Fw: Particle size and time scale

[Ahmad Reza Motezakker]

---

From: Ahmad Reza Motezakker
Sent: Tuesday, March 2, 2021 11:40 AM
To: Peter Košovan
Subject: Re: Particle size and time scale

 

Dear Peter,

Thank you very much for the reply.

Exactly as you mentioned, I compare MSD and diffusion of the rod shape particle with the famous theories of stiff polymer physics in books and papers (for example, Doi and Edwards, Odijk and ...). I have checked most of the papers which espressoMD (CG+LBM) has been used in them and to my knowledge, they have not mentioned the way that they have defined the time scale.

I really be thankful if you give me some hints on considering viscosity in calculating time scale or any paper. Thank you.

Best regards,

Ahmad Reza

---

From: Peter Košovan <peter.kosovan@natur.cuni.cz>
Sent: Tuesday, March 2, 2021 11:22:26 AM
To: Ahmad Reza Motezakker
Subject: Re: Particle size and time scale

 

Dear Ahmad,

Please, be aware that in implicit-solvent simulations you cannot simply calculate the time scale by calculating the unit of time from the simulation units of mass, energy and length. The reason is that the implicit solvent (or LBM coupling in your case) determines the viscosity of the solvent, which does not enter in the calculation of the time unit. Therefore, dynamics in CG simulations is accelerated. In other words, your actual time unit is several orders of magnitude greater than what you get from simple conversion of units.

You can calculate your dynamical time scales by measuring the diffusion coefficient of a single particle with the same parameters.

It might be also helpful if you specify the theory to which you are comparing your results.

With regards,

Peter

On Tue, Mar 2, 2021 at 11:13 AM Ahmad Reza Motezakker <armot@kth.se> wrote:

Dear EspressoMD users,

I have a polymer system coupled with LBM in espressomd. The problem is that particle dynamic is around 10^3 or 10^4 times faster than predicted results from predicted theory and previous resulted results. I have considered each sigma σ equal to 10 nm in reality or SI units. In other words, for a particle with length of 150 nm , I have considered 15 beads with σ =1 . 

Following that, the box lengh of 450 nm is set to 45 in espresso unit in the same way. All the rest of parameters have been calculated according to the mass of particle and the length unit of 10 nm.

Calculating time unit, I come up with time unit of [t]=9*10^-9 seconds. When it comes to running the simulation, I should set time step to maximum 0.001 (which will be equal to pico seconds in SI) to run, for bigger ones it crashes. And the result which I get from the simulation, it shows that the dynamic is so fast compare to theory. 

I will be really thankful if you share with me your thoughts and show me what I am doing wrong here. Thank you all in advance.

P.S. , I log the data each 1000 step which will be around nano seconds roughly.

Best regards,

Ahmad Reza 

--

Dr. Peter Košovan

Department of Physical and Macromolecular Chemistry

Faculty of Science, Charles University in Prague, Czech Republic

Katedra fyzikální a makromolekulární chemie

Přírodovědecká fakulta Univerzity Karlovy v Praze

www.natur.cuni.cz/chemistry/fyzchem/
Tel. +420221951029

Fax +420224919752