Surfaces can be characterized by three fractal parameters: root-mean-square (RMS) roughness, roughness exponent and lateral correlation length. Little work has been done on correlating these parameters with adhesion. In this study, we simulated the adhesion between an atomic force microscope (AFM) tip and sample surfaces with varying fractal parameters. And experimentally, we performed adhesion measurements on polycrystalline silicon sidewalls of varying topography using AFM. Both the simulations and the experimental data support the conclusion that surface roughness is a significant predictor of adhesion, with the adhesion dropping by more than an order of magnitude for a roughness change from 1 to 10 nm. For the roughness exponent, the simulations reveal a 20% decrease in adhesion as the roughness exponent varies from 0 to 1. The scatter in the experimental data was large since the range of the roughness exponent varied only from 0.85 to 0.99. For the lateral correlation length, the experiment showed a wide range of adhesion values for smaller correlation lengths and low adhesion for larger correlation lengths and we are still investigating the theoretical basis of this observation. Although much work is still needed, the work presented here should advance the fundamental understanding of the role of nanoscale fractal roughness in adhesion and might enable better design of future nanoscale devices. (C) Koninklijke Brill NV, Leiden, 2010