Topics

31

General Questions and Answers / the reliability of Huckel method on the calculation of IV curve

« on: June 2, 2017, 03:29 »

Dear experts,

Who can tell me something about the reliability of the IV curves calculated using ATK SE: Extended Huckel method?

It's much faster than DFT method as calculating the IV curve of devices. In many cases, it's almost mission impossible to use DFT methods to calculate the multi-points IV curves for a device with many atoms on a local computer, even on a small computer cluster. Thus, ATK SE: Extended Huckel method could be the only choice if there is no suitable Slater-Korster basis set for a particular system.

However, I want to know the difference between the results calculated using DFT or Huckel method. Who can give me some comments on the reliability of the transmission spectrum calculated using Huckel method? or tell me where to find such information?

Thank you very much for your time!

32

General Questions and Answers / which configuration is reasonable?

« on: May 31, 2017, 16:47 »

Dear experts,

I built an interface using "Interface" plugin. However, after the geometry optimization, it resulted in a strange interface configuration (please see attached files), is it reasonable?
The settings for the Calculator and geometry optimization blocks are:

(1) Calculator: GGA-OMX-high,  k_point_sampling=(1, 9, 9), density_mesh_cutoff=200.0*Hartree
(2) Geometry optimization: max_forces=0.01*eV/Ang, max_stress=0.001*eV/Ang**3, disable_stress=True (constrain cell x, y, z)

Thanks very much for anyone who is willing to help me！Thank you for your time and kind help!

lknife

33

General Questions and Answers / Why the MPI process cannot help me to speedup the calculation?

« on: May 28, 2017, 04:32 »

Dear experts,

I am studying the tutorial "Spin-orbit transport calculations: Bi2Se3 topological insulator thin-film device" and want to reproduce the results of the tutorial. I downloaded the script "electrode1.py" and ran it on my local computer using ATK2016 with the default setting of the Job manager. It took me a long time (nearly 5 days) to get the result. The configuration of my computer is :
   
     Intel Core (TM)  i7 6700 CPU, quad core, 3.4 GHz, 16GB RAM

Since it took to too long time, I want to run it on the computer cluster of our university using MPI process to speedup the calculation. Because of the license of ATK, only ATK2015 can be used on the cluster. I modified some parts of the script "electrod1.py" using ATK2015 and submitted it to the cluster using these MPI settings:

     Number of Nodes: 8
     Tasks per Node: 2
     memory per CPU =16GB
     as some other's suggestion, I disabled the threading calculation using "export OMP_NUM_THREADS=1, export OMP_DYNAMIC=FALSE" and set 16 MPI processes to run the calculation.
     (Please see the attached files: electrode1gkm_argo5.py is the python file for calculation and elelctrode1gkm_argo5.sh (txt) is the file for submitting the job to the cluster.)

Now the calculation has been running on the cluster for 2D21hrs, much slower than I expected.
What I want to ask is: Is there anything wrong with my MPI setting or the scripts? Why it's almost no use for speeding up the calculation?

Thank you very much for your kind reply!

34

General Questions and Answers / Do I need to relax the interface before constructing the device?

« on: May 25, 2017, 22:42 »

Dear all,

I am studying the interface based devices following the tutorial "Modeling metal–semiconductor contacts: The Ag–Si interface". I have two questions on this topic, expecting your kind help!

(1) After constructing the interface using "Interface" plugin, such as a in-plane MoS2-WSe2 interface (seen in the attached file), can I use a smaller unit cell (about 2nm) to represent the interface-based device instead of constructing a full-length device (such as 5nm or longer)? That is, I want to use this smaller unit cell for geometry optimization in order to save time. After that, I can construct the full-length device configuration by increasing the size of the central region at the interface of the respective electrode. Is it reasonable?

(2)  After relaxation of the unit cell, I want to construct a device using "device from bulk" plugin. Do I need to relax the electrode and central region, as well as the whole device configuration again according to the tutorial? Or just use the device obtained from "device from bulk" for further calculation?

Thank you very much for your time and kind help!

35

General Questions and Answers / How to "shift surface" in "Interface" plugin?

« on: May 21, 2017, 16:44 »

Dear all,

When building an interface using the "Interface" plugin, I am always confused by the usage of "shift surfaces" menu.

In the tutorial "Building an interface between Ag(100) and Au(111)", it is said that "From this menu it is possible to either adjust the relative position of the two crystal manually, or compute the optimal distance using the ATK-Classical calculator." For the Ag(100)/Au(111) interface, the “EAM_Zhou_2004” potential was used to find the optimal distance.

However, for many systems, there is no classical potential to use. Then, how to adjust the position manually? Do we just need to adjust the distance along z direction or the shifting of the lattice along x and y directions are also needed? What's the criterion for the shifting of the relative position of the two crystal?

