efficiency of classical force fields. This allows simulations ranging from atoms, molecules, and biosystems, to solid and bulk materials, surfaces, nanomaterials, and their interfaces and complex interactions. A recent class of advanced MLIPs, which use equivariant representations and deep graph neural networks, is known as universal models. These models are proposed as foundational models suitable for any system, covering most elements from the periodic table. Current universal MLIPs (UIPs) have been trained with the largest consistent dataset available nowadays. However, these are composed mostly of bulk materials' DFT calculations. In this article, we assess the universality of all openly available UIPs, namely MACE, CHGNet, and M3GNet, in a representative task of generalization: calculation of surface energies. We find that the out-of-the-box foundational models have significant shortcomings in this task, with errors correlated to the total energy of surface simulations, having an out-of-domain distance from the training dataset. Our results show that while UIPs are an efficient starting point for fine-tuning specialized models, we envision the potential of increasing the coverage of the materials space towards universal training datasets for MLIPs.
2023
Lelis, Gabrielle Coelho; Fonseca, Wilson Tiago; Lima, Alessandro Henrique; Okazaki, Anderson Kenji; Figueiredo, Eduardo Costa; Jr, Antonio Riul; Schleder, Gabriel R.; Samorì, Paolo; Oliveira, Rafael Furlan
Em: ACS Applied Materials & Interfaces, vol. 0, não 0, pp. null, 2023, (PMID: 38134415).
@article{doi:10.1021/acsami.3c16699,
title = {Harnessing Small-Molecule Analyte Detection in Complex Media: Combining Molecularly Imprinted Polymers, Electrolytic Transistors, and Machine Learning},
author = {Gabrielle Coelho Lelis and Wilson Tiago Fonseca and Alessandro Henrique Lima and Anderson Kenji Okazaki and Eduardo Costa Figueiredo and Antonio Riul Jr and Gabriel R. Schleder and Paolo Samorì and Rafael Furlan Oliveira},
url = {https://doi.org/10.1021/acsami.3c16699},
doi = {10.1021/acsami.3c16699},
year = {2023},
date = {2023-12-22},
journal = {ACS Applied Materials & Interfaces},
volume = {0},
number = {0},
pages = {null},
abstract = {Small-molecule analyte detection is key for improving quality of life, particularly in health monitoring through the early detection of diseases. However, detecting specific markers in complex multicomponent media using devices compatible with point-of-care (PoC) technologies is still a major challenge. Here, we introduce a novel approach that combines molecularly imprinted polymers (MIPs), electrolyte-gated transistors (EGTs) based on 2D materials, and machine learning (ML) to detect hippuric acid (HA) in artificial urine, being a critical marker for toluene intoxication, parasitic infections, and kidney and bowel inflammation. Reduced graphene oxide (rGO) was used as the sensory material and molecularly imprinted polymer (MIP) as supramolecular receptors. Employing supervised ML techniques based on symbolic regression and compressive sensing enabled us to comprehensively analyze the EGT transfer curves, eliminating the need for arbitrary signal selection and allowing a multivariate analysis during HA detection. The resulting device displayed simultaneously low operating voltages (<0.5 V), rapid response times (≤10 s), operation across a wide range of HA concentrations (from 0.05 to 200 nmol L–1), and a low limit of detection (LoD) of 39 pmol L–1. Thanks to the ML multivariate analysis, we achieved a 2.5-fold increase in the device sensitivity (1.007 μA/nmol L–1) with respect to the human data analysis (0.388 μA/nmol L–1). Our method represents a major advance in PoC technologies, by enabling the accurate determination of small-molecule markers in complex media via the combination of ML analysis, supramolecular analyte recognition, and electrolytic transistors.},
note = {PMID: 38134415},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Small-molecule analyte detection is key for improving quality of life, particularly in health monitoring through the early detection of diseases. However, detecting specific markers in complex multicomponent media using devices compatible with point-of-care (PoC) technologies is still a major challenge. Here, we introduce a novel approach that combines molecularly imprinted polymers (MIPs), electrolyte-gated transistors (EGTs) based on 2D materials, and machine learning (ML) to detect hippuric acid (HA) in artificial urine, being a critical marker for toluene intoxication, parasitic infections, and kidney and bowel inflammation. Reduced graphene oxide (rGO) was used as the sensory material and molecularly imprinted polymer (MIP) as supramolecular receptors. Employing supervised ML techniques based on symbolic regression and compressive sensing enabled us to comprehensively analyze the EGT transfer curves, eliminating the need for arbitrary signal selection and allowing a multivariate analysis during HA detection. The resulting device displayed simultaneously low operating voltages (<0.5 V), rapid response times (≤10 s), operation across a wide range of HA concentrations (from 0.05 to 200 nmol L–1), and a low limit of detection (LoD) of 39 pmol L–1. Thanks to the ML multivariate analysis, we achieved a 2.5-fold increase in the device sensitivity (1.007 μA/nmol L–1) with respect to the human data analysis (0.388 μA/nmol L–1). Our method represents a major advance in PoC technologies, by enabling the accurate determination of small-molecule markers in complex media via the combination of ML analysis, supramolecular analyte recognition, and electrolytic transistors.
