I was working on a problem asking 'in an oscillation of a bond in a single O2 molecule with a pair-wise Lennard-Jones (LJ) potential how does the total energy $E$ vs $t$ change?' it goes on to provide values of constants like $σ$ the but initial position/velocity are not stated. Is it possible to use the Verlet algorithm without those initial conditions (make assumptions)? Or is there a way to obtain those?
-
2$\begingroup$ You might consider to ask this on Matter Modeling. $\endgroup$– Tobias FünkeCommented Feb 2, 2023 at 20:54
-
1$\begingroup$ @TobiasFünke It would be a sad day if a question like this could not be answered in a site about physics. This idea that every question mentioning numerical methods is not Physics is quite dangerous for this community. $\endgroup$– GiorgioP-DoomsdayClockIsAt-90Commented Feb 2, 2023 at 21:41
-
$\begingroup$ I am confused.. is there a practical setting where you know the initial conditions of a molecular dynamics simulation? Typically you choose some random initial condition (provided that you fulfill the constraints of the problem at hand) $\endgroup$– QuilloCommented Feb 2, 2023 at 22:07
-
$\begingroup$ @GiorgioP-DoomsdayClockIsAt-90 I don't say it is not physics. I mentioned the other SE site because there are many experts regarding numerical simulations and more in the field of chemistry, material science and condensed matter... If someone asks a math question related to physics which could be answered better (IMHO) on math SE, I'd also recommend to ask the question there. I did not vote to close and, to reiterate, neither said it is not physics or anything. $\endgroup$– Tobias FünkeCommented Feb 3, 2023 at 7:52
1 Answer
The O$_2$ molecule has only one vibrational normal mode. Whatever is a suitable form for a classical interaction potential between the two oxygens, the system behaves as a simple harmonic oscillator in the regime of small oscillations around the equilibrium position.
To check numerically how observable quantities vary with time, the choice of the time origin is irrelevant. Therefore you can choose whatever initial condition you like, provided the system remains in the linear regime of the force.
However, I would say that it is probably useless to perform a numerical simulation to find the results unless the influence of the numerical algorithm is the real issue.