• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
Science
Report

Perovskite Solar Cells Achieve ~27.8% Efficiency in Recent Research

Science & Tech Daily roundup highlighted recent peer-reviewed work on perovskite solar cells reaching approximately 27.8% efficiency, representing continued progress in next-generation photovoltaic technology.

RY
1 Source, 4h ago, first seen 4h ago

TLDR

Perovskite solar cells have long been pursued as a potential alternative to silicon, offering lower manufacturing costs and potential for flexible applications. Reaching 27.8% efficiency approaches the theoretical limits for single-junction cells and demonstrates continued material science advances. The inclusion in daily science roundups indicates sustained researcher and public interest in renewable energy breakthroughs, even when not garnering standalone viral attention.

Combined views

128

1 Source, first seen 4h ago

2 likes
Featured Source

Combined views

128

1 Source, first seen 4h ago

2 likes

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

1 Source

@Goeun_6121Science & Tech Daily | October 2, 2026 Eight developments worth tracking from October 1 peer-reviewed publications. 1. Researchers in Korea traced how glucose metabolism drives outer membrane vesicle production in Acinetobacter baumannii. Only glucose-oxidizing strains showed acidification and enhanced vesiculation. Adding the PQQ biosynthesis gene cluster to a PQQ-deficient strain recreated both. The vesicles carried the adhesin Ata and the beta-lactamase OXA-51, linking carbon metabolism to envelope remodeling. A culture-based mechanism. Whether this changes infection persistence or antibiotic response in a real host is untested. 2. A 52-week randomized trial mapped semaglutide's kidney effects in type 2 diabetes with MRI, paired biopsies and transcriptomics. Three coprimary MRI endpoints missed significance. Secondaries differed: renal artery resistive index fell, and cortical diffusion suggested slower fibrosis. Biopsy transcriptomics flagged two responsive glomerular endothelial programs and fewer local immune cell signals. The biopsy arm was small, 33 biopsied, 22 with paired single-nucleus data. Secondary analyses are not a confirmed mechanism, and this was a mechanism study, not an outcomes trial. 3. A benchmarking study showed that how we score gene-perturbation AI can change which models look like they work. Across 14 datasets and 18 metrics, mean squared error and control-based correlation diluted real signals concentrated in a few genes. With calibrated metrics and positive controls, several deep-learning models beat an uninformative baseline that had previously matched them. A better scoreboard is not a better therapy. Whether rankings hold in new cells needs prospective testing. 4. Oxygen doping pushed monolayer WSe2 p-type transistors to record currents at a 16-nanometer channel. Mobility rose from 55 to 218.3 cm2/Vs, contact resistance fell from 1,869 to 104 ohm-micrometers. The device hit 1,635 microamps per micrometer of saturation current at room temperature with an 81 percent ballistic ratio. A champion single device. Doping uniformity, threshold control, yield and n-type integration remain open. 5. A proposal prepared surface-code logical states without measurement. A reinforcement-learning strategy found on a small code generalized to arbitrary code distance with only local gates, preserving the protection distance. Under depolarizing-noise simulation, the ancilla-free approach cut logical error rates by up to an order of magnitude versus standard stabilizer measurement. Simulation, not hardware. Circuit depth grows with code distance, and the payoff is largest where measurement is costly, such as neutral-atom and ion-trap platforms. 6. A perovskite-organic tandem solar cell reached a certified 27.82 percent efficiency. An additive homogenized crystallization and suppressed light-induced halide segregation. The champion cell hit 28.20 percent and kept 90 percent of its efficiency after 700 hours of maximum-power-point tracking. One small champion cell does not predict module yield or decades of outdoor life. 7. JWST compared the mineral makeup of 21 extreme debris disks around young stars. About a third were silica-rich, appearing only around stars younger than 300 million years, read as dust from Mars-sized-body collisions like the Moon-forming impact. Silica-poor disks persist across ages. Collisions are inferred, not seen. A larger sample with time-resolved spectra would tighten the case. 8. Microcapsules grown on ice became multi-enzyme reaction spaces. Monomers concentrated in the quasi-liquid layer of an ice sphere template formed a polyamide shell, then the ice melted away. Capsules 30 to 500 micrometers wide drove visible-light CO2-to-methanol conversion up to 80 times higher than the comparison condition. A lab comparison, not an industrial efficiency. Scale-up, defects and real-sample stability are unproven. Across these studies the pattern repeats: deeper observation, but each carries a boundary. A mechanism in culture, a secondary endpoint, a champion cell, a model inference. The question is rarely whether the result is real. It is where the measurement stops.
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    1 Source

