Filtering of NaCl, CaCl2, (NH4)2SO4 and NiSO4 aerosol particles 7-20 nm in diameter by a stainless steel grid was studied in order to find out if there is perfect sticking or partial rebound. Our experiment used particles from a spray-drying process, the majority of which were electrically neutral. Penetration through the grid was measured by comparing the concentration downstream of the grid with the upstream concentration under otherwise identical conditions. Size selection was done with a scanning mobility particle sizer (SMPS). Filter penetration P as function of the particle diameter dp was expressed by - ln (P (dp)) = C dp- x. The values of x determined were smaller than the theoretical value of 1.29, indicating enhanced penetration of small particles and deviation from the classical filtration model. Because of possible systematic errors in the size selection, we focus on the differences of x from material to material, which indicate different sticking probabilities. We apply a statistical test, which yields a 90% confidence level for the result. There is a sticking probability of <100% at least for NaCl particles and even more so for NiSO4. This result is in contrast to former findings using metal and/or charged particles, and we speculate that the discrepancy is due to the smaller Hamaker constant of salts and that particle charge is important for the sticking probability.