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 - Petr Khomyakov

Pages: 1 ... 69 70 [71] 72 73 ... 86
1051
For a given k-point, I guess that the number of eigenvectors is the same as the number of phonon modes = 3xN, where N is the number of atoms in the unit cell you have adopted to calculate phonons in your system; the eigenvector dimension is given by the dimension of the Dynamical Matrix as briefly discussed in http://docs.quantumwise.com/manuals/Types/VibrationalMode/VibrationalMode.html.

If N=2, e.g., for the primitive cell of a single graphene sheet, the matrix dimension is then 6x6, meaning that each eigenvector has 6 elements. 

1052
You have to converge the total energy and other physical quantities of interest with respect to the mesh density for a given pressure applied. I guess this is the only way to actually see if there exists any significant dependence of the density mesh cut-off on the external pressure. It is very likely to be insignificant for low pressure, and it might become an issue when the pressure is high, but this is a matter of numerical experiment to figure out what is low and high for your specific system.

1053
General Questions and Answers / Re: job manager error
« on: May 12, 2017, 22:47 »
The 13.8.1 version of ATK is rather old, and not supported anymore. Please upgrade to a newer version.   

1054
General Questions and Answers / Re: data points complex bs
« on: May 10, 2017, 16:09 »
You may use Text Representation to see the complex band structure in xy-format for the propagating modes. For the evanescent modes (scroll down to see the corresponding data in the text file), it is given in xyz format, where x->energy, y->k, z->kappa.   

1055
I do not think you can distinguish the longitudinal and transverse modes in the phonon band structure plot without doing the analysis of the actual phonon eigenmodes resulted from the Dynamical Matrix diagonalization.
 

1056
General Questions and Answers / Re: transmission spectrum
« on: May 9, 2017, 09:15 »
Thank you for enclosing the scripts, but as I said the two transmission spectrum calculations are done for two different systems. One system is a periodically repeated defect, whereas the second one is a single defect. There is no reason why the two spectra should be identical.

I would suggest reading a book on the electron transport theory to understand the physics behind the electron transport calculations, e.g., see Supriyo Datta, Electronic Transport in Mesoscopic Systems, http://www.cambridge.org/catalogue/catalogue.asp?isbn=0521599431.

1057
General Questions and Answers / Re: transmission spectrum
« on: May 8, 2017, 13:44 »
Your electrodes in the device are not made of the same material as the scattering region as you do have a sulfur atom (a scatterer) in the middle of the device, whereas there is not sulfur in the electrodes.

In addition, you should always check if the (k_A, k_B)-grid is the same in the two calculations. You have adopted the default values, but the defaults might be different for these two kinds of calculations, in principle. To see the defaults, you would need to save the script details with "Show defaults" instead of "Minimal".   

1058
I am sorry I still do not understand your actual question. The bulk MoS2, WS2, and h-BN materials are available in the Database in the Builder. So, you can build the corresponding multilayer materials in the same way you did for 1ML-MoS2. The band structure calculations can also be done in the way you did for 1ML-MoS2.

1059
General Questions and Answers / Re: transmission spectrum
« on: May 8, 2017, 13:11 »
As I said the computational settings for the two calculations are to be effectively similar to get the same result. From the images, it seems that you have adopted too coarse k-grid for the 2D Brillouin zone integration of the Transmission Spectrum in the bulk calculation. You may see that the device-derived spectrum is much smoother compared to that from the bulk calculation. If you posted your python scripts related to these calculations it might make it easier to figure out the cause for this discrepancy.   

1060
General Questions and Answers / Re: Builder error
« on: May 8, 2017, 11:59 »
Please contact the QuantumWise support (support@quantumwise.com) for this issue. You would also need to describe all the steps you did to get this error. 

1061
General Questions and Answers / Re: transmission spectrum
« on: May 8, 2017, 11:51 »
In principle, it should not make any difference as soon as the both calculations of the transmission are numerically converged with respect to computational settings.

1062
I am not sure I understand the question. Which elements are you talking about and why this is related to the band structure calculation of single and double layer materials? In most cases, you can build a single layer or multilayer system from the corresponding 3D layered material. The band structure is then calculated in the same way as for 3D materials, but you just have to use a single k-point (e.g., Gamma point) in the out-of-plane direction.   

1063
I would suggest going through getting started tutorials at http://docs.quantumwise.com/tutorials/getting_started.html. Here you may see particular example of the usage of the basic VNL-ATK tools.

Regarding building a new structure in the VNL, this is done in the Builder of the VNL; you may take a look at the Builder manual for more details. Another approach is to go through the tutorials and case studies at http://docs.quantumwise.com/ to see the examples of how particular structures have been created in the VNL.   

1064
I see no obvious reason why the Kubo-Greenwood formula should fail for layered materials. You can do the Optical Spectrum calculation in a usual way as you do for three-dimensional systems. The only difference is that you have to adopt a thick vacuum layer in-between the periodically-repeating layers in the out-of-plane direction.

1065
General Questions and Answers / Re: what's DFT+Σ method?
« on: April 25, 2017, 10:23 »
I can only guess that DFT+Σ might refer to a many-body method such as GW. This kind of many-body approaches are not implemented in the ATK yet. I note that the GW is computationally extensive, not really practical for large systems. We are currently working on implementing the hybrid functional approach such as HSE, which is less expensive, but often provides the GW accuracy for energy gap calculations.

Pages: 1 ... 69 70 [71] 72 73 ... 86