A new two-dimensional aerosol bin scheme, which resolves both aerosol size and black carbon (BC) mixing state for BC aging processes (e.g., condensation and coagulation) with 12 size × 10 mixing state bins, has been developed and implemented into the WRF-chem model (MS-resolved WRF-chem). The mixing state of BC simulated by this model is compared with direct measurements over the East Asian region in spring 2009. Model simulations generally reproduce the observed features of the BC mixing state, such as the size-dependent number fractions of BC-containing and BC-free particles and the coating thickness of BC-containing particles. This result shows that the model can simulate realistic BC mixing states in the atmosphere if condensation and coagulation processes are calculated explicitly with the detailed treatment of BC mixing state. Sensitivity simulations show that the condensation process is dominant for the growth of thinly coated BC particles, while the coagulation process is necessary to produce thickly coated BC particles. Off-line optical and radiative calculations assuming an average mixing state for each size bin show that the domain- and period-averaged absorption coefficient and heating rate by aerosols are overestimated by 30–40% in the boundary layer, compared with a benchmark simulation with the detailed treatment of mixing state. The absolute value of aerosol radiative forcing is also overestimated (10%, 3 W m–2) at the surface. However, these overestimations are reduced considerably when all the parameters (including mass and number concentration) are calculated with the simple treatment of mixing state. This is because the overestimation of radiative parameters due to higher absorption efficiency (compared with the benchmark simulation) is largely canceled by the underestimation of BC concentrations due to efficient wet removal processes. The overall errors in radiative forcing can be much smaller because of this cancellation, but for the wrong reasons.