Hi everybody,

I'm trying to get the evolution of the storage and loss modulus of polymer nanocomposites. For the moment, I'm using the Fourier transform of the stress autocorrelation function (Green-Kubo relations). I run NPT simulations with Nose-Hoover thermostat and Parrinello barostat and compute the results with Matlab. However, it requires really long simulation to get non-aberrant results and they are either not really realistic or unphysical. Hence different questions:
* Do you know any other methods which could be more proficient?
* If I keep this method (Fourier transform of stress acf), how the dynamic simulation should be run to get more precise results?
- With or without stress?
- What kind of stress? Ideally, I would like to apply a sinusoidal stress but the period wouldn't be too small?
- Which barostat should be used? (Souza Martins? Parrinello?)

I'm ok with writing scripts so feel free to propose whatever you think of!

I thank you very much by advance for your time!!

Best :)