Electrospun Nanofibrous Architectures for Food Packaging: A Review of Pore Engineering, Permeability Control, and Sustainability
Mahmoud Farghaly, Chen Huang, Ziyan Li, Chenchao Zhang, Huawei Li, Shuaichao Du, Xuesen Zeng, Xueliang XiaoAbstract
Fresh food spoils rapidly when packaging cannot balance gas exchange with moisture and microbial barriers. Conventional plastic films generally fall short of these combined demands. This review examines electrospun nanofibrous architectures (NFAs) as next-generation sustainable food packaging platforms. Beyond covering fabrication fundamentals, we explore how processing parameters, polymer crystallinity, and pore structure together control oxygen permeability (OP) and water vapor permeability (WVP), which determine preservation performance. Building on these insights, we highlight the emerging role of machine learning (ML) in predicting fiber morphology and optimizing pore design, offering a data-driven route to minimize empirical trial-and-error in material design. However, its current applicability is constrained by data scarcity and generalizability challenges. We also identify greener fabrication routes, including solvent-free melt electrospinning and bio-based, renewable solvents, as key steps toward sustainability. The review further covers smart packaging systems, in which biopolymer blends act simultaneously as integrated freshness sensors that detect spoilage indicators, such as total volatile basic nitrogen (TVB-N), and as carriers for the controlled release of antimicrobial and antioxidant agents. In addition, a techno-economic assessment shows that scaling up the production of electrospun NFAs is commercially attractive for specific high-value food sectors, particularly with high-throughput configurations. Collectively, electrospun NFAs represent a scalable, tunable platform for active food preservation. Future progress will rely on standardizing safety and migration protocols, developing closed-loop solvent recovery systems, and building the large, shared datasets needed to make ML models genuinely transferable across laboratories and polymer systems.