916
General Questions and Answers / Re: Electron Density...
« on: May 21, 2015, 08:40 »
It is the average of the electron density, projected along an axis. It says nothing about which part(s) of the electron distribution are carriers.
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.
917
General Questions and Answers / Re: Gate-Drain Capacitance
« on: May 20, 2015, 19:20 »
Perhaps this tutorial will help you: http://quantumwise.com/publications/tutorials/item/511-calculation-of-atomic-scale-capacitance
918
General Questions and Answers / Re: Is the experimental conductance of graphene comparable with ATK results?
« on: May 20, 2015, 19:18 »
That is a really interesting scientific question, but I am not quite sure what the answer is. It may even be that you would get different answers depending on who you ask 
If the measurement is really done under conditions that can be accurately modeled by theory, I would expect ATK to be the weapon of choice.
If the measurement is really done under conditions that can be accurately modeled by theory, I would expect ATK to be the weapon of choice.
919
General Questions and Answers / Re: methods used for spin transport measurements
« on: May 20, 2015, 18:50 »
The theoretical foundation of DFT electron transport calculations using ATK is pretty well covered in a scientific paper (http://journals.aps.org/prb/abstract/10.1103/PhysRevB.65.165401) as well as the ATK reference manual (http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/chap.negf.html).
We use the non-equilibrium Greens function method, also for the case of spin transport. Several examples of this are available of tutorials on spintronics calculations, including MTJs (http://quantumwise.com/publications/tutorials/itemlist/category/95-spintronics).
Doping is easily handled by adding or removing background charge. This modifies the external potential acting on the Kohn-Sham one-electron states. See "http://quantumwise.com/publications/tutorials/item/820-ni-silicide-si-interfaces" and "http://quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.atomiccompensationcharge.html" for more info.
Hope this helps.
- Jess
We use the non-equilibrium Greens function method, also for the case of spin transport. Several examples of this are available of tutorials on spintronics calculations, including MTJs (http://quantumwise.com/publications/tutorials/itemlist/category/95-spintronics).
Doping is easily handled by adding or removing background charge. This modifies the external potential acting on the Kohn-Sham one-electron states. See "http://quantumwise.com/publications/tutorials/item/820-ni-silicide-si-interfaces" and "http://quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.atomiccompensationcharge.html" for more info.
Hope this helps.
- Jess
920
General Questions and Answers / Re: spin-dependent Seeback coefficients
« on: May 20, 2015, 14:56 »
You are right, VNL currently calculates the Seebeck coefficient from the total electronic transmission spectrum (the sum of spin-up and -down spectra). So no, you are currently not able to get spin-resolved Seebeck coefficients.
- Jess
- Jess
921
General Questions and Answers / Re: Why a delocalized spin-up HOMO can not give rise to a strong spin-up transmisson
« on: May 20, 2015, 13:34 »
Perhaps the spin-up HOMO is unlikely to be populated?
922
General Questions and Answers / Re: Question on antiresonance in transmission
« on: May 20, 2015, 13:30 »
The dip does not look like a numerical artifact. Is it quantum interference? Hard to say.
923
General Questions and Answers / Re: Boundary Conditions Set between two gate and one gate FET
« on: May 20, 2015, 13:24 »
Let my first correct my previous answer a bit: You can use metal gates that extend all the way to the supercell boundaries. However, this is not likely to change your results.
Real-space derivative: What I mean is that Neumann boundary conditions impose the condition that the field is constant at the boundary, but imposes no conditions on the strength of the field at the boundary.
Field strength: Do you mean the electric field strength between the gates?
Real-space derivative: What I mean is that Neumann boundary conditions impose the condition that the field is constant at the boundary, but imposes no conditions on the strength of the field at the boundary.
Field strength: Do you mean the electric field strength between the gates?
924
General Questions and Answers / Re: Building the structure of blue phosphorene using optimization
« on: May 20, 2015, 13:15 »
So the atomic structure of blue phosporene is really like a graphene sheet, except that all atoms are phosphor and every second atom is displaced down as compared to the others. However, this is not the most stable allotrope of solid phosphor, so it is most likely meta-stable. Not all initial guesses for the structure will therefore relax to blue phosphorene when using Geometry Optimization.
One quick way to set up this structure is using the Nanosheet plugin in the Builder: Use (m,n) = (1,1) for the C-vector alignment with the unit cell. This produces a graphene sheet (4 atoms in the rectangular unit cell). Next, change all atoms to P, and use Coordinate Tools -> Translate to displace 2 of the atoms a few Angstroms along the x axis. Lastly, center the atoms in the unit cell, and change the unit cell lattice parameters to (A, B, C) = (10, 6, 3) Å while keeping the fractional coordinates of the atoms constant.
I found that relaxing the system produced by the above procedure leads to blue phosporene with P-P distances of 2.27 Å (see attached script).
