Brake wear particles (BWPs) represent a major source of non-exhaust particulate matter from road traffic, contributing substantially to human exposure, particularly in urban environments. While traditionally associated with coarse and fine fractions, mounting evidence shows that brake systems emit large quantities of ultrafine particles (UFPs; <100 nm), which dominate number concentrations despite contributing little to mass. This paper synthesizes current knowledge on BWP formation mechanisms, physicochemical characteristics, environmental behavior, and toxicological effects, with a specific emphasis on UFPs. Mechanical friction and high-temperature degradation of pad and disc materials generate nanoscale primary particles that rapidly agglomerate yet retain ultrafine structural features. Reported real-world and laboratory number concentrations commonly range from 10³ to over 10⁶ particles/cm³, with diameters between 10 and 100 nm, rising sharply during intensive braking. Toxicological studies consistently demonstrate that UFP-rich and metal-laden BWPs, particularly those containing Fe, Cu, Mn, Cd, and Sb compounds, induce oxidative stress, inflammation, mitochondrial dysfunction, genotoxicity, and epithelial barrier disruption in human lung and immune cells. Ecotoxicological studies further reveal adverse impacts across aquatic organisms, plants, soil invertebrates, and mammals, with evidence of environmental persistence and food-chain transfer. Despite these findings, current regulatory frameworks address only the mass of particulate matter from brakes and omit UFP number-based limits, leaving a major gap in emission control.

Brake wear is an increasingly important source of non-exhaust traffic-related particulate matter, yet the relationship between emissions measured under standardized brake-dynamometer conditions and those observed during complete-vehicle operation remains insufficiently resolved. This study presents a systematic critical analysis and methodological synthesis of laboratory, chassis-dynamometer, and on-road evidence, emphasizing physical system boundaries, brake duty, thermal conditions, sampling efficiency, component conditioning, and measurement reproducibility. Only three experimental programs provided sufficiently comparable paired or quasi-paired road-to-dynamometer data for the principal quantitative comparison; their study-specific emission-factor ratios (EF_ratio) ranged from 0.33 to 1.44 and occurred on both sides of unity. The small, heterogeneous evidence base therefore does not support a pooled meta-analysis or a universal road-to-dynamometer correction factor. For particulate mass, braking energy, friction material, conditioning, and particle recovery are major determinants of cross-method differences. Particle number is particularly temperature-sensitive because thermally activated evaporation and decomposition can trigger nucleation of ultrafine particles. Mobile samplers may therefore both lose particles after generation and perturb the source by altering brake cooling. To structure these effects, the study applies a four-level system-boundary framework and proposes a three-stage validation sequence combining metrological benchmarking, thermal non-interference assessment, and matched road-laboratory testing. The proposed acceptance targets are explicitly presented as provisional research criteria requiring further interlaboratory validation.