[DFTB-Plus-User] R: R: R: negf calculation

Fri Apr 27 16:51:33 CEST 2018

mpifx is part of dftb+  (is a nice fortran interface to mpi)

In order to compile the new code you need to follow the instructions in INSTALL.rst

Go to line 50. 

git clone (or download of the sources)

cd dftbplus

git submodule update  --init  --recursive

the latter downloads the submodules mpifx and scalapackfx

Hope this helps

Alex

Da: DFTB-Plus-User [mailto:dftb-plus-user-bounces at mailman.zfn.uni-bremen.de] Per conto di Theodoros Leontiou
Inviato: venerdì 27 aprile 2018 16:26
A: User list for DFTB+ related questions
Oggetto: Re: [DFTB-Plus-User] R: R: negf calculation

Dear Alessandro,

I have tried to compile the parallel negf version at the link you suggested at

https://github.com/aradi/dftbplus

However, I get errors which to the best of my knowledge are due to missing files.

For example I get "*** No rule to make target .... etc etc "  when try to compile mpifx.

The directory seems to empty of any source files.

Should I be able to compile this code yet? 

I had a look at INSTALL.rsf but the instructions are not for the negf version.

I appreciate your help and time

Theodoros

---- Original Message ----
From: "Theodoros Leontiou" <eng.lt at frederick.ac.cy>
To: "User list for DFTB+ related questions" <dftb-plus-user at mailman.zfn.uni-bremen.de>
Sent: Thu, Apr 26, 2018, 12:39 PM
Subject: Re: [DFTB-Plus-User] R: R: negf calculation

Dear Alessandro,

I have downloaded the version at the dftb+ website and used the examples also found there.

These is the same format also discussed at dftb+ recipes

http://www.dftbplus.org/fileadmin/DFTBPLUS/public/recipes-negf/html/index.html

I will download and try the beta version as you suggested.

Thanks

Theodoros

---- Original Message ----
From: "Alessandro Pecchia" <alessandro.pecchia at ismn.cnr.it>
To: "'User list for DFTB+ related questions'" <dftb-plus-user at mailman.zfn.uni-bremen.de>
Sent: Thu, Apr 26, 2018, 10:29 AM
Subject: [DFTB-Plus-User] R: R: negf calculation

Hi,

I think you are using an old version of the manual or I really have to look at it (!)

The definition of the Transport block is outdated.

You should move FermiLevel in each contact block as well as Potential (one per contact).

The energy range must be specified in the block 

Analysis{ 

   TunnelingAndDos{ …here… } 

}

What version of the code are you using? 

You could download and test the beta version from github:

https://github.com/aradi/dftbplus

there is a parallel-negf branch we are currently merging for a forthcoming official release. 

We are happy if you can test it and report problems. 

Alessandro

Task = UploadContacts {

    FirstLayerAtoms =  1   61   121   181   241   301   361   421   481   541 

    FermiLevel [eV] = SetForAll { -3.112636 }

    ContactPotentials {

      source = 0.0

      drain [eV] = 0.0

    }

    EnergyRange [eV] = -6.0  -1.0

    EnergyStep [eV] = 0.02 

  }

}

Again many thanks for your help!

Regards

Theodoros                                                                                                                                                                       

---- Original Message ----
From: "alessandro.pecchia" <alessandro.pecchia at ismn.cnr.it>
To: "User list for DFTB+ related questions" <dftb-plus-user at mailman.zfn.uni-bremen.de>
Sent: Wed, Apr 25, 2018, 11:38 PM
Subject: [DFTB-Plus-User] R: negf calculation

Hi,

 If I interpret correctly the problem is that the Geometry{} block has been removed in newest versions, so please remove it from your input. The parser expects Device in the Transport block.

Thanks for signaling this problem.

Regards,

Alessandro

Inviato dal mio dispositivo Samsung

-------- Messaggio originale --------
Da: Theodoros Leontiou <eng.lt at frederick.ac.cy> 
Data: 25/04/18 8:05 PM (GMT+01:00) 
A: dftb-plus-user at mailman.zfn.uni-bremen.de 
Oggetto: [DFTB-Plus-User] negf calculation 

Dear dftb+ specialist

I am trying to do my first dftb+ negf calculation.

However, I get an error I don't know how to resolve. 

I am running a test using the input files provided for the Silicon nanowire 

example. When I try to calculate the contact hamiltonian I get the

following:

ERROR!

-> Missing child: Device

Path: dftb_in/Transport

Line: 6-23 (File: dftb_in.hsd)

What is the meaning of the above error message?

I am grateful for your help

Regards

Theodoros

ps: I am attaching my input file below

####### BEGINNING OF INPUT FILE ##########################

Geometry = GenFormat {

<<< 'SiNW.gen'

}

Transport {

  Geometry {

    Device {

     AtomRange = 1 600

    }

    Contact {

      Id = "source"

      AtomRange = 601 720

    }

    Contact {

      Id = "drain"

      AtomRange = 721 840

    }

  }

  Task = ContactHamiltonian{

  ContactId = "source"

  }

}

Driver = {}

Hamiltonian = DFTB {

  SCC = Yes

  SCCTolerance = 1e-7

  MaxSCCIterations = 20

  MaxAngularMomentum = {

    Si ="p"

    H = "s"

  }

  Filling = Fermi {

    Temperature [Kelvin] = 0

  }

  SlaterKosterFiles = Type2FileNames{

    Prefix = "/home/leontiou/Dropbox/research/tight_binding/dftb+/dftb+mpi-negf.r4732/examples/SiNW/"

    Separator = "-"

    Suffix = ".skf"

  }

  KpointsAndWeights = SupercellFolding{

    8 0 0

    0 1 0

    0 0 1

    0.50 0.0 0.0

  }

Electrostatics = GammaFunctional {}

Eigensolver = DivideAndConquer{}

}

################ END OF INPUT FILE ##########################

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20180427/2057fd0f/attachment.htm>