Barcelos, Ingrid D.; Oliveira, Raphaela; Schleder, Gabriel R.; Matos, Matheus J. S.; Longuinhos, Raphael; Ribeiro-Soares, Jenaina; Barboza, Ana Paula M.; Prado, Mariana C.; Pinto, Elisângela S.; Gobato, Yara Galvão; Chacham, Hélio; Neves, Bernardo R. A.; Cadore, Alisson R.
Phyllosilicates as earth-abundant layered materials for electronics and optoelectronics: Prospects and challenges in their ultrathin limit Journal Article
Em: vol. 134, não 9, 2023, ISSN: 1089-7550.
@article{Barcelos2023,
title = {Phyllosilicates as earth-abundant layered materials for electronics and optoelectronics: Prospects and challenges in their ultrathin limit},
author = {Ingrid D. Barcelos and Raphaela Oliveira and Gabriel R. Schleder and Matheus J. S. Matos and Raphael Longuinhos and Jenaina Ribeiro-Soares and Ana Paula M. Barboza and Mariana C. Prado and Elisângela S. Pinto and Yara Galvão Gobato and Hélio Chacham and Bernardo R. A. Neves and Alisson R. Cadore},
doi = {10.1063/5.0161736},
issn = {1089-7550},
year = {2023},
date = {2023-09-07},
volume = {134},
number = {9},
publisher = {AIP Publishing},
abstract = {<jats:p>Phyllosilicate minerals are an emerging class of naturally occurring layered insulators with large bandgap energy that have gained attention from the scientific community. This class of lamellar materials has been recently explored at the ultrathin two-dimensional level due to their specific mechanical, electrical, magnetic, and optoelectronic properties, which are crucial for engineering novel devices (including heterostructures). Due to these properties, phyllosilicate minerals can be considered promising low-cost nanomaterials for future applications. In this Perspective article, we will present relevant features of these materials for their use in potential 2D-based electronic and optoelectronic applications, also discussing some of the major challenges in working with them.</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:p>Phyllosilicate minerals are an emerging class of naturally occurring layered insulators with large bandgap energy that have gained attention from the scientific community. This class of lamellar materials has been recently explored at the ultrathin two-dimensional level due to their specific mechanical, electrical, magnetic, and optoelectronic properties, which are crucial for engineering novel devices (including heterostructures). Due to these properties, phyllosilicate minerals can be considered promising low-cost nanomaterials for future applications. In this Perspective article, we will present relevant features of these materials for their use in potential 2D-based electronic and optoelectronic applications, also discussing some of the major challenges in working with them.</jats:p>
Morales-Durán, Nicolás; Wang, Jie; Schleder, Gabriel R.; Angeli, Mattia; Zhu, Ziyan; Kaxiras, Efthimios; Repellin, Cécile; Cano, Jennifer
Pressure-enhanced fractional Chern insulators along a magic line in moiré transition metal dichalcogenides Journal Article
Em: Phys. Rev. Res., vol. 5, iss. 3, pp. L032022, 2023.
@article{PhysRevResearch.5.L032022,
title = {Pressure-enhanced fractional Chern insulators along a magic line in moiré transition metal dichalcogenides},
author = {Nicolás Morales-Durán and Jie Wang and Gabriel R. Schleder and Mattia Angeli and Ziyan Zhu and Efthimios Kaxiras and Cécile Repellin and Jennifer Cano},
url = {https://link.aps.org/doi/10.1103/PhysRevResearch.5.L032022},
doi = {10.1103/PhysRevResearch.5.L032022},
year = {2023},
date = {2023-08-01},
journal = {Phys. Rev. Res.},
volume = {5},
issue = {3},
pages = {L032022},
publisher = {American Physical Society},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Claro, Pedro I. C.; Borges, Egon P. B. S.; Schleder, Gabriel R.; Archilha, Nathaly L.; Pinto, Allan; Carvalho, Murilo; Driemeier, Carlos E.; Fazzio, Adalberto; Gouveia, Rubia F.
