Messages

31

General Questions and Answers / Re: MemoryError

« on: January 12, 2021, 15:16 »

What is the QuantumATK version you are using? It would be helpful to see your script and log file, and know how you run this job, 48 Gb is it for a node or core (MPI process)? 

32

General Questions and Answers / Re: Query regarding calculation of capacitance using ATK

« on: January 12, 2021, 14:02 »

Device calculation means that you have a central region of whatever system sandwiched between two semi-infinite electrodes, this is why these layers are added, in fact, the represent semi-infinite graphite-like leads.

33

General Questions and Answers / Re: Slater-Koster method for charge transfer calculations

« on: January 12, 2021, 14:00 »

I am not sure SK would be work for this kind of problems. Did you see any example in literature? You may always verify that by benchmarking SK against DFT, e.g., calculating absorption energy or any physical quantity of interest, at least for some configuration, to see how that performs. 

For the second one, you may introduce a solvent dielectric constant in the Poisson solver for DFT as well. Also, did you do self-consistent calculation with SK (you would need to unselect "No SCF iteration" in the Calculator)?  Seeing your script might be helpful.

34

General Questions and Answers / Re: Band structure calculation of BN monolayer

« on: January 12, 2021, 11:57 »

It is not really about running it with different codes; this matter is fully geometrical related Brillouin zone folding when enlarging unit cell size.

If 1x1 BN refers to 1x1 primitive cell (not some other unit cell) of hexagonal BN, then K-point of the primitive Brillouin zone folds onto the Gamma-point of the 6x6 hexagonal BN Brillouin zone.

For example, 2x2 BN would have K-point folded on K-point, the same as for graphene, but 3x3 BN would have it on Gamma-point, meaning that 6x6 BN also has it folded on the Gamma-point.

I am guessing that there must be some issue with interpretation of the VASP band structure data.

35

General Questions and Answers / Re: Effect of less (less than 15 angstrom) vacuum region in context of periodic inte

« on: January 12, 2021, 10:57 »

I do not think there is any magic vacuum thickness such as 15 Ang to assume that artifacts of using periodic boundary conditions are eliminated. One should always verify if the results are converged with respect to this parameter.

Note that the best way to avoid this issue is to use Dirichlet/Neumann boundary conditions  in left/right vacuum region for slab calculations, and Dirichlet and Neumann for the left electrode and right vacuum region for surface Green's function calculations, i.e., for one-probe device. One still has to choose vacuum regions carefully, checking that there exist sufficiently large, flat potential (e.g., HartreeDifferencePotential) regions in vacuum near the boundaries.

36

General Questions and Answers / Re: Poisson Ratio nanotube

« on: January 8, 2021, 11:37 »

You can definitely calculate and print out total energy as given in the usage examples in the QuantumATK manual, https://docs.quantumatk.com/manual/Types/TotalEnergy/TotalEnergy.html. You can print other quantities (strain value, radius and so on) in a similar manner. Plotting can be done, e.g., using pyplot, https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.html, https://matplotlib.org/3.2.1/api/_as_gen/matplotlib.pyplot.plot.html, that requires some python scripting and then one can execute the script by sending it to the Job Manager,

37

General Questions and Answers / Re: Query regarding calculation of capacitance using ATK

« on: January 8, 2021, 11:31 »

Could you clarify what capacitance you are trying to compute? You mentioned capacitance of bilayer graphene. Would you then mean quantum capacitance as discussed in-here https://en.wikipedia.org/wiki/Quantum_capacitance?  Otherwise, you are now calculating capacitance of a capacitor comprised of graphite electrodes for a given separation distance between the graphite plate. In that case I am sure that you have enough monolayers for your plates.

38

General Questions and Answers / Re: question on transistor simulation tutorial

« on: January 8, 2021, 11:25 »

The result shows a very low transmission in -1 to 1 eV range (Vg=0) using GGA-SZP which is in good agreement with Armchair GNR bandgap IF we consider it very long along z-direction, but in this tutorial, we are studying an armchair in z shape which is very short and if we just split this device3.py and add 20 angstrom to the center and getting bandstructure we will see that the bandgap is reduced to 0.2eV which means that due to quantum confinement along z-direction the new (z shape armchair ) has small bandgap which is not in agreed with the transmission results.
I can not understand the channel is considered as a  long armchair ? or is it a small one? or due to very little subbands forming a 0.2 eV gap in this quasi-quantum dot armchair shape structure! we can not get the transmission around it?

Could you provide more information on this calculation, python script, log file and possibly hdf5 output or transmission plot at least? What is the QunatumATK version you are using?

39

General Questions and Answers / Re: error in estimating thermoelectic coefficient through script file

« on: January 8, 2021, 11:18 »

OSError: The specified NC file does not exist, or does not contain a transmission spectrum.

40

General Questions and Answers / Re: error in estimating thermoelectic coefficient through script file

« on: January 8, 2021, 11:18 »

OSError: The specified hdf5 file does not exist, or does not contain a transmission spectrum.

41

General Questions and Answers / Re: Editing Transmission Plot

« on: January 7, 2021, 16:17 »

Unfortunately, you have to export transmission spectrum data in xy-format (right-click on the plot and then export data) and replot it. In fact, you can still do it in QuantumATK, writing a plotting script using pyplot, https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.html, or similar, and then submit it to the Job Manager or execute it from the command line: atkpython plotting_script.py.

42

General Questions and Answers / Re: apply an electric or magnetic field to a molecule

« on: November 17, 2020, 09:54 »

Regarding a molecule in electric field, please take a look at these tutorials, https://docs.quantumatk.com/tutorials/molecular_electronics.html, in particular, the one on Molecular device. One may also apply external electric field without using device configuration, see https://docs.quantumatk.com/tutorials/opening_a_band_gap/opening_a_band_gap.html.

QuantumATK does not allow for external magnetic field application.

43

General Questions and Answers / Re: How do I save the GUI inteface in the QATK Script Generator

« on: November 16, 2020, 14:49 »

You may use Templates in the GUI, as described in this guide, https://docs.quantumatk.com/guides/scripter/templates/templates.html.

44

General Questions and Answers / Re: calculation electron affinity of graphene

« on: November 16, 2020, 14:47 »

Work function of a free-standing or supported graphene monolayer can be calculated as done, e.g., in PHYSICAL REVIEW B 79, 195425 (2009). For accurate calculations, I would suggest using DFT-PlaneWaveCalculator with PseudoDojo pseudopotential and mixed Neumann-Dirichlet boundary conditions in the out-of-graphene-plane direction.

Note that you should make sure that your DFT-PW calculations are converged with respect to wave function cutoff and k-point sampling in the in-plane directions.

You may also find extra information on work function calculations at the following tutorial, https://docs.quantumatk.com/tutorials/work_function_ag_100/work_function_ag_100.html. There is one important difference when using LCAO (tutorial) and PW (present suggestion) basis sets. When using PW basis set one does not need any ghost atoms, as often required for LCAO-based work function calculations.

45

General Questions and Answers / Re: SlaterKoster Parameters for Tungsten

« on: October 19, 2020, 11:28 »

In general, I would say that one should do SCF. To really validate Non-SCF, one would have to benchmark it against SCF-obtained results I think.