Messages

1

General Questions and Answers / Re: how many atoms atk can deal with

« on: May 15, 2014, 20:48 »

Hi njuxyh,

Density-function-theory (DFT) based models are computationally  prohibitive for systems having more than about 200 atoms [1].

Depending on the selected method, calculations can be performed on structures with several hundred (DFT/DFTB and Hückel) to tens of thousands (tight-binding) of atoms [2].

[1] H. Raza e E. C. Kan, «An extended Hückel theory based atomistic model for graphene nanoelectronics», J Comput Electron, vol. 7, n. 3, pagg. 372–375, set. 2008.
[2] «Graphene studies with ATK», QuantumWise. [online]. Available at: http://quantumwise.com/applications/graphene/16-graphene-studies-with-atk.

Hope it helps
Luis

2

General Questions and Answers / BandStructure - Extended Huckel Method

« on: May 2, 2014, 06:30 »

Good day ATK user,

I have been following this tutorial:
http://www.quantumwise.com/documents/tutorials/ATK-11.8/BasicGrapheneTutorial/index.html/chap.armchairribbon.html

At the end, they recommended to redo the calculations for different types of ribbons (armchair/zigzag) and for different widths because band gap is strongly dependent on both factors!

I have been testing upon Na-AGNR (Na ranging from 4 to 13) and indeed the band gap varied as expected. I could even recongnize the 3m, 3m+1 and 3m-1 group.

When I try to analyze the same AGNRs but repeted 15 times in the z direction, I get something strange, or at least something that I do not understand. I get too many straight and almost parallel lines (see attached 1 and 2). Also the Ef is not at the center anymore. Can you help me understanding the physical meaning of this?

btw the bandStructure is calculated using Extended Huckel method; to what extent can I trust these results? If for example, I have distorted edges or vacancies,  can I still trust the results obtained?

The attachments represent the Band Structure for a 4AGNRx15.


Many many thanks,
Luis

3

General Questions and Answers / Re: Bandgap vs. DOS curve

« on: April 25, 2014, 18:44 »

Hello all,

Can you clarify why Mr. Hassan is using YXZG route. In the forum (http://quantumwise.com/forum/index.php?topic=2479.0#.U1qDefl5N1Y) Mr Dipankar Saha and Dr. Blom say that G,Z is sufficient. what is the drawback/advantage of using this route?

Thanks,
Luis M. VIllamagua

4

General Questions and Answers / ZGNR with withds from 4 to 9

« on: March 25, 2014, 18:48 »

I am reading a paper where the autor says that he constructed ZGNR with widhts ranging from 4 to 9 accross the ribbon. VNL is not allowing me to do that. It only allows me to create zGNR with odd number of lines accross the ribbon, for example 4, 6, 8, 10, etc.

Am I doing something wrong, or am I missunderstanding what the author on this papers is saying?

Regards,
Luis

5

General Questions and Answers / Number of bands to plot

« on: March 14, 2014, 21:25 »

Hello everyone,

Do you know guys how many bands should I plot when computing the BandStructure? When I plot all bands I get a graph that I cannot understand and I have seen some paper that they plot only a finite number of bands.

Regards,
Luis

6

General Questions and Answers / Re: Plotting postcar from vasp with VNL

« on: March 14, 2014, 21:22 »

It did not work when the coordintes where direct. I used VMD to transform to cartesians and ploted it with VNL and did work.

Regards,
luis

7

General Questions and Answers / Plotting postcar from vasp with VNL

« on: March 14, 2014, 17:18 »

Hi guys,

Can you please give me some information about how to plot the POSCAR from VASP with VNL?


Regards,
Luis

8

Installation and License Questions / Re: Error while starting license server for ATK 10.8

« on: February 28, 2014, 11:24 »

Sir,

Today, when re-installing the license I had the same problem. This is how I overcame the problem:
Changed the owner of the "tmp"  directory to be the user and NOT the root. In a fresh linux installation this directory belongs to the root.

sudo chown -R <user> /var/tmp

Regards,
Luis

9

General Questions and Answers / Re: Tight binding - calculation

« on: February 18, 2014, 11:38 »

For example, to calculate the electronic properties ...

(1) If I continue using DFT, should this ensures that the results I will get will be accurate?
(2) If I use TB, could I not get accurate results?

