Yangtze River Delta Physics Research Center (China)

The aim of this study was to investigate whether complex auditory brainstem response (c-ABR) outcomes and speech-in-noise perception performance are associated with occupational noise exposure in young noise-exposed workers with normal hearing thresholds and preserved otoacoustic emissions, a population at high risk of cochlear synaptopathy. Eighty-eight male participants (aged 18–40 years) from Zhejiang province, China, were recruited into two groups: 44 workers exposed to occupational noise and 44 controls from quieter workplaces. All participants had normal hearing thresholds and presented with distortion product otoacoustic emissions. For the group of noise-exposed workers, noise exposure levels were measured using shift-long recordings, and cumulative noise exposure (CNE) was calculated. Speech-in-noise perception was assessed using the Mandarin Bamford-Kowal-Bench (BKB) test. c-ABR recordings in response to a 40-ms/da/syllable were obtained. Peak-to-trough amplitudes and latencies of the frequency-following response (FFR) waves were analyzed to assess neuronal encoding fidelity at the brainstem level. The noise-exposed group exhibited a significant delay in C-wave latency but no other c-ABR differences were found. No significant group differences were observed in BKB scores, though a correlation emerged between noise exposure levels and speech-in-noise perception within the noise-exposed group. Higher noise exposure levels were correlated with better speech perception in noise (lower signal-to-noise ratio results for the BKB test). Our results do not strongly support an effect of occupational noise exposure on speech encoding at the brainstem level. In this sample, occupational noise exposure was not linked to indicators of cochlear synaptopathy, as assessed through c-ABR and BKB measures.

Photovoltaic (PV) technology is the core pathway foraddressingglobal climate change and advancing energy system decarbonization,yet the rapid expansion of PV manufacturing capacity has triggereda surge in life-cycle greenhouse gas emissions, sparking mountingconcerns. We integrated multisource heterogeneous data from China’sPV industry (2005–2024) to develop a life-cycle accountingframework, which quantifies industrial carbon emissions and theirevolutionary patterns across production stages and multiscale spatiotemporaldimensions. We also deconstructed the emission impacts of scale, technology,and structural factors, and predicted future trends. Over two decades,China’s PV industry-wide carbon emissions soared from 0.24to 205 million tonnes, while product-level emission intensity plummetedfrom 1,300 to 380 kg CO2eq/kWp. The contribution of technologicalprogress to emission reduction rose from about 3% of the observedincrease in emissions in 2005–2007 to nearly 100% in 2020–2024.Spatially, raw material and monocrystalline cell production have shiftedinland for cost advantages, while module assembly remains concentratedin coastal hubs like the Yangtze River Delta. Capacity utilization,grid decarbonization, and technical learning will dictate future emissions.Against surging global PV demand, coordinated capacity planning, acceleratedtech progress, optimized spatial distribution, and established incentivepolicies are pivotal to steering China’s PV manufacturing ontoa sustainable low-carbon path.

Membrane distillation (MD) provides an energy-efficient method to remove PFAS using vapor pressure differences. This study develops glucose-derived carbon-coated ceramic membranes, demonstrating their potential for PFAS removal. The membranes, prepared by heat-treating an aqueous glucose solution on porous alumina under a nitrogen atmosphere, exhibited a water contact angle of approximately 118°. Scanning electron microscopy (SEM) and electron probe microanalysis (EPMA) analyses confirmed carbon coverage on the alumina particles, while mercury porosimetry showed that the pore size decreased with increasing carbon content. The optimal membrane, with 18 mg of carbon and a pore size of 257 nm, achieved over 95% PFAS removal and a water vapor flux of 15 kg/m2h.

ABSTRACT In the node classification task on graphs, mainstream graph neural networks often adopt the same neighborhood aggregation strategy for all nodes. This approach results in biased learning toward features from the majority nodes while devaluing those from the minority nodes under class imbalance. In particular, for marginal nodes, limited neighboring information can severely impact classification performance. To address this challenge, this paper proposes a feature fusion‐based hybrid quantum‐classical graph residual neural network (QGRNN). Leveraging the nonlinear expressive capacity of qubits in modeling complex feature interactions, the model innovatively integrates a structure‐driven node selection mechanism with a quantum feature enhancement module, while also dynamically fusing classical features and Hamiltonian expectation values through a gated residual fusion mechanism to compensate for representational deficiencies of marginal nodes overlooked by classical methods. Experimental results show that QGRNN consistently outperforms baselines across a range of node classification tasks. In binary and ternary classification settings, it exhibits strong discriminative capability and robustness, especially maintaining high accuracy under severe class imbalance. In addition, QGRNN also demonstrates strong generality in other tasks, in recommendation scenarios, achieving an average improvement of 36.3% on NDCG@5, 44.3% on Recall@5, and 21.9% on Recall@10 compared to the baseline.

In this Letter, we present a mode manipulation strategy based on femtosecond-laser-induced local modification of a fluoride microsphere resonator. Two symmetric modification regions are precisely inscribed on both sides of the equatorial plane of the resonator, introducing a permanent refractive index modification through nonlinear laser-material interactions. This engineered modification effectively suppresses high-order axial modes, enabling stable single-mode lasing at 1918.58 nm with a low threshold of 2 mW. The proposed approach simplifies the resonance spectrum and provides a built-in solution to the intrinsic multimode limitations of whispering-gallery-mode (WGM) lasers, offering great potential for integrated photonics and mid-infrared light sources.