Messages

61

General Questions and Answers / Re: LDA+U for transport calculation

« on: April 22, 2011, 10:13 »

Can we apply LDA+U (or GGA+U..or whatever+U) to transport calculation?

Yes
One example is attached, which mimics the calculation presented in Appl. Phys. Lett. 94, 252103 (2009).

62

General Questions and Answers / Re: Why the color bar is always red?

« on: April 22, 2011, 08:15 »

It may depend on the graphic card( or driver) you use.
Could you confirm whether or not you get the same picture with a different computer?

63

General Questions and Answers / Re: silicon nanowire bandgap query

« on: April 22, 2011, 07:18 »

Dear ramkrishna,

First of all, both structures you configured are inadequate.
In the case of SiNW_1.py, you should define the structure as not a bulk but a molecule.
In the case of silicon_wire.py, the length of C vector of the unit cell is wrong.
The correct one is "5.4306" angstrom.

Could you try "silicon_wire-1_modified.py" and verify that the appropriate calculation can be done with it?

64

General Questions and Answers / Re: some questions about calculation of MPSH

« on: April 17, 2011, 06:42 »

You can calculate MPSH eigenstates and spectrum by using Eigenstate() and MolecularEnergySpectrum() respectively.

http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.eigenstate.html
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.molecularenergyspectrum.html

65

General Questions and Answers / Re: the periodic boundry of system

« on: April 16, 2011, 04:44 »

we cann't exculde the influence of electrodes on whole device!

Yes, so the modeling you should do depends on what you want.
For instance, if you want to investigate a contact resistance between electrodes and molecule or the influence of electrodes against the electron transport properties more accurately, you should prepare 3D electrodes.
On the other hand, if you want to focus mainly on the intrinsic electron transport properties of molecules, using virtual electrodes may be permitted.

how can we assure the sense of periodic boundry at X,Y direction if we use the virtual electrodes which have large padding region!

In such a case, periodic boundary condition is not indispensable but only an option.
Using Neumann boundary condition gives the same result as periodic boundary condition if the cell size is enough large.

66

General Questions and Answers / Re: the periodic boundry of system

« on: April 16, 2011, 03:08 »

But if make the padding region larger enough at X,Y direction ,can we avoid the interation between the molecule and its mirror?

Exactly, and the calculation for the system you introduced should be done so.
I guess the electrodes is actually not 3D but 1D-chain.
In molecular devices, the electron transport properties strongly depend on the electronic structure of molecules bridged between electrodes.
That's why the calculation using virtual electrodes could make some sense.
("virtual" means there is not such a electrode in real)

67

General Questions and Answers / Re: Convergence problem of Heusler alloy

« on: March 7, 2011, 10:40 »

How to do a bulk calculation to find the spin value as initial guess...

Examples are attached.
You can obtain spin polarization of each atom by executing "co2crsi_bulk.py" and "cu2cral_bulk.py" and calculating the difference between up spin charge and down spin charge.
And then, you can estimate the adequate initial guess, for instance, for Cr is (spin polarization/4).
http://www.quantumwise.com/documents/manuals/ATK-2008.10/ref.electrondensityparameters.html#ref.electrondensityparameters.description

the central spacer Cu2CrAl is a non-magnetic material

It may be true in real, but ATK 2008.10 predicts that Cu2CrAl has also spin-polarization.
I doubt that the electronic structure of Cr can't be treated properly in ATK 2008.10.
So, I strongly recommend you get upgrade of ATK to 10.8 or 11.2 because this issue is fixed in new ATK.
http://quantumwise.com/forum/index.php?topic=667.msg3531#msg3531

68

General Questions and Answers / Re: Convergence problem of Heusler alloy

« on: February 14, 2011, 05:35 »

Dear baizq,

Thanks for your help with our problem. I have tried your script and convergenve is really achieved after 23 loops. I have to say that your modified script is a great support to our work because we struggled to tune the parameters but the system was always failing to converge.

It's my pleasure

I also have a question here: why do you set the initial elemental spins as those in your script?

I determined each initial scaled spin from spin polarizations of bulk Co2CrSi and Cu2CrAl.

I have to calculate the transmission properties for both antiparallel and parallel cases, and I have no idea how to set the initial spins for each kind of element involved, could you please kindly give me some suggestion on this issue?

In anti-parallel case, indeed it's not obvious.
One candidate is that spin reversal occurs at the center of Cu2CrAl.
How about attached?

69

General Questions and Answers / Re: Help! How to calculate the TMR values of Co2MnSi/MgO/Co2MnSi MTJ?

« on: January 27, 2011, 05:14 »

If you have the script, please give me

Attached is the sample structure of Co2MnSi/MgO/Co2MnSi for ATK 10.8.
Please enjoy 

70

Scripts, Tutorials and Applications / Re: How to get the value of work function with ATK 10.8.2

« on: January 21, 2011, 03:15 »

I modified workfunction-analyzer.py as workfunction-analyzer2.py so that work function is calculated from effective potential.
I've found that in some cases(e.g. Al), a vacuum level is NOT the largest in electro static potential.
That's why, it is more efficient to calculate work function from effective potential

71

General Questions and Answers / Re: error when calculate scf of LaMnO3

« on: October 15, 2010, 06:27 »

I have calculated transmission coefficients and transmission eigenvalues for a two-probe system, I found transmission coefficient and the sum of transmission eigenvalues are not equal at (0.0, 0.0) k-point, one is 8.59771e-11 and the other is 2.10421766e-13. I then calculated the two values at other k-points, some k-points they are equal and some are not,but the difference is small.
Why they are different? Smiley

Such small values should be regarded as "0" and the very little difference comes from numerical error.
That's why transmission coefficient and the sum of transmission eigenvalues at (0.0, 0.0) k-point are equivalent physically.

72

General Questions and Answers / Re: Help:why the transmission spectrum is negative value？

« on: September 12, 2010, 08:38 »

It may be due to the issue of Krylov algorithm.
It is expected that using recursion algorithm alternatively will avoid this issue.

73

General Questions and Answers / Re: Very slow Ni convergence

« on: September 3, 2010, 06:04 »

Please try attached.
I could get the system converged with it in 69 steps.

Please note that exchange_correlation(LSDA or SGGA) must be designated explicitly in the script in order to do spin-polarized calculation properly.

74

Scripts, Tutorials and Applications / Re: sp3-passivation with H for bare Si nanorod

« on: August 27, 2010, 01:47 »

Dear kstokbro,

Thank you for the correction and giving us the example of usage

75

General Questions and Answers / Re: Convergence problem of Heusler alloy

« on: August 26, 2010, 02:44 »

Dear phylyh,

Could you try attached?
I could get your system converged after 23 loops with it
I'm not sure what is the main cause about this issue because I modified several things from yours but anyway it would help you.