    @Goeun_6121Science & Tech Daily | October 2, 2026 Eight developments worth tracking from October 1 peer-reviewed publications. 1. Researchers in Korea traced how glucose metabolism drives outer membrane vesicle production in Acinetobacter baumannii. Only glucose-oxidizing strains showed acidification and enhanced vesiculation. Adding the PQQ biosynthesis gene cluster to a PQQ-deficient strain recreated both. The vesicles carried the adhesin Ata and the beta-lactamase OXA-51, linking carbon metabolism to envelope remodeling. A culture-based mechanism. Whether this changes infection persistence or antibiotic response in a real host is untested. 2. A 52-week randomized trial mapped semaglutide's kidney effects in type 2 diabetes with MRI, paired biopsies and transcriptomics. Three coprimary MRI endpoints missed significance. Secondaries differed: renal artery resistive index fell, and cortical diffusion suggested slower fibrosis. Biopsy transcriptomics flagged two responsive glomerular endothelial programs and fewer local immune cell signals. The biopsy arm was small, 33 biopsied, 22 with paired single-nucleus data. Secondary analyses are not a confirmed mechanism, and this was a mechanism study, not an outcomes trial. 3. A benchmarking study showed that how we score gene-perturbation AI can change which models look like they work. Across 14 datasets and 18 metrics, mean squared error and control-based correlation diluted real signals concentrated in a few genes. With calibrated metrics and positive controls, several deep-learning models beat an uninformative baseline that had previously matched them. A better scoreboard is not a better therapy. Whether rankings hold in new cells needs prospective testing. 4. Oxygen doping pushed monolayer WSe2 p-type transistors to record currents at a 16-nanometer channel. Mobility rose from 55 to 218.3 cm2/Vs, contact resistance fell from 1,869 to 104 ohm-micrometers. The device hit 1,635 microamps per micrometer of saturation current at room temperature with an 81 percent ballistic ratio. A champion single device. Doping uniformity, threshold control, yield and n-type integration remain open. 5. A proposal prepared surface-code logical states without measurement. A reinforcement-learning strategy found on a small code generalized to arbitrary code distance with only local gates, preserving the protection distance. Under depolarizing-noise simulation, the ancilla-free approach cut logical error rates by up to an order of magnitude versus standard stabilizer measurement. Simulation, not hardware. Circuit depth grows with code distance, and the payoff is largest where measurement is costly, such as neutral-atom and ion-trap platforms. 6. A perovskite-organic tandem solar cell reached a certified 27.82 percent efficiency. An additive homogenized crystallization and suppressed light-induced halide segregation. The champion cell hit 28.20 percent and kept 90 percent of its efficiency after 700 hours of maximum-power-point tracking. One small champion cell does not predict module yield or decades of outdoor life. 7. JWST compared the mineral makeup of 21 extreme debris disks around young stars. About a third were silica-rich, appearing only around stars younger than 300 million years, read as dust from Mars-sized-body collisions like the Moon-forming impact. Silica-poor disks persist across ages. Collisions are inferred, not seen. A larger sample with time-resolved spectra would tighten the case. 8. Microcapsules grown on ice became multi-enzyme reaction spaces. Monomers concentrated in the quasi-liquid layer of an ice sphere template formed a polyamide shell, then the ice melted away. Capsules 30 to 500 micrometers wide drove visible-light CO2-to-methanol conversion up to 80 times higher than the comparison condition. A lab comparison, not an industrial efficiency. Scale-up, defects and real-sample stability are unproven. Across these studies the pattern repeats: deeper observation, but each carries a boundary. A mechanism in culture, a secondary endpoint, a champion cell, a model inference. The question is rarely whether the result is real. It is where the measurement stops.
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet