TellWell
← Back to feed
Publications3h ago88% confidenceConfidence 88% — the share of independent, credible sources corroborating the core facts.

Machine Learning and Simulation Accelerate Discovery of High-Efficiency Lead-Free Perovskite Solar Cells

Center 100%
1 source

Researchers combined SCAPS-1D device simulations with machine learning to systematically screen 125 perovskite solar cell architectures, identifying lead-free designs with predicted power conversion efficiencies up to 28.85%. The study used a representative subset of configurations to train an ML model that reliably ranked unexplored structures, with SHAP analysis revealing that hole transport layer band gap, absorber band gap, and electron transport layer electron affinity are the most influential performance factors. This physics-based, data-driven approach could accelerate development of efficient, environmentally friendly perovskite solar cells while providing transparent design principles applicable to other optoelectronic systems.

Researchers integrated SCAPS-1D device simulations with machine learning to address the combinatorial challenge of designing multilayer perovskite solar cells. The study systematically explored 125 device architectures constructed from five electron transport layers, five absorbers (including lead-free double perovskites), and five hole transport layers. A representative subset was used to train an ML model that predicted power conversion efficiency (PCE) for unexplored structures, with Leave-One-Group-Out cross-validation demonstrating reliable ranking capability. SHAP analysis identified the most influential material descriptors: hole transport layer band gap, absorber band gap, and electron transport layer electron affinity. The workflow identified several high-performance configurations verified through full simulations, with the device FTO/TiO₂/Cs₂AgBiBr₆/NiO/Ag achieving 28.85% PCE and the ML-suggested FTO/SnO₂/Cs₂AgInBr₆/NiO/Ag reaching 28.62%, outperforming related literature architectures by approximately 4% absolute. Notably, eight of the top-11 structures employed the lead-free double perovskite Cs₂AgInBr₆.

What's missing

The study relies on SCAPS-1D simulations rather than experimental validation of the predicted high-efficiency devices, which represents a limitation in confirming whether the theoretical PCE values are achievable in practice. The paper does not discuss stability, manufacturing feasibility, or cost considerations of the proposed lead-free perovskite architectures.

What different sources said

  • Multilayer Screening of Double and Conventional Perovskite Solar Cells Using SCAPS-1D and Machine Learning: Optimization of ETL, HTL, and Absorber for High-Efficiency Architectures

Related

PublicationsConfidence 82% — the share of independent, credible sources corroborating the core facts.

Genetic Drift, Not Selection, Drives Rapid Feather Color Evolution in Island Bird Radiation

A new study of an island bird radiation found that rapid evolution of feather coloration is driven primarily by genetic drift in small populations rather than sexual or ecological selection. The research integrated whole-genome data with detailed plumage measurements across complete species sampling to test whether signaling trait evolution correlates with speciation rates. The findings suggest that neutral demographic processes play a central role in generating phenotypic diversity during island radiations, challenging assumptions about the mechanisms driving rapid evolution.

1 source7m ago
PublicationsConfidence 82% — the share of independent, credible sources corroborating the core facts.

New AI Model Improves Prediction of Therapeutic Peptide Function from Protein Sequences

Researchers developed a lightweight CNN classifier that predicts whether peptide sequences have therapeutic properties, trained on a database of 54,655 peptides across 48 functional categories. The model uses a novel negative sampling strategy to reduce false positive rates from over 60% in previous approaches to 2.1%. This advancement could accelerate drug discovery by enabling faster computational screening of peptide candidates before expensive experimental testing.

1 source14m ago
PublicationsConfidence 82% — the share of independent, credible sources corroborating the core facts.

Study Shows Different Metabolic Stress Models Produce Distinct Effects on Human Neuronal Networks

Researchers tested three common in vitro metabolic stress models on human-derived neuronal networks and found each produced different patterns of neuronal activity and cell damage. The models tested were hypoxia alone, oxygen-glucose deprivation (OGD), and hypoxia combined with glutamate exposure. The findings suggest that choice of experimental model significantly affects results and that combining electrophysiological and structural analyses is important for accurately assessing metabolic stress in stroke research.

1 source14m ago