I am new with the topic, so I would appreciate if you can help me choosing the more convenient route.

Thanks again,
Luis

10

General Questions and Answers / Tight binding - calculation

« on: February 17, 2014, 09:45 »

I was said that geometry optimization had to be carried out by DFT. Should I calculate DOS and Band Structure with tight binding? or should I continue using the same framework?

Regards,
Luis

11

General Questions and Answers / How to set the mesh grid cutoff

« on: February 14, 2014, 17:36 »

Hi guys,

Can someone explains to me what Grid Mesh cutoff is for?? VNL is giving me default values. I do not know if I should Increase or decrease it value? Or should I optimize it the same way I optimize the k-point? Does it depends on the number of atoms my nanoribbon has??

Thanks in advance for your help.

Regards,
Luis

12

General Questions and Answers / DOS and bandstructure of a ZGNR

« on: February 11, 2014, 14:33 »

Can you kindly help me understand my results for bulk zigzag Chirality? Could be that I am not reading the results properly

I analysed the DOS and Band Structure of AGNR, obtaining results quite reasonable. However, when I carried out the same analyses for ZGNR I got results that does not make any sense to me; there are many lines crossing.

I followed this tutorial (http://www.quantumwise.com/documents/tutorials/latest/GrapheneBloch/index.html/chap.bandstructure.html) and I expected to see something close to that.


This is what I did .....

(1) Relaxed a pristine 8ZGNR 12 times repeated

Calculator: DFT
k-point mesh: 1x1x9
Grid mesh cut-off: 75 Hartree
Basis set: DoubleZetaPolazired for both Hydrogen and Carbon

(2) On the border of the 8ZGNR a Stone-Wale defect was inserted

The SW defect was inserted at the middle of the ribbon along the z direction


------------------X-------------------



---------------------------------------


(3) Relaxed the nanoribbon with the defect included

I allowed to move the atoms that were at maximum 6AA around the defect.

(4) Calculated the DOS and Band structure

Calculator: DFT
k-point mesh: 1x1x21
Grid mesh cut-off: 75 Hartree
Basis set: DoubleZetaPolazired for both Hydrogen and Carbon

DOS block
k-point grid 1x1x21
Energy Zero: Fermi Level
Bands above the Fermi level: All

BandStructure block
points per segment: 200
Brillioum zone route: G, Z

As I understand, if I insert a defect on the border of a ZigZag Nanoribbon (as specified above), the bandgap should increase. In my case however, this is not happening.

I have no clue about what could the problem be. I attach the graphs of my simulations.


Regards,
Luis

13

General Questions and Answers / Re: Geometry optimization 3AGNR repeated 9 times in the z direction

« on: February 4, 2014, 16:23 »

I understand Sir.

Thanks again,
Luis

14

General Questions and Answers / Re: Geometry optimization 3AGNR repeated 9 times in the z direction

« on: February 4, 2014, 16:16 »

For the moment I want to calculate the electronic characteristics (DOS, bandstructure, etc.) of the nanoribbon as bulk. Is then 1x1x9  ok for this part?

Afterwards, I will use VNL to create a device in order to compute the transport characteristics (transmission spectrum, transmission pathway, etc). I suppose I should calculate the k-point mesh again at this point. It should be 1x1x80 approximately, right?

Regards,
Luis

15

General Questions and Answers / Re: Geometry optimization 3AGNR repeated 9 times in the z direction

« on: February 4, 2014, 11:32 »

Very very clear now  dear Humberto

About k-points: yes, it worked!

I computed the k-point grid for two different nanoribbons (bulk configuration), getting:

For 8ZGNR (12 times repeated  in the z direction) => with DFT (1x1x5) with Huckel (1x1x5)
For 16ZGNR, (12 times repeated in the z direction) => with DFT (1x1x6) with Huckel (1x1x7)

With Huckel took me a couple of minutes while with DFT took hours

When relaxing both geometries I used 1x1x9 in order to ensure a perfect sampling onto the BZ; I hope that this will not poor the accuracy of my calculations.

By the way, (correct me if I am wrong, please) the k-points should be found only once at the beginning. I mean, one must find the k-points of an specific nanoribbons (by changing the kc point and seeing where the energy converges) and then use this grid until the end of my project, right? For example if I want to insert a defect (SW defect in my case), should I still use the same k-point grid?

Regards,
Luis M. Villamagua