From micro- to nano- and time-resolved x-ray computed tomography: Bio-based applications, synchrotron capabilities, and data-driven processing Journal Article
Em: vol. 10, não 2, 2023, ISSN: 1931-9401.
@article{Claro2023,
title = {From micro- to nano- and time-resolved x-ray computed tomography: Bio-based applications, synchrotron capabilities, and data-driven processing},
author = {Pedro I. C. Claro and Egon P. B. S. Borges and Gabriel R. Schleder and Nathaly L. Archilha and Allan Pinto and Murilo Carvalho and Carlos E. Driemeier and Adalberto Fazzio and Rubia F. Gouveia},
doi = {10.1063/5.0129324},
issn = {1931-9401},
year = {2023},
date = {2023-06-01},
volume = {10},
number = {2},
publisher = {AIP Publishing},
abstract = {<jats:p>X-ray computed microtomography (μCT) is an innovative and nondestructive versatile technique that has been used extensively to investigate bio-based systems in multiple application areas. Emerging progress in this field has brought countless studies using μCT characterization, revealing three-dimensional (3D) material structures and quantifying features such as defects, pores, secondary phases, filler dispersions, and internal interfaces. Recently, x-ray computed tomography (CT) beamlines coupled to synchrotron light sources have also enabled computed nanotomography (nCT) and four-dimensional (4D) characterization, allowing in situ, in vivo, and in operando characterization from the micro- to nanostructure. This increase in temporal and spatial resolutions produces a deluge of data to be processed, including real-time processing, to provide feedback during experiments. To overcome this issue, deep learning techniques have risen as a powerful tool that permits the automation of large amounts of data processing, availing the maximum beamline capabilities. In this context, this review outlines applications, synchrotron capabilities, and data-driven processing, focusing on the urgency of combining computational tools with experimental data. We bring a recent overview on this topic to researchers and professionals working not only in this and related areas but also to readers starting their contact with x-ray CT techniques and deep learning.</jats:p>},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
<jats:p>X-ray computed microtomography (μCT) is an innovative and nondestructive versatile technique that has been used extensively to investigate bio-based systems in multiple application areas. Emerging progress in this field has brought countless studies using μCT characterization, revealing three-dimensional (3D) material structures and quantifying features such as defects, pores, secondary phases, filler dispersions, and internal interfaces. Recently, x-ray computed tomography (CT) beamlines coupled to synchrotron light sources have also enabled computed nanotomography (nCT) and four-dimensional (4D) characterization, allowing in situ, in vivo, and in operando characterization from the micro- to nanostructure. This increase in temporal and spatial resolutions produces a deluge of data to be processed, including real-time processing, to provide feedback during experiments. To overcome this issue, deep learning techniques have risen as a powerful tool that permits the automation of large amounts of data processing, availing the maximum beamline capabilities. In this context, this review outlines applications, synchrotron capabilities, and data-driven processing, focusing on the urgency of combining computational tools with experimental data. We bring a recent overview on this topic to researchers and professionals working not only in this and related areas but also to readers starting their contact with x-ray CT techniques and deep learning.</jats:p>
Costa, Marcio; Focassio, Bruno; Canonico, Luis M.; Cysne, Tarik P.; Schleder, Gabriel R.; Muniz, R. B.; Fazzio, Adalberto; Rappoport, Tatiana G.
Connecting Higher-Order Topology with the Orbital Hall Effect in Monolayers of Transition Metal Dichalcogenides Journal Article
Em: Phys. Rev. Lett., vol. 130, iss. 11, pp. 116204, 2023.
@article{PhysRevLett.130.116204,
title = {Connecting Higher-Order Topology with the Orbital Hall Effect in Monolayers of Transition Metal Dichalcogenides},
author = {Marcio Costa and Bruno Focassio and Luis M. Canonico and Tarik P. Cysne and Gabriel R. Schleder and R. B. Muniz and Adalberto Fazzio and Tatiana G. Rappoport},
url = {https://link.aps.org/doi/10.1103/PhysRevLett.130.116204},
doi = {10.1103/PhysRevLett.130.116204},
year = {2023},
date = {2023-03-01},
journal = {Phys. Rev. Lett.},
volume = {130},
issue = {11},
pages = {116204},
publisher = {American Physical Society},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Hsieh, Valerie; Halbertal, Dorri; Finney, Nathan R.; Zhu, Ziyan; Gerber, Eli; Pizzochero, Michele; Kucukbenli, Emine; Schleder, Gabriel R.; Angeli, Mattia; Watanabe, Kenji; Taniguchi, Takashi; Kim, Eun-Ah; Kaxiras, Efthimios; Hone, James; Dean, Cory R.; Basov, D. N.
