The dynamic shape factor and the exponents, eta and D(fm), which characterize the power law dependence of friction coefficient on the number of primary spheres and the mass on the mobility diameter, have been determined for silver agglomerates using the differential mobility-aerosol particle mass (DMA-APM) analyzer method. This method provides characterization of nearly monodisperse agglomerates and is able to analyze thousands of particles over a 10 min period. A quantitative uncertainty analysis finds that the calibration of the APM is the major source of uncertainty and that the combined uncertainties are about 6-7% for the dynamic shape factor and about 3% for the exponents eta and D(fm). The dynamic shape factor obtained based on free molecular analysis is larger than the measured results. The observed decrease in q by about 15% with increasing agglomerate size compared to almost constant values for the model predictions suggests a flow interaction between the primary particles not included in the models which are based on free molecular dynamics. An empirical equation is given for the N dependence of the ratio of the measured friction coefficient to a free molecular expression based on a computer simulation. Model predictions indicate that eta is independent of agglomerate size while D(fm) is sensitive to agglomerate size. Experimentally, it appears the opposite is true: the dependence of eta on particle size is greater than for D(fm). The near constancy for the measured D(fm) results from the decreasing value in eta being compensated by the slip correction term in the expression relating d(m) to the friction coefficient. (C) 2009 Elsevier Ltd. All rights reserved.