General comments:

The present work describes molecular Dynamics (MD) simulations of collisions of molecules / clusters and ions and the determination of collision rate constants from these simulations as well as from theory, using analytical equations with different approximations. The study increases our understanding of the validity range and limitations of these approximations.

Specific comments:

Eq 11: to my understanding, β_L is the rate constant for the collision of an ion with a polarizable molecule and β_SC that for an ion with a polarizable dipolar molecule. Hence, I would expect that K should approach one for μ_d (and hence x) approaching 0, but it seems to be 0.025. Please comment on this behaviour.

I find the description of the equilibration process difficult to follow (lines 222-233). There is a statement “The collision partners were first separately equilibrated for 50 ps using a Langevin thermostat with a damping factor of 0.1 ps. During the equilibration, both the center-of-mass motion of each collision partner and the angular momentum of the total system were removed … Both collision partners were then given a velocity along the x-direction” If they are equilibrated separately (i.e. in separate simulations), both fragments should not have any rotational angular momentum. I have the impression rather “together, but separated with a large distance” is meant. If the latter is the case, have you checked the correct distribution of rotational angular momenta? Only a check of the energy distribution is mentioned. This this is a scalar and the angular momentum is a vector, therefore this may not be sufficient.

Line 261: “The center-of-mass distance criterion for a successful collision was determined for each system by taking the distance at which the value of the PMF was 5kBT (∼ 0.13 eV at 300 K) higher than its minimum,” There are two such distances, right and left from the minimum. Which one have you chosen?

Line 270, Discarding and adding new trajectories in case of dissociation: Is that the correct procedure? To me, it would seem most appropriate to define 3 outcomes of the encounters: no collision, association, and dissociation of an existing dimer. Then one would use the total number of trajectories in the denominator for each of the rate calculations. This would seem in line with what is done in Master Equation calculations for barrierless reactions. Please comment on the justification for your approach.   

Line 279: What do you mean by “small oscillations in the interaction energy”?

Can you specify statistical uncertainties, for the MD results, e.g. in table 2?

Technical comment:

Figs 4 and 5: The Circles to not show up on my Adobe Acrobat on android, but they do on windows.

This manuscript explores the collision dynamics of eight ion-dipole systems using potential of mean force (PMF) calculations and molecular dynamics (MD) simulations. Collision probability maps are obtained by MD to determine the dynamic collision cross sections. The collision rate coefficient results obtained by PFM and MD are compared to the classic Su and Chesnavich and the Langevin-Gioumousis-Stevenson models. 

The manuscript is in general well written and well organized and manages to provide an understanding of the conditions that PMF calculations and central field models can be used reliably for the determination of the collision rate coefficient. 

To further improve the manuscript, I suggest adding a comment on how long the clusters are traced after collision, i.e. what is the lifetime of the clusters shown here. In addition, the value of the probability, P(v,b), at which the dynamic collision cross sections of Fig. 4(e-h) and Fig. 5(e-h) were obtained by MD, should be provided.