Domain-Dependent Surface Adhesion in Twisted Few-Layer Graphene: Platform for Moiré-Assisted Chemistry Journal Article
Em: Nano Letters, vol. 23, não 8, pp. 3137-3143, 2023, (PMID: 37036942).
@article{doi:10.1021/acs.nanolett.2c04137,
title = {Domain-Dependent Surface Adhesion in Twisted Few-Layer Graphene: Platform for Moiré-Assisted Chemistry},
author = {Valerie Hsieh and Dorri Halbertal and Nathan R. Finney and Ziyan Zhu and Eli Gerber and Michele Pizzochero and Emine Kucukbenli and Gabriel R. Schleder and Mattia Angeli and Kenji Watanabe and Takashi Taniguchi and Eun-Ah Kim and Efthimios Kaxiras and James Hone and Cory R. Dean and D. N. Basov},
url = {https://doi.org/10.1021/acs.nanolett.2c04137},
doi = {10.1021/acs.nanolett.2c04137},
year = {2023},
date = {2023-01-01},
journal = {Nano Letters},
volume = {23},
number = {8},
pages = {3137-3143},
abstract = {Twisted van der Waals multilayers are widely regarded as a rich platform to access novel electronic phases thanks to the multiple degrees of freedom available for controlling their electronic and chemical properties. Here, we propose that the stacking domains that form naturally due to the relative twist between successive layers act as an additional ”knob” for controlling the behavior of these systems and report the emergence and engineering of stacking domain-dependent surface chemistry in twisted few-layer graphene. Using mid-infrared near-field optical microscopy and atomic force microscopy, we observe a selective adhesion of metallic nanoparticles and liquid water at the domains with rhombohedral stacking configurations of minimally twisted double bi- and trilayer graphene. Furthermore, we demonstrate that the manipulation of nanoparticles located at certain stacking domains can locally reconfigure the moiré superlattice in their vicinity at the micrometer scale. Our findings establish a new approach to controlling moiré-assisted chemistry and nanoengineering.},
note = {PMID: 37036942},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Twisted van der Waals multilayers are widely regarded as a rich platform to access novel electronic phases thanks to the multiple degrees of freedom available for controlling their electronic and chemical properties. Here, we propose that the stacking domains that form naturally due to the relative twist between successive layers act as an additional ”knob” for controlling the behavior of these systems and report the emergence and engineering of stacking domain-dependent surface chemistry in twisted few-layer graphene. Using mid-infrared near-field optical microscopy and atomic force microscopy, we observe a selective adhesion of metallic nanoparticles and liquid water at the domains with rhombohedral stacking configurations of minimally twisted double bi- and trilayer graphene. Furthermore, we demonstrate that the manipulation of nanoparticles located at certain stacking domains can locally reconfigure the moiré superlattice in their vicinity at the micrometer scale. Our findings establish a new approach to controlling moiré-assisted chemistry and nanoengineering.
Morales-Durán, Nicolás; Wang, Jie; Schleder, Gabriel R.; Angeli, Mattia; Zhu, Ziyan; Kaxiras, Efthimios; Repellin, Cécile; Cano, Jennifer
Pressure–enhanced fractional Chern insulators in moiré transition metal dichalcogenides along a magic line Working paper
2023.
@workingpaper{moralesdurán2023pressureenhanced,
title = {Pressure–enhanced fractional Chern insulators in moiré transition metal dichalcogenides along a magic line},
author = {Nicolás Morales-Durán and Jie Wang and Gabriel R. Schleder and Mattia Angeli and Ziyan Zhu and Efthimios Kaxiras and Cécile Repellin and Jennifer Cano},
url = {https://arxiv.org/abs/2304.06669},
doi = {10.48550/arXiv.2304.06669},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
abstract = {We show that pressure applied to twisted WSe2 can enhance the many-body gap and region of stability of a fractional Chern insulator at filling ν=1/3. Our results are based on exact diagonalization of a continuum model, whose pressure-dependence is obtained through it ab initio methods. We interpret our results in terms of a it magic line in the pressure-it vs-twist angle phase diagram: along the magic line, the bandwidth of the topmost moiré valence band is minimized while simultaneously its quantum geometry nearly resembles that of an ideal Chern band. We expect our results to generalize to other twisted transition metal dichalcogenide homobilayers.},
keywords = {},
pubstate = {published},
tppubtype = {workingpaper}
}
We show that pressure applied to twisted WSe2 can enhance the many-body gap and region of stability of a fractional Chern insulator at filling ν=1/3. Our results are based on exact diagonalization of a continuum model, whose pressure-dependence is obtained through it ab initio methods. We interpret our results in terms of a it magic line in the pressure-it vs-twist angle phase diagram: along the magic line, the bandwidth of the topmost moiré valence band is minimized while simultaneously its quantum geometry nearly resembles that of an ideal Chern band. We expect our results to generalize to other twisted transition metal dichalcogenide homobilayers.
