Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How do we model metal nanoparticles for MD simulations?

  1. Jan 18, 2016 #1
    I am trying to model metal NPs capped with organic agents to simulate their phase transfer behavior from aqueous to organic. Not doing computational chemistry courses has had me to disadvantage. Would love some initiation or a jump start primer for the same.

    Looking forward for cues to get me working. Thanks.
  2. jcsd
  3. Jan 23, 2016 #2
    Thanks for the post! This is an automated courtesy bump. Sorry you aren't generating responses at the moment. Do you have any further information, come to any new conclusions or is it possible to reword the post?
  4. Jan 25, 2016 #3
    Since you're working with metals, you might have to employ Density Functional Theory, and even that will be difficult! Depending on the size of your system and your computing resources, you have a few different options. Whether you decide to do that or not, here's a good introduction to the theory behind it (go to Youtube for the full playlist):

    I don't know if molecular dynamics (MD) software will do what you need, from the limited knowledge I have it's usually completely organic systems modeled with molecular dynamics. But if you end up needing to do quantum chemistry, a couple good places to start are the program ORCA (which has the best documentation I've ever seen), or the program Firefly (which has less experimental algorithms than ORCA, and may be faster for your purposes). If you need something to create your nanoparticles with, I've used Avogadro and Gabedit before, and have liked them both (I think Gabedit has an ORCA input file maker, too, which is nice).

    Since I don't know your topic well enough, I can't promise this is right, but if you're looking for a reaction or change of phase, using the softwares mentioned above you may want to do transition state searches.

    Good luck!
  5. Jan 26, 2016 #4
    Thanks ModestyKing for the insight. I consulted my university professor regarding the same, for which we came down to simplifying the problem so as to reduce computational requirements. We intend to employ 'hard sphere'/'billiard ball' model for the metal nanoparticles - since we are not really studying the interaction of metal nanoparticle with medium, but the stablizing/capping agent on it - however, when I began calculating and generating the biphasic medium using small molecule building software as packmol and VMD, I realized that the number of molecules is too large even if I want to simulate single nanoparticle of diameter 10 nm ~ 12nm. I am definitely going to look into ORCA and Firefly though I am pretty certain we are not looking into quantum chemistry perspective at this stage. We will mostly be looking at classical model interpretation of vdW fields.

    I did a fairly exhausting literature survey on the topic and have finally come down to a simplified system to start the work with, for which I need some direction in creating a 'patchy hard sphere' for my nanoparticle. Would there be any tool that you know of that can make hard sphere model, because molecular builders such as VMDs Molefacture doesn't seem to have such features.

    Thanks again. Meanwhile I will be looking into Avogadro and Gabedit.
  6. Jan 26, 2016 #5
    I can't remember if Avogadro and Gabedit have a nanoparticle builder, but one has a nanotube builder, so it stands to reason they may also do nanoparticles. One professor of mine simulated a buckyball (C60 nanoparticle) using quantum mechanically driven molecular dynamics with the software TeraChem, which uses GPUs to severely speed up simulations, and it works with VMD to illustrate the particles.

    However, if you are not using that (it's an expensive commercial software), it would be nice to know - do you mean you want to do molecular dynamics calculations (hard spheres and "springs" representing bonds, with VdW and electrostatic interactions), or you are going to make your own program? If it's MD calculations, the main tool for that is NAMD, but AMBER is also helpful. I don't know enough about MD yet (starting research with it now), so I might not be able to help more. Good luck!
  7. Jan 26, 2016 #6
    I don't thinks this work lies completely in the purview of Molecular Dynamics, but since the experiment needs to solve for energetics of equilibration of particles in a biphasic medium I don't know what other technique can help me out. As I saw Avogadro too, it is pretty same as Molfacture that comes with VMD. The current state of my problem is not of making nanoparticle like a molecule, but just as ball. If so to say a constant void of dimension of nanoparticle should work too if I can program that void surface to have that property as that of the surface of nanoparticle+capping agent.

    Yes, I am inclined on using NAMD only, and I myself have started this research a couple weeks back now. I am not sure about the AMBER package, it is also a commercial licence, while I am trying my best to use FOSS for the work unless its dearly required otherwise. I'll post any update or progress I make. Thanks again.

    Good Luck to you too.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: How do we model metal nanoparticles for MD simulations?
  1. How do metals work (Replies: 1)