- Jess
One quick way to set up this structure is using the Nanosheet plugin in the Builder: Use (m,n) = (1,1) for the C-vector alignment with the unit cell. This produces a graphene sheet (4 atoms in the rectangular unit cell). Next, change all atoms to P, and use Coordinate Tools -> Translate to displace 2 of the atoms a few Angstroms along the x axis. Lastly, center the atoms in the unit cell, and change the unit cell lattice parameters to (A, B, C) = (10, 6, 3) Å while keeping the fractional coordinates of the atoms constant.
I found that relaxing the system produced by the above procedure leads to blue phosporene with P-P distances of 2.27 Å (see attached script).
- Jess
925
General Questions and Answers / Re: The error of Phonon transmission spectrum
« on: May 19, 2015, 16:54 »
We have confirmed that this is a bug. It will be fixed in the 2015 version of ATK.
Tre phonon transmission spectrum turns out to look like illustrated in the attached pdf.
- Jess
Tre phonon transmission spectrum turns out to look like illustrated in the attached pdf.
- Jess
926
Future Releases / Re: Suggestion about Page Shrink in VNL
« on: May 19, 2015, 16:05 »
Dear Fang Yong,
thank you for this suggestion. We will make a note of it.
- Jess
thank you for this suggestion. We will make a note of it.
- Jess
927
General Questions and Answers / Re: brillouin zone route
« on: May 19, 2015, 16:02 »
Yes, I believe that ought to work, provided that the MoS2 K-point is located at (kx, ky, kz)=(1/3, 1/3, 0).
928
Installation and License Questions / Re: ATK on CRAY XC Series
« on: May 19, 2015, 15:00 »
Hi.
That sounds very interesting. Let me try to answer your questions:
1) ATK uses Intel MKL for matrix operations. Provided that the CRAY XC series offer MKL, the ATK software should work well on such a system.
2) The ATK computation time vs. number of CPUs depends very much on the type of system you are considering. For bulk calculations with DFT we parallelize efficiently over k-points and more, but at some point the Hartree potential will start hurting you. For device calculations, we take advantage of sparse methods and k-point parallelization to drive down the time per SCF step. Obviously, device calculations with semi-empirical methods will scale differently than with DFT.
3) The number of atoms than can be computed with 8 GB available RAM will depend a lot on a) the physical system (which elements? how big a supercell? etc.), and b) the DFT method (how many basis functions per atom? should the calculation be spin-polarized, noncolinear, or include spin-orbit interactions? etc.). The scaling is not gonna be linear. It will most likely be quadratic wrt. total number of basis functions in the calculations. The curse of matrix algebra
What sort of systems are you thinking of running? It would be easier to estimate the scaling you will get if you could provide an example of a calculation (in private if [email protected] is used).
Best,
Jess
That sounds very interesting. Let me try to answer your questions:
1) ATK uses Intel MKL for matrix operations. Provided that the CRAY XC series offer MKL, the ATK software should work well on such a system.
2) The ATK computation time vs. number of CPUs depends very much on the type of system you are considering. For bulk calculations with DFT we parallelize efficiently over k-points and more, but at some point the Hartree potential will start hurting you. For device calculations, we take advantage of sparse methods and k-point parallelization to drive down the time per SCF step. Obviously, device calculations with semi-empirical methods will scale differently than with DFT.
3) The number of atoms than can be computed with 8 GB available RAM will depend a lot on a) the physical system (which elements? how big a supercell? etc.), and b) the DFT method (how many basis functions per atom? should the calculation be spin-polarized, noncolinear, or include spin-orbit interactions? etc.). The scaling is not gonna be linear. It will most likely be quadratic wrt. total number of basis functions in the calculations. The curse of matrix algebra
What sort of systems are you thinking of running? It would be easier to estimate the scaling you will get if you could provide an example of a calculation (in private if [email protected] is used).
Best,
Jess
929
General Questions and Answers / Re: Boundary Conditions Set between two gate and one gate FET
« on: May 19, 2015, 14:10 »
- The FET models in 2.png and 3.png are not appropriate for computations. Your gates should not extend all the way to the boundaries along B of the supercell.
- As for 1.png and 4.png, I can not really "suggest" any of them over the other. The choice of model device depends on the physical device you want to simulate.
- Neumann boundary conditions along B should work well for both 1.png and 4.png. You want the electrostatic field on the boundaries above and below the device to be constant (finite or zero) but with zero real-space derivative.
- As for 1.png and 4.png, I can not really "suggest" any of them over the other. The choice of model device depends on the physical device you want to simulate.
- Neumann boundary conditions along B should work well for both 1.png and 4.png. You want the electrostatic field on the boundaries above and below the device to be constant (finite or zero) but with zero real-space derivative.
930
General Questions and Answers / Re: convergence problem
« on: May 19, 2015, 13:47 »
How did you increase the maximum number of SCF steps? Did you increase "Maximum steps" in "Iteration Control" using the Script Generator?