Li, Zhijie; Tabataba-Vakili, Farsane; Zhao, Shen; Rupp, Anna; Bilgin, Ismail; Herdegen, Ziria; März, Benjamin; Watanabe, Kenji; Taniguchi, Takashi; Schleder, Gabriel R.; Baimuratov, Anvar S.; Kaxiras, Efthimios; Müller-Caspary, Knut; Högele, Alexander
Lattice Reconstruction in MoSe2–WSe2 Heterobilayers Synthesized by Chemical Vapor Deposition Journal Article
Em: Nano Letters, vol. 23, não 10, pp. 4160-4166, 2023, (PMID: 37141148).
@article{doi:10.1021/acs.nanolett.2c05094,
title = {Lattice Reconstruction in MoSe2–WSe2 Heterobilayers Synthesized by Chemical Vapor Deposition},
author = {Zhijie Li and Farsane Tabataba-Vakili and Shen Zhao and Anna Rupp and Ismail Bilgin and Ziria Herdegen and Benjamin März and Kenji Watanabe and Takashi Taniguchi and Gabriel R. Schleder and Anvar S. Baimuratov and Efthimios Kaxiras and Knut Müller-Caspary and Alexander Högele},
url = {https://doi.org/10.1021/acs.nanolett.2c05094},
doi = {10.1021/acs.nanolett.2c05094},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Nano Letters},
volume = {23},
number = {10},
pages = {4160-4166},
abstract = {Vertical van der Waals heterostructures of semiconducting transition metal dichalcogenides realize moiré systems with rich correlated electron phases and moiré exciton phenomena. For material combinations with small lattice mismatch and twist angles as in MoSe2–WSe2, however, lattice reconstruction eliminates the canonical moiré pattern and instead gives rise to arrays of periodically reconstructed nanoscale domains and mesoscopically extended areas of one atomic registry. Here, we elucidate the role of atomic reconstruction in MoSe2–WSe2 heterostructures synthesized by chemical vapor deposition. With complementary imaging down to the atomic scale, simulations, and optical spectroscopy methods, we identify the coexistence of moiré-type cores and extended moiré-free regions in heterostacks with parallel and antiparallel alignment. Our work highlights the potential of chemical vapor deposition for applications requiring laterally extended heterosystems of one atomic registry or exciton-confining heterostack arrays.},
note = {PMID: 37141148},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Vertical van der Waals heterostructures of semiconducting transition metal dichalcogenides realize moiré systems with rich correlated electron phases and moiré exciton phenomena. For material combinations with small lattice mismatch and twist angles as in MoSe2–WSe2, however, lattice reconstruction eliminates the canonical moiré pattern and instead gives rise to arrays of periodically reconstructed nanoscale domains and mesoscopically extended areas of one atomic registry. Here, we elucidate the role of atomic reconstruction in MoSe2–WSe2 heterostructures synthesized by chemical vapor deposition. With complementary imaging down to the atomic scale, simulations, and optical spectroscopy methods, we identify the coexistence of moiré-type cores and extended moiré-free regions in heterostacks with parallel and antiparallel alignment. Our work highlights the potential of chemical vapor deposition for applications requiring laterally extended heterosystems of one atomic registry or exciton-confining heterostack arrays.
Nicoliche, Caroline Y. N.; Silva, Giulia S.; Gomes-de-Pontes, Leticia; Schleder, Gabriel R.; Lima, Renato S.
Em: Garcia-Cordero, Jose L.; Revzin, Alexander (Ed.): Microfluidic Systems for Cancer Diagnosis, pp. 83–94, Springer US, New York, NY, 2023, ISBN: 978-1-0716-3271-0.
