The increasing urgency to mitigate plastic pollution has accelerated the shift from linear manufacturing toward circular systems. This review synthesizes current advances in mechanical, chemical, biological, and upcycling pathways, emphasizing how artificial intelligence (AI) is reshaping decision-making, performance prediction, and system-level optimization. Intelligent sensing technologies—such as FTIR, Raman spectroscopy, hyperspectral imaging, and LIBS—combined with Machine Learning (ML) classifiers have improved material identification, reduced reject rates, and enhanced sorting precision. AI-assisted kinetic modeling, catalyst performance prediction, and enzyme design tools have improved process intensification for pyrolysis, solvolysis, depolymerization, and biocatalysis. Life Cycle Assessment (LCA)-integrated datasets reveal that environmental benefits depend strongly on functional-unit selection, energy decarbonization, and substitution factors rather than mass-based comparisons alone. Case studies across Europe, Latin America, and Asia show that digital traceability, Extended Producer Responsibility (EPR), and full-system costing are pivotal to robust circular outcomes. Upcycling strategies increasingly generate high-value materials and composites, supported by digital twins and surrogate models. Collectively, evidence indicates that AI moves from supportive instrumentation to a structural enabler of transparency, performance assurance, and predictive environmental planning. The convergence of AI-based design, standardized LCA frameworks, and inclusive governance emerges as a necessary foundation for scaling circular plastic systems sustainably.

Gamma-oryzanol (GO) is a bioactive compound that, due to its biological characteristics, can be added to a food matrix. However, the bioactive compound is difficult to incorporate due to its low solubility and stability. A nanoemulsion allows substances to be packaged in nanometric sizes, improving their bioavailability. In this work, a GO nanoemulsion was developed using high-energy techniques. The methodological process began with the formulation of the coarse emulsion, where the emulsifiers (sodium caseinate and citrus pectin), diluent (rice bran oil), and pH were varied to find the most stable formulation. The coarse emulsion was subjected to four high-energy techniques (conventional homogenization, high-pressure homogenization, ultra-high-pressure homogenization, and ultrasonication) to reduce the droplet size. A physical-stability test, rheological-behavior test, image analysis, and particle-size-and-distribution test were conducted to determine which was the best technique. The formulation with the highest stability (pH 5.3) was composed of 87% water, 6.1% sodium caseinate, 0.6% citrus pectin, 6.1% rice bran oil, and 0.2% GO. The ultrasonic treatment obtains the smallest particle size (30.1 ± 1 nm), and the high-pressure treatment obtains the greatest stability (TSI < 0.3), both at 0 and 7 days of storage. High-energy treatments significantly reduce the droplet size of the emulsion, with important differences between each technique.

Chickpea (Cicer arietinum L.) flour is a promising raw material for the development of biodegradable packaging due to its protein and polyphenol content. In this study, thermocompressed chickpea flour sheets were reinforced with cellulose nanocrystals (CNC) to improve their barrier, mechanical, thermal, and structural properties. Preliminary trials identified 22% moisture as the most suitable condition for consistent sheet formation. CNC was incorporated at 0, 2.5, 5.0, and 7.5% (w/w), and the resulting sheets were evaluated for phenolic content, antioxidant activity, water vapor permeability (WVP), optical properties, thermal behavior, morphology, and structural characteristics. Thermocompression reduced the measurable phenolic fractions, although antioxidant activity was not significantly affected. CNC markedly reduced WVP, from 5.16x10^-10 (control) to 5.93x10^-12 g∙m^-1∙s^-1∙Pa^-1 at 7.5% CNC. Tensile strength and Young's modulus increased with CNC loading, while elongation at break was highest at intermediate concentrations. SEM, DSC, XRD, and FTIR analyses indicated matrix reorganization and modified thermo-structural behavior. Overall, CNC improved the barrier and mechanical performance of thermocompressed chickpea flour sheets, supporting their potential for biodegradable packaging applications.

The increasing urgency to mitigate plastic pollution has accelerated the shift from linear manufacturing toward circular systems. This review synthesizes current advances in mechanical, chemical, biological, and upcycling pathways, emphasizing how artificial intelligence (AI) is reshaping decision-making, performance prediction, and system-level optimization. Intelligent sensing technologies — such as FTIR, Raman spectroscopy, hyperspectral imaging, and LIBS — combined with machine-learning classifiers have improved material identification, reduced reject rates, and enhanced sorting precision. AI-assisted kinetic modeling, catalyst performance prediction, and enzyme design tools have improved process intensification for pyrolysis, solvolysis, depolymerization, and biocatalysis. Life Cycle Assessment (LCA)–integrated datasets reveal that environmental benefits depend strongly on functional-unit selection, energy decarbonization, and substitution factors rather than mass-based comparisons alone. Case studies across Europe, Latin America, and Asia show that digital traceability, Extended Producer Responsibility (EPR), and full-system costing are pivotal to robust circular outcomes. Upcycling strategies increasingly generate high-value materials and composites, supported by digital twins and surrogate models. Collectively, evidence indicates that AI moves from supportive instrumentation to a structural enabler of transparency, performance assurance, and predictive environmental planning. The convergence of AI-based design, standardized LCA frameworks, and inclusive governance emerges as a necessary foundation for scaling circular plastic systems sustainably.