Experimental setup affects the particle formation rate and its slope d(log J)/d(log C)

We have simulated the formation of sulfuric acid-dimethylamine clusters in different experimental setups and in the atmosphere. We demonstrate that the cluster formation rate may be different in flow tube experiments than in chamber experiments due to the depletion of some trace compound. We also show that applying the nucleation theorem in a situation where all its assumptions do not hold may lead to erroneous results: the slope d(log J)/d(log C) of the formation rate versus the sulfuric acid concentration is in many conditions 4, although each step of the formation process decreases the Gibbs energy of the system and there is no critical cluster.