@inbook{Nicoliche2023,
title = {Single-Response Electronic Tongue and Machine Learning Enable the Multidetermination of Extracellular Vesicle Biomarkers for Cancer Diagnostics Without Recognition Elements},
author = {Caroline Y. N. Nicoliche and Giulia S. Silva and Leticia Gomes-de-Pontes and Gabriel R. Schleder and Renato S. Lima},
editor = {Jose L. Garcia-Cordero and Alexander Revzin},
url = {https://doi.org/10.1007/978-1-0716-3271-0_6},
doi = {10.1007/978-1-0716-3271-0_6},
isbn = {978-1-0716-3271-0},
year = {2023},
date = {2023-01-01},
booktitle = {Microfluidic Systems for Cancer Diagnosis},
pages = {83–94},
publisher = {Springer US},
address = {New York, NY},
abstract = {Platforms based on impedimetric electronic tongue (nonselective sensor) and machine learning are promising to bring disease screening biosensors into mainstream use toward straightforward, fast, and accurate analyses at the point-of-care, thus contributing to rationalize and decentralize laboratory tests with social and economic impacts being achieved. By combining a low-cost and scalable electronic tongue with machine learning, in this chapter, we describe the simultaneous determination of two extracellular vesicle (EV) biomarkers, i.e., the concentrations of EV and carried proteins, in mice blood with Ehrlich tumor from a single impedance spectrum without using biorecognizing elements. This tumor shows primary features of mammary tumor cells. Pencil HB core electrodes are integrated into polydimethylsiloxane (PDMS) microfluidic chip. The platform shows the highest throughput in comparison with the methods addressed in the literature to determine EV biomarkers.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Platforms based on impedimetric electronic tongue (nonselective sensor) and machine learning are promising to bring disease screening biosensors into mainstream use toward straightforward, fast, and accurate analyses at the point-of-care, thus contributing to rationalize and decentralize laboratory tests with social and economic impacts being achieved. By combining a low-cost and scalable electronic tongue with machine learning, in this chapter, we describe the simultaneous determination of two extracellular vesicle (EV) biomarkers, i.e., the concentrations of EV and carried proteins, in mice blood with Ehrlich tumor from a single impedance spectrum without using biorecognizing elements. This tumor shows primary features of mammary tumor cells. Pencil HB core electrodes are integrated into polydimethylsiloxane (PDMS) microfluidic chip. The platform shows the highest throughput in comparison with the methods addressed in the literature to determine EV biomarkers.
Schleder, Gabriel R.; Pizzochero, Michele; Kaxiras, Efthimios
One-Dimensional Moiré Physics and Chemistry in Heterostrained Bilayer Graphene Working paper
2023.
@workingpaper{schleder2023onedimensional,
title = {One-Dimensional Moiré Physics and Chemistry in Heterostrained Bilayer Graphene},
author = {Gabriel R. Schleder and Michele Pizzochero and Efthimios Kaxiras},
url = {https://arxiv.org/abs/2306.09799},
doi = {10.48550/arXiv.2306.09},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
abstract = {Twisted bilayer graphene (tBLG) has emerged as a promising platform to explore exotic electronic phases. However, the formation of moiré patterns in tBLG has thus far been confined to the introduction of twist angles between the layers. Here, we propose heterostrained bilayer graphene (hBLG), as an alternative avenue to access twist-angle-free moiré physics via lattice mismatch. Using atomistic and first-principles calculations, we demonstrate that uniaxial heterostrain can promote isolated flat electronic bands around the Fermi level. Furthermore, the heterostrain-induced out-of-plane lattice relaxation may lead to a spatially modulated reactivity of the surface layer, paving the way for the moiré-driven chemistry and magnetism. We anticipate that our findings can be readily generalized to other layered materials.},
keywords = {},
pubstate = {published},
tppubtype = {workingpaper}
}
Twisted bilayer graphene (tBLG) has emerged as a promising platform to explore exotic electronic phases. However, the formation of moiré patterns in tBLG has thus far been confined to the introduction of twist angles between the layers. Here, we propose heterostrained bilayer graphene (hBLG), as an alternative avenue to access twist-angle-free moiré physics via lattice mismatch. Using atomistic and first-principles calculations, we demonstrate that uniaxial heterostrain can promote isolated flat electronic bands around the Fermi level. Furthermore, the heterostrain-induced out-of-plane lattice relaxation may lead to a spatially modulated reactivity of the surface layer, paving the way for the moiré-driven chemistry and magnetism. We anticipate that our findings can be readily generalized to other layered materials.