Accurate modeling of lithium transport in polycrystalline cathodes is essential for reliable battery simulations. However, reported solid-state diffusion coefficients often vary widely due to particle-scale effects. In this work, we investigate the influence of secondary particle size on reaction kinetics and solidstate transport in NMC622 cathodes by combining particle-size-fractioned model electrode experiments with single-particle electrochemistry (SPE) on the same material batch. To our knowledge, this is the first direct experimental comparison of SPE and electrode-level kinetic and transport measurements on identical active material. By systematically varying the secondary particle size and comparing the results across both experimental scales, a consistent picture of lithium transport in polycrystalline agglomerates
emerges. The exchange current density, when normalized by the electrochemically active mass, shows no systematic dependence on secondary particle size. Likewise, the characteristic diffusion time constant obtained from small-signal measurements is largely independent of agglomerate diameter. These findings
contradict the Monolithic Particle Model and instead support the Cracked Particle Model, in which electrolyte penetration into intergranular cracks decouples effective reaction and diffusion lengths from the external particle dimensions. As a consequence, solid-state diffusion coefficients derived under monolithic assumptions systematically overestimate intrinsic lithium diffusivity, providing an explanation for the wide scatter of reported values in the literature. Mass-normalized exchange current densities and diffusion
time constants therefore represent more robust metrics for comparing polycrystalline active materials and for parameterizing battery models. In this context, implementations of particle size distributions (PSD) should be interpreted with caution under small-signal conditions. Finally, particle-size-resolved model electrodes combined with GITT- and impedance-based analyses offer a robust and experimentally
accessible alternative to SPE techniques and can be transferred to other material classes to systematically investigate particle-scale transport phenomena.