QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: zhySarah on November 10, 2015, 21:16

Title: is there a way to simulate a cluster of quantum dots?
Post by: zhySarah on November 10, 2015, 21:16
Hi,

I am thinking of using a molecule (Cd2S2) to simulate a cluster of CdS Quantum dots, but not sure how to build one. Any comments or suggestions?  :)

PS: Is there a tutorial teaching how to adsorb a molecule to a nanotube (or any other configuration?)

Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Umberto Martinez on November 11, 2015, 12:53
You can start from this tutorial: http://docs.quantumwise.com/tutorials/benzene_au111.html
and also go through these one to familiarize with the Builder:
http://docs.quantumwise.com/tutorials/vnl_builder.html
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on September 22, 2016, 19:14
After building quantum dots, is it possible to use them in Synopsys tools for TCAD simulations?
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Anders Blom on September 22, 2016, 22:20
No, that is not possible. TCAD has no atomic resolution for things like this.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on September 28, 2016, 20:50
Sir,
I want to analyze the variation in bandgap with the change in number of atoms of Quantum dots. For this, I am using silicon Quantum dots. I have noticed the change in bandstructure but not in bandgap when considering the bandgap as minimum energy difference between conduction and valence band.
Attached are the bandstructures that I have obtained by varying the number of atoms in 3D.

I have used the repeat option to repeat the atoms in the three directions. It does not allow the control on number of atoms. Could you any other way by which the atoms number can be increased in the structure.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Anders Blom on September 28, 2016, 21:26
Seems like you are just computing a larger and larger periodic supercell of Si, so that will not change anything (just take longer time). A cluster will have vacuum around it, and you can build one using the  Wulff construction tool in the Builder.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on September 28, 2016, 21:36
Thank you sir!
So I should build a cluster using Wulff construction tool for analyzing the variation in bandgap with the change in number of atoms of Quantum dots?
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Anders Blom on September 28, 2016, 21:45
You can try that, yes, although real quantum dots are rarely as regular in shape as the Wulff construction gives; more typically they are pyramidicial or similar.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on September 30, 2016, 19:24
Sir,
I tried the Wulff construction tool for analyzing the variation in bandgap. Unfortunately, after having Wulff cluster the bandstructure option in analysis does not show up. So, how can I find bandgap after having Wulff cluster.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Anders Blom on October 1, 2016, 00:54
Only periodic structure have a band structure... You can use the HOMO/LUMO gap instead for a cluster.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on October 6, 2016, 18:30
Sir,
The Wulff construction tool does not provide an option to increase the atoms in  a discrete manner. I noticed that number of atoms in a cluster is dependent on the radius of the cluster. Is there any way I can set up the number of atoms as my wish in a cluster.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Ulrik G. Vej-Hansen on October 6, 2016, 20:26
Just to clarify, do you want to add single atoms to an existing Wulff construction, or would you like to add whole layers?

In both cases, the resulting structures are not really Wulff constructions anymore, though they can of course be perfectly valid nanoparticle structures.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on October 6, 2016, 20:31
I want to vary the number of atoms by certain number (e.g. increase 2 atoms). In Wulff construction, number increases with radius. There is no such control on number of atoms.

Let say I want to create a structure with 15 atoms. How will I do it?
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Anders Blom on October 7, 2016, 01:57
That's more a question for you than for us.  What does a nanocluster with 15 atoms look like? I certainly don't know and wouldn't know without doing the simulation. I guess there are many answers, but you could just put 15 atoms in a cube relatively close to each other (so they bond reasonably but also don't repel dramatically) and optimize the geometry with DFT. Or make a Wulff construction that is close to 15, then remove some atoms (somewhat symmetrically) to make it 15, and optimize. The other natural question I would have is, why 15? What's magical about that number? Have you or someone else synthesized clusters of that exact atom count in the lab? If so, or otherwise, is there experimental evidence of the shape?
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: sumit34 on October 7, 2016, 07:44
I had put 15 as a random number. 
Actually, I want to obtain the bandgap (HOMO-LUMO gap) variation with respect to the number of atoms. So, I need to vary the number of atoms in a definite proportion so that appropriate graph is obtained.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Ulrik G. Vej-Hansen on October 7, 2016, 10:13
As Anders says, predicting nanoparticle shapes with certainty is quite complicated. See for example: http://www.sciencedirect.com/science/article/pii/S0021951711002673

If you want to investigate the HOMO-LUMO gap for a certain geometry and vary the number of atoms, I would make a Wulff construction and then remove the highest-coordinated atoms. This is no guarantee that your shapes will have a large weight in the actual distribution of shapes, but it should at least provide you with fairly low-energy shapes.
Title: Re: is there a way to simulate a cluster of quantum dots?
Post by: Anders Blom on October 7, 2016, 20:10
Probably it's more relevant to look at various properties as function of size (radius) rather than the exact number of atoms.