This paper investigates a coupled Navier-Stokes-dual-porosity-poroelasticity model, which consists of three components: conduits, matrix and micro-fractures. This model characterizes fluid flow and deformation phenomena in free flow region and dual-porosity poroelastic system. To overcome potential numerical locking, a total pressure is introduced for the dual-porosity poroelastic system, while an upwind scheme is applied to the convective term to enhance numerical stability. Based on this, a semi-discrete scheme is first constructed using a discontinuous Galerkin method on polygonal meshes, and its well-posedness is established under a suitable small data condition. A fully discrete scheme is subsequently derived by applying the backward Euler method for temporal discretization. Additionally, we conduct the error estimates for these two discrete schemes. Numerical results demonstrate that the proposed method achieves good convergence across different mesh types and exhibits robustness and high accuracy in several representative scenarios. These include two-dimensional cases, such as coupled surface and subsurface hydrological systems and open-hole/cased-hole completions, and a three-dimensional channel filtration problem. These results confirm the method's effectiveness and applicability in simulating practical complex flows.
PET circularity is often evaluated by recovery volumes, although performance ultimately depends on whether recovered material can be converted into quality-assured outputs and absorbed by viable end markets. Yet the mechanisms connecting policy, markets, technology, and stakeholder behavior remain dispersed across heterogeneous evidence. We developed a source-attributed, time-stamped causal-network approach using 750 Australian government, industry and NGO, corporate, and media documents published during 2010–2025. Causal extraction, semantic normalization, sentiment assignment, and graph construction yielded 2794 factor nodes and 1588 unique causal edges. The network was strongly externalized: policy, market, and social factors were more numerous and highly connected than internal technical and operational factors. Persistent barriers combined cost and price volatility, inconsistent recyclate quality, infrastructure gaps, and weak policy implementation. Positive mechanisms clustered around standards and measurement, design for recyclability, operational capacity, and end-market development; their circular value depended on retained material quality and the end use reached. Source attribution revealed systematic framing differences, with media coverage concentrating negative implementation narratives. Temporal and regional analyses indicated a shift from limited attention to rapid expansion and subsequent consolidation, alongside national structural hubs and differentiated state pathways. A full-text academic shadow corpus preserved the broad problem domains but changed their hierarchy: academic evidence foregrounded material and processing mechanisms, whereas the primary corpus foregrounded institutional coordination and market implementation. The contribution is therefore not a universal ranking of Australian interventions, but an auditable framework for identifying how technical feasibility becomes, or fails to become, operational circularity under different institutional conditions.
Developing high-performance reactant-adaptive electrocatalysts for oxygen evolution reaction (OER) and urea oxidation reaction (UOR) is of great importance for sustained green hydrogen generation and simultaneously water remediation. Enabling the efficient and reversible interconversion of active sites can intrinsically address the sluggish reaction kinetics. Herein, we fabricated a heterostructure (Ni(OH)2/Co MOF/NF) featuring nickel (Ni) sites with efficient and reversible interconversion between Ni3+ and Ni2+. These sites exhibit exceptional reactant-adaptive catalytic performance: the current density of 100 mA cm- 2 is achieved in the OER process at an overpotential of only 290 mV, and current densities of 10 and 100 mA cm- 2 are attained in the UOR process at potentials of merely 1.25 and 1.33 V versus reversible hydrogen electrode, respectively. The Ni(OH)2/Co MOF/ NF catalyst maintains stability for more than 110 h at 10 mA cm- 2 for both OER and UOR. In situ spectroscopy measurements and density functional theory calculations reveal that the heterostructure regulates the electronic structure of nickel sites, modulates the interaction between key intermediates and nickel sites, and lowers the reaction energy barrier, thereby accelerating reaction kinetics and enhancing UOR and OER performance. This work provides new insights for developing highly efficient electrocatalysts in the clean energy and wastewater treatment fields.
Starch digestibility is strongly influenced by crystalline polymorphism and processing conditions, yet comparative studies under practical cooking conditions remain limited. This study investigated the multi-scale structural reorganization and in vitro digestibility of A-, B-, and C-type starches after conventional cooking (boiling at 100 degrees C for 20 min). Cooking disrupted granule surface barriers and lamellar structures, thereby increasing alpha-amylase accessibility and content of rapidly digestible starch (RDS) for all samples. Among the cooked starches, A-type starch exhibited the highest hydrolysis rate constant (k = 0.294 min(-1)) and lowest resistant starch (RS) retention (27.83%), whereas C-type starch showed the lowest k value (0.230 min(-1)) and highest RS retention (31.25%), indicating greater resistance to enzymatic digestion. Multi-scale analyses using X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FT-IR), and solid-state C-13 cross-polarization/magic-anglespinning nuclear magnetic resonance spectroscopy (C-13 CP/MAS NMR) revealed that C-type starch may retain partially ordered B-type-associated regions during cooking, while A-type starch undergoes extensive crystalline lattice disruption. B-type starch formed V-type complexes, which may have partially compensated for the cooking-induced loss of digestion resistance. These transformations provide a structural basis for the divergent digestion kinetics of the three polymorphs. The results demonstrate how crystalline polymorphism and cooking interactively regulate starch digestibility and suggest that C-type starch may serve as a low-glycemic index (lowGI) ingredient, whereas A-type starch may be suitable for starch-based formulations designed for rapid digestibility.
Software defect prediction is an important technique for improving software quality and reducing maintenance costs. In recent years, neural network based graph representations have attracted widespread attention in defect prediction for their ability to capture both structural and semantic features of source codes. However, existing studies largely focus on learning effective graph representations, with little systematic analysis of how different program graphs and graph neural network (GNN) models interact. To address this gap, we present an empirical study that systematically investigates the collaborative effects of 24 unique Graph-GNN combinations on defect prediction performance, which are formed from 6 program graphs (i.e., AST, CFG, CDG, DDG, PDG, and CPG) and 4 GNN models (i.e., GCN, GAT, GIN, and GraphSAGE). Extensive experiments are conducted on 15 versions of 7 open-source Java projects under both within-project defect prediction (WPDP) and cross-project defect prediction (CPDP) settings. Five widely used evaluation measures and the SKESD statistical significance test are applied to compare and rank the performance of different combinations. The results show significant differences in both traditional evaluation measures and effort-aware performance measures among the Graph-GNN combinations, with CFG or CPG combined with GAT or GCN demonstrate the strongest overall performance, suggesting that CFG-GCN is recommended for WPDP, while CPG-GAT is preferred for CPDP. Overall, we provide systematic empirical evidence and practical guidance for selecting effective graph-based GNN models for future defect prediction research.
