Why do you need MD simulations to perform binding energy calculations on metal surfaces? Static DFT calculations are usually enough.. I would try with these and comparing the trends with the experiments.

What kind of MD did you perform?    Why?

Did you try to relax the system, and them compare the energies?

Relaxing the systems to their respective local minima puts them on a common denominator, making the comparison of energies logical.

Monte Carlo simulation adsorption locator implemented in Material studio, I relaxed the systems, but all halo compounds show higher bindind energy compared to their anologues. This contradicts with some experimental results. Although DFT calculations using Dmol3 give consistant results with experimental to the greatest extend. But molecular dynamics has some discrepancy with experimental results.

"Although DFT calculations using Dmol3 give consistant results with experimental to the greatest extend. But molecular dynamics has some discrepancy with experimental results."  I don't get what you mean by this.

What do you mean by the first statement?

And if DFT is giving consistent results, what do you mean MD is showing discrepancy?

What dynamics did you carry out and how did you compare the energies?

The way I see it, you only need to relax/optimize the coordinates of the system at DFT level of theory. And then compare the relaxed systems by calculating their adsorption energies. Also, how do you calculate the adsorption energies?

We study the inhibition efficiency of some organic compounds towards carbon steel in aqueous solution. I used different experimental techniques electrochemical such as EIS, EFM, and polarization. From these experiential results, the arrangements of these compounds based on the activities has been determined. Now, I know the arrangement of these compounds based on the its inhibitory activity. For the theoretical part of the study, DFT calculation has to confirm the above mentioned sequence of the compounds from the calculated parameters such as the gap energy and LUMO, HOMO energies., from which a correlation has been established between the experimental results and DFT descriptors. At this level everything is OK and non halo-compounds show less activity than its counterparts. When  Molecular dynamic simulation were performed to simulate the system using standard procedure for Forcite (Quench, Dynamics,  Montecarlo simulation).  There is discrepancy and halo-compounds show better efficiency based on the adsorption energy calculated between metals and these compounds. Halo-compounds show more negative adsorption energy (more positive binding energy ) with metal. My question "Is there a trend for halo-comounds to show more negative adsorption for Molecular Dynamic simulation? Or Does the discrepancy come from other factors relating to the molecular dynamic calculation level that some factors are approximated?

Thank you in advance

My problem is what you are referring to as "standard procedure".

1. What is that?

Info provided by you does not answer questions raised by me in my last reply.

2. If you have relaxed the system, what dynamics did you carry out, and why?

    I do not understand why you need to carry out any MD, after the system is relaxed.

    And what energy, from the dynamics, do you choose? And how can one justify this

    choice?

    MD/relaxation performed at any lower levels of theory, (than DFT here) will not be comparable to that performed by DFT itself. I mean that MD/relaxation theory level has to be the same for comparison. Not just the theory, but other parameters, like the basis, the energy cutoffs, the K-points, etc. must also be the same.

3. How did you calculate the adsorption energies?