Thanks a lot for your kind help!

36

Scripts, Tutorials and Applications / How to make a checkpoint file after a fixed period of time?

« on: May 20, 2017, 17:40 »

On many computer clusters, there would be a hard time limit, such as 12hr, 24 hr or even longer. I want to know if it is possible to make a checkpoint file for later use before it is killed because of time limit. That is, if the calculation is not not converged before the time limit, how can I save the checkpoint file after a fixed period of time?

37

General Questions and Answers / how to center the lattice through python script?

« on: May 18, 2017, 22:04 »

Dear all,

Followed the tutorial, I made a script to adjust the interlayer distance of a bilayer system with various D values. I wanted to calculate the "total energy vs interlayer distance" curve. I found that the value of lattice vectors had some effects on the total energy. Thus, I don't want to change the value of vector c as the tutorial does:
--------------------------------------------   
vc = vc + [0.0, 0.0, d]*Angstrom
--------------------------------------------
However, if I delete this code line, the final bulk configuration will not be at the center of the lattice, which will also affect the successive calculation.

My question is: how to center the bulk configuration through python script?

I am looking forward to your kind help!

Thank you very much for your time!

38

General Questions and Answers / How to set the Brillouin zone route for band calculation of a unit cell

« on: May 16, 2017, 16:51 »

Dear all,

I made a bilayer hetero-structure  using the "interface" plugin and wanted to calculate its band structure. However, I did not know how to set the Brillouin zone route for such a unit cell. I used the default route for unit cell "G, Z" but the result was not correct. Attached is the .py file for this calculation. Can anybody can kindly help me?

Thanks a lot for anyone who is willing to help me!

lknife

39

General Questions and Answers / How to make a loop for a interlayer distance scan?

« on: May 15, 2017, 21:37 »

Dear all,

I constructed a hetero-bilayer using the "Interface" plugin, as seen in attached .py file. However, since there is no  potentials preseted for optimizing the interlayer distance of the two layers, I want to get the value by calculating the total energy of the bilayer system with various interlayer distance values. I tried to follow the tutorial "Carbon Nanotube Junctions" where there is a script dealing with different distance between two graphene nanotubes. However, because of my limited knowledge about python script, there are so many errors with my script that I'm shamed to post it here.   

What I want to do is to make a loop with various interlayer diatance. There are two layers in the bulk configuration: each of them is signed a tag named "Left_layer" and "Right_layer", respectively. Based on an original bulk configuration, each time, I want to change the distance a between these two layers a little to form a new configuration, and then calculate the total energy of this new configuration. By observing the total energy vs. interlayer distance curve, I can get the idea which value of interlayer distance is the optimum one.

There is another question: if the van der Waals interaction is not considered, can I get the right figure of the total energy of the system? Can I get the correct value of interlayer distance by a scan without considering the van der Waals interaction? I tried to follow the tutorial "DFT-D and basis-set superposition error", however, it took me too long the time that I don't want to use it any more.

Thanks a lot for anyone who is willing to help me!

40

General Questions and Answers / Is there anything wrong with my band structure calculation?

« on: May 10, 2017, 21:42 »

Dear all,

Attached are the band structure of a 1-dimension WTe2 nanoribbon together with its .py script. Who can tell me if there is something wrong with the calculation? It’s totally different with other band structure figures so that I hardly have any idea how to analyze it.

Thanks very much for anybody who would like to help me!

41

General Questions and Answers / How to set parallel calculation for ATK 2015?

« on: May 5, 2017, 16:33 »

Dear all,

As mentioned in tutorial “Spin-orbit transport calculations: Bi2Se3 topological insulator thin-film device”, DFT device calculations including spin-orbit coupling can be computationally heavy and may require a lot of memory. Thus, MPI is a good solution for these calculations.

I am using ATK2016 on my local computer to do some scientific calculations. If I want to run MPI calculation on the computer cluster of our university, only ATK2015-MPI can be used. In ATK2016, one can specify the parallel parameters, such as “processes per NEB image”, “processes per individual”, “processes per bias point” and “processes per saddle search” directly in the python script. However, in ATK2015, no such parameters can be used in the script. One can only set the MPI calculation through “Job manager”.

The default script of a “calculator” related to MPI settings for ATK2015 and ATK2016 are as followings:

ATK2015:
equilibrium_method = GreensFunction (
    processes_per_contour_point=1,
    )
non_equilibrium_method = GreensFunction(
    processes_per_contour_point=1,
)

ATK2016:
equilibrium_method = GreensFunction(
    processes_per_contour_point=1,
    )
non_equilibrium_method = GreensFunction(
    processes_per_contour_point=1,
)
parallel_parameters = ParallelParameters(
    processes_per_bias_point=None,
    processes_per_neb_image=None,
    processes_per_saddle_search=1,
    processes_per_individual=None,
    )

One question is: How to set “processes_per_bias_point” in ATK2015? I do think it’s very helpful to save time for I-V calculating if you have enough processes.

