Modeling Energy Transfer Efficiency in Collision-Induced Dissociation and Unfolding Using Molecular Dynamics Simulations

Collisional activation is available ubiquitously in commercial mass spectrometers, but, due to differences in instrumentation, it can be difficult to "translate" instrument settings from one instrument in a collisional activation experiment to another instrument to achieve quantitatively predictable results. With the goal of a more universal understanding and translatability of collision-induced dissociation and unfolding (CID/U) experiments, we recently introduced software ("IonSPA") for simulating ion heating, cooling, and kinetic energy damping during collisional activation. A key parameter in IonSPA is the "pseudo-atom mass", which controls the efficiency of energy transfer between an ion's vibrational modes and the gas particle on a collision-by-collision basis. Here, we use molecular dynamic (MD) simulations to study energy transfer efficiency for native charge states of three different protein ions. Energy transfer efficiency is found to depend primarily on the collision gas type, relative speed of the collision, and total vibrational energy of the ion. Efficiency trends toward unity (completely inelastic and thermalizing) at zero collision speed, decreases to a minimum at a moderate collision speed, and increases again toward highly inelastic at high collision speeds. Dependence on collision geometry and the role of the centrifugal barrier and interaction potential are also explored. A pseudo-atom mass function based on gas type, ion vibrational temperature, and collision speed is derived from these MD simulations and implemented into IonSPA to calculate thermochemical barrier values in CID and CIU.