The remarkable optoelectronic properties of metal halide perovskites (MHPs) have established them as highly promising photovoltaic absorber materials, propelling the rapid advancement of perovskite solar cells (PSCs) that outperform many traditional alternatives in terms of power conversion efficiency (PCE). However, despite their advancements, PSC devices encounter significant non-radiative recombination losses, encompassing trap-assisted (Shockley–Read–Hall) recombination in bulk and interfaces of PSCs, which restricts their open-circuit voltage (V_OC) and overall PCE, dragging it below the Shockley–Queisser (SQ) limit. The ongoing debate regarding the role of grain boundary (GB) recombination, whether it primarily manifests as bulk or surface recombination, has spurred extensive research aimed at elucidating these mechanisms. This review provides a critical comprehensive analysis of the thermodynamic correlations related to V_OC losses, bridging the gap between the theoretical SQ limit and practical device performance. Subsequently, it delves into recent findings that aim to decipher the multifaced nature and origin of radiative and non-radiative recombination-induced losses within the device stack, assessing their impacts on overall performance. Furthermore, this review emphasizes the application of advanced machine learning techniques to discern dominant recombination mechanisms in PSCs. Finally, it summarizes the notable advanced strategies to mitigate undesirable non-radiative recombination losses, which pave the way to the thermodynamic efficiency limit.

The interaction between CO and Pb²⁺ regulated grain growth, enhanced crystallinity that effectively passivated non-radiative recombination centres and CF₃ group stabilizes the perovskite structure by interacting with FA and Pb cationic defects.

The fabrication of sustainable packaging films based on chitosan/starch (CTS/Starch) blends, reinforced with Chitosan Nanoparticles (CNPs), was achieved via the casting blend technique. This research explored the impact of varying CNPs loading on critical physicochemical properties, including water vapor permeation (WVP), thermal stability, and mechanical strength. To elucidate the structural and chemical complexities of the blend films, surface morphology was investigated via Scanning Electron Microscopy (SEM), internal architecture was visualized using Transmission Electron Microscopy (TEM), and molecular interactions were probed through Fourier Transform Infrared (FTIR) spectroscopy. The reduction in WVP from 6.18 ± 0.54 to 5.38 ± 0.93 g.m⁻¹.s⁻¹.pa⁻¹, equilibrium moisture content (EMC) from 16.52 ± 1.03% to 12.5 ± 1.05%, and water absorbency (WA) from 340 ± 1.63% to 88.65 ± 1.12% in CTS/Starch blend films demonstrated loaded with (0–8 wt%) CNPs loading. Concurrently, films with 2–8 wt% CNP loading exhibited an increase in opacity from 2.38 ± 1.01 mm⁻¹ to 4.83 ± 0.83 mm⁻¹, accompanied by a decrease in transmittance from 89.20 ± 0.50% to 79.70 ± 1.20%. These findings collectively indicated that the CNP-incorporated chitosan/starch composites offer enhanced ultraviolet light shielding and improved water barrier capabilities compared to the non-reinforced chitosan/starch films, underscoring their promising utility in food and pharmaceutical packaging applications.