Think about a box comprised of people from a conservative cultural background that has been undergoing some changes due to the advent of external stimuli, such as, but not limited to, a foreign norm. Well, those people will start disseminating their own opinion on this new culture; some will accept and some not. This can hamper personal relationships among people within the community. For example, many will move from one social group or circle to another only to justify their stance. Don’t be so skeptical, it’s just a metaphor. Well, in large, visible scales, we can visualize these changes and know the social positions of individuals.
But, think for a moment. These similar movements happen in non-visible scales around us. Well, not exactly similar, but similar enough to compare. For instance, we can feel the evaporation of our sweat in airflow wherein atoms detach from one another, although we can’t see this with our naked eyes. Surely, atoms and molecules are everywhere. They also move, transfer, rotate, spin, and even, detach from one another. They too respond to external stimuli, i.e. strain, vibration, friction.
For a box of atoms under a stimulus, such as elongation, these atoms will tend to move in different directions, just like in the aforementioned norm-related case. But, we don’t know the exact position where these atoms will move to. One atom can move to the right, or the left, or any other possible direction. These movements depend on other atoms’ movements. Like, how far one atom is from other atoms so that attraction and repulsion can come into play. So, to understand where these atoms will end up after a certain time, we need to perform numerous simulations considering every possible directional movement, as in molecular dynamics simulations in LAMMPS, to calculate all possible movements and compare them to find the optimal position. This process is done for each atom. After considering all the possible locations, the computer will move each atom to a position of maximum stability, for the time being. This is just an example of what computer simulations do to these atoms. You can very easily work this out on your own by solving hundreds of equations (depending on your system size) and then comparing the output at each time step with the optimum values whether the output aligns with the desired temperature, energy, or any other properties in consideration. In future blogs, other topics related to molecular dynamics simulation will be dissected.
To sum up, it can be said that to know the optimal result out of many possible results, which are dependent on several factors, simulations are the perfect samaritan. You set the parameters, and the computer does the iterative, boring work for you.