For the tutorial above mentioned, the MPI settings involved are the followings:
(1)   “4 MPI processes per k-point”
(2)   “evaluation of the 48 contour points for the Green’s function method is done using 3 MPI processes per contour point”
(3)   “if your cores have only little memory, you may need to allocate 2 cores or more per MPI process”

The guide to use our computer cluster for MPI calculations can be described as these codes:

___________________________________________________________________________________
## N is the number of nodes you want to span you MPI processes
#SBATCH --nodes N
## Set the number of processes per node, this must not exceed the node core count
#SBATCH --ntasks-per-node n
## Request for memory per task, default is 2G so request more if needed. Specify in MB
#SBATCH --mem-per-cpu 4098
____________________________________________________________________________________

I have questions about (1): Can I set “4 MPI processes per k-point” through python script using ATK2015 before submitting the .py file to the cluster? If not, how to set the parameters above for this calculation? Do I Just need to set the “ntasks-per-node” to 4?

I am looking forward to your kind help!  Thank you very much!

42

General Questions and Answers / the topological invariants in tutorial "Bi2Se3 topological insulator"

« on: April 3, 2017, 19:43 »

Dear Sir/all

I am now studying the tutorial "Bi2Se3 topological insulator" and have some questions about it, hoping that someone in this forum can help me!

I followed the directions in this tutorial and have done all the steps except the last section "Python script for the calculating the topological invariants". After downloading the python script "TopologicalInvariant3D.py" without any modification, I executed the scripts below:

——————————————————————————————————————————————————
from TopologicalInvariant3D import topologicalInvariant3D

bulk_configuration = nlread("bi2se3_bulk_sogga.nc", BulkConfiguration)[0]

print '-----------------------------------'
print 'Topological Invariants = ', topologicalInvariant3D(bulk_configuration)
print '-----------------------------------'
——————————————————————————————————————————————————

I replaced the bulk_configuration with the results previously obtained. However, the final results is "Topological Invariants =  [0 0 0 0]", no matter which bulk_configuration was used, GGA or SOGGA. Also, I am not sure how to get the "bi2Se3_bulk_sogga.nc" since it's not in the tutorial.

Can anyone here help me to answer this question? I appreciate your kindness!

Thanks again! 

43

General Questions and Answers / the *.py files in tutorial "Bi2Se3 topological insulator"

« on: April 1, 2017, 06:37 »

Dear Sir,

I am now studying the tutorial "Bi2Se3 topological insulator" and wanted to download the .py files for learning. However, some of the .py files (such as  "bloch_states.py", "fermi_surface.py", "TopologicalInvariant3D.py "   ) can only be saved as .html file which can not be displayed correctly in the Editor of ATK package. Because I am a beginner of python, I do not have very clear idea about the organization of the codes. Who can kindly tell me where I can download these .py file?

Thanks a lot!

44

General Questions and Answers / Which one would be more suitable, LDA or GGA?

« on: March 22, 2017, 16:56 »

Dear all,

I am trying to do some optimization calculation on my layered TMDs. However, when I tried different "Exchange correlation", say, GGA and LDA, there were much differences between the results. The most difference is the interlayer distance.

For example, for my MoS2/WSe2 hetero-bilayer system, the optimized interlayer was 6.1723 Angstrom for LDA-OMX-medium-200 (here 200 is the density mesh cut-off ) and 6.728 Angstrom for GGA-SG15-high-75 (here 75 is the density mesh cut-off, too), respectively. Let alone some other differences such as the lattice vectors and the band structures. There was so much significant differences that I can hardly believe in both of them.

Which one would be more suitable for the optimization of these layered structures? and why? Can anybody kindly help me?

Thanks a lot to anyone who views this message!
 

45

General Questions and Answers / asking computer configuration for running ATK

« on: December 6, 2016, 16:48 »

Hello, everyone,

when I tried to calculate the transmission spectrum of MoS2 nanoribbon using ATK-DFT method, I got an error message like this: ** Back Engine Exception : The parallel solver cannot allocate the required workspace. Consider enabling disk workspace. ** Location of Exception : parallelsolver.cpp:568". 
Since the system I calculated is not a large system, I don't know what's wrong with it. Can anyone help me? Thank you very much!

The following is the configuration of my computer:
Intel(R)core(TM) i7-6700 CPU @3.4GHz,
Installed memory: 16GB,
System type: 64-bit operating system
hard driver: 128G solid state hard driver with  1TB portable hard driver

Thanks again for your kindness!