Decoupling conservative and dissipative forces in frequency modulation atomic force microscopy

Experiments and theoretical calculations of conservative forces measured by frequency modulation atomic force microscopy (FM-AFM) in vacuum are generally in reasonable agreement. This contrasts with dissipative forces, where experiment and theory often disagree by several orders of magnitude. These discrepancies have repeatedly been attributed to instrumental artifacts, the cause of which remains elusive. We demonstrate that the frequency response of the piezoacoustic cantilever excitation system, traditionally assumed flat, can actually lead to surprisingly large apparent damping by the coupling of the frequency shift to the drive-amplitude signal, typically referred to as the "dissipation" signal. Our theory predicts large quantitative and qualitative variability observed in dissipation spectroscopy experiments, contrast inversion at step edges and in atomic-scale dissipation imaging, as well as changes in the power-law relationship between the drive signal and bias voltage in dissipation spectroscopy. The magnitude of apparent damping can escalate by more than an order of magnitude at cryogenic temperatures. We present a simple nondestructive method for correcting this source of apparent damping, which will allow dissipation measurements to be reliably and quantitatively compared to theoretical models.