How to perform multi lattice kMC calculations in kmcos?

[Ghaem Hasanzadeh]

Dear kmcos developers,

I know kmos was used to perform multi-lattice kmc calculations (please see: https://pubs.acs.org/doi/10.1021/cs501352t), but I don't know how I can use this is the code. Can you please help me with this? 

Best wishes,

Ghaem

[Aditya "Ashi" Savara]

While I was not involved in that study, I can tell you a general approach that will work.

(1) Recognize that the "KMC lattice" defined in the beginning is basically just for visualization. Reactants and products can appear and disappear "anywhere"  you define a site (even overlapping radius). Only one species may occupy a given site, however. KMCOS models processes as you define them.

Now consider that there are a few cases: 

(a) one case is you have a crystal lattice that will transforms into another crystal lattice during the course of simulation. In this case, you define *all* sites at the beginning. With positions that are defined relative to your KMC lattice.  Then, during the course of reaction, your reactants will simply proceed as intended.  You could have some kind of "filler" species on the sites corresponding to the second chemical crystal lattice, and could remove those species to make the sites "open" as the surface changes, and you would put filler species on the sites corresponding to the first chemical crystal lattice as those sites disappear from chemical conversion. There is more than one way to do this. The key point is that your KMC lattice defines the relative coordinates, but it does not define the sites. You define where in geometry space sites will exist, and then you define the processes that can occur, which can include making them active or inactive. You could even make a species called "inactive" and put it on a site to make that site inactive, until some process takes that species off. Since the site would not be shown as "open", it would effectively be inactive.

(b) the other case is that you have two overlapping lattices at the surface at the same time.  This is perhaps even more easy, because it's the same as case a except you start without any filler species. All sites are open and active to begin with. 

(c) supposing there is a case where you want the left half of the surface to be one chemical lattice, and the right half to be another chemical lattice. This just requires making one large unit cell, and then recognizing that the KMC lattice is just for visualization purposes, you'd define your chemical lattices as desired across your one unit cell. So you could have one half of the unit cell a certain way and the other half of the unit cell another way.

Such KMC models do take time to create, but that is how one would create it.