Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Anders Blom

Pages: 1 2 [3] 4 5 ... 334
31
Btw, there are forcefields for Bi2Te3, such as

Qiu, B.; Ruan, X. L. Molecular dynamics simulations of lattice thermal conductivity of bismuth telluride using two-body interatomic potentials Phys. Rev. B 2009, 80, 165203
Huang, B. L.; Kaviany, M. Ab initio and molecular dynamics predictions for electron and phonon transport in bismuth telluride Phys. Rev. B 2008, 77, 125209

But I could not find one for Bi2Se3... Maybe a case for our new machine-learned Moment Tensor Potentials...!

32
There are a few things here and there that could be finetuned, but the most important thing you should fix first is that the automatically detected repetitions for the system is 9x9x3. Since this is a 2D sheet, you really want this to be 9x9x1, or, to run a smaller test calculation, perhaps just 5x5x1 or 7x7x1. You could make the Automatic repetition return 1 in C by including more vacuum, but 28 Å is probably enough.

So, instead, simply set the repetitions manually instead of using the Automatic keyword, and the calculation will use much, much less memory. Since there are 15 atoms in the original cell, currently you are running calculations with 9*9*3*15=3645 atom, which is very heavy, and completely unnecessary, since it should just have been 9*9*15 in the most accurate case. Still 1200 atoms or so, but that is much more manageable. I would however suggest first running 5x5x1 (375 atoms) and then 7x7x1 (735 atoms). If these results are similar, that should be accurate enough.

And finally, you are running on a single machine, so yes, these calculations will take time. If you have any chance to run in parallel on multiple nodes, that will speed things up essentially linearly by the number of additional computers you use.

33
General Questions / Re: Zig-Zag CNT
« on: June 18, 2021, 20:52 »
I always enjoy interacting with our users and one has to remember that we are all new to a subject at some point :) I also take it as feedback that perhaps it could make sense to use the labels zigzag/armchair even in the nanotube builder, as shortcuts. Note that we do have these in the graphene nanoribbon builder!

34
This is not implemented, and I am not even sure it's possible to do so. It seems the common approach is to compute the Born charges without spin-orbit included, even for spin applications, the assumption being that the Born charges themselves are not strongly sensitive to the inclusion of spin or not. Of course that assumption is hard to prove, if you cannot compute the charges with spin-orbit and check, but it is a reasonable one plus the Born charges are extremely to compute accurately anyway, the error estimates are often 10-20%, thus a few extra percent do not really matter. See e.g. https://arxiv.org/pdf/1207.2916 and https://arxiv.org/pdf/0906.3803.pdf

35
General Questions / Re: Zig-Zag CNT
« on: June 17, 2021, 22:06 »
Zigzag nanotubes are of the chiral index type (n,0) and armchair are (n,n).

A friendly suggestion is to catch up on the basic properties on nanotubes via some review article or the many, many resources you can easily find online. A Google search on "zigzag cnt" immediately shows pictures of the property I mentioned above, e.g. via http://www.photon.t.u-tokyo.ac.jp/~maruyama/kataura/chirality.html. That way it will be easier for you to successfully run the calculations you are interested in.

36
Yes, you have the right idea, you need to set up the electrodes as supercells, rather than using the automatically detected repetitions (3x in the Y-direction).

If you still have the setup in the Builder, just go back to "Device from Bulk" and choose "Set up electrode cross-section as a supercell"

I would also recommend to add 1 or 2 more gold layer inside the central region, you currently only have an exact copy of the electrode in there, which can cause some problems, esp. if you later want to change the device.

37
We have not yet implemented symmetry reduction for the main SCF loop in the LCAO code because it's technically a bit harder than with planewaves. We use solid harmonics, and the phase relationships when you apply the symmetry operations are not so trivial.

If you use the PlaneWave calculator in QuantumATK you will see that it does use symmetry reduction.

38
Questions and Answers / Re: Zero Bias Transmission
« on: June 16, 2021, 23:01 »
There are many examples in the literature showing that just extrapolating the zero-bias transmission spectrum into a I-V curve can be completely different from the real solution, where you converge the SCF at each bias point. The most obvious effect which you cannot capture with the zero-bias approach is negative differential resistance.

See e.g. https://journals.aps.org/prb/abstract/10.1103/PhysRevB.85.184426 and note how different the transmission is as a function of bias. The band edges will often even "follow" the bias window.

39
Click Text representation or Export on the labfloor

40
When you see very large charge transfer from one atom like Carbon number 235 (it has gained 12 electrons!) you can tell something is not physical in the simulation. it's not  a numerical issue, or convergence problem, but rather there is something strange with all those H atoms...  Since graphene is sp2, I am not sure you can just add CH3 (sp3) like that. And even if you can, the H atoms are too close to each other.

41
Questions and Answers / Re: error
« on: June 16, 2021, 22:50 »
It looks like the HDF5 file has become corrupt. Might be easier to just rerun the calculation...

42
Questions and Answers / Re: Electrostatic Doping
« on: June 16, 2021, 22:49 »
The implicit or electrostatic doping as you put it is similar to explicit actual dopant atoms in that results in a similar shift of the Fermi level, which in turn is a result of charge redistribution, so yes it should be able to describe the depletion region. It does not however, naturally, add any dopant scattering as real dopants would, but if that is not a major concern (and other effects like dopant clustering or position-dependent effects) then I would say that the electrostatic doping provides a physically equivalent picture as real dopants, but with major benefits such as the ability to handle any (and in particular low) concentration without making enormous supercells plus also the possibility to use semi-empirical methods which do not have parameters for the dopant atoms.

43
Questions and Answers / Re: Build silicene nanoribbons
« on: June 16, 2021, 22:25 »
It's a bit trickier with the buckled sheet, since the automatic passivation tool does not recognize it as sp2 when it's not flat, but you are absolutely right that the Custom Passivator (should really be called Advanced Passivator...) will manage it when you explicitly select sp2!

44
Questions and Answers / Re: Build silicene nanoribbons
« on: June 16, 2021, 20:32 »
That is the definition of a nanoribbon, yes... And hydrogen passivation of the edges

45
Questions and Answers / Re: Build silicene nanoribbons
« on: June 16, 2021, 02:23 »
Yes you might need to swap axes to align the transport direction, but that will be needed anyway because the C axis (the transport axis in QuantumATK) is perpendicular to the sheet in the original system.

After the supercell operation as indicated above, it's very simple to make armchair and zigzag by noting that the new A direction is along the zigzag direction, and B along armchair, as you can see if you make some repeats (picture below).

Pages: 1 2 [3] 4 5 ... 334