QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: asanchez on January 21, 2016, 19:22

Title: Problem when calculating InelasticTransmissionSpectrum
Post by: asanchez on January 21, 2016, 19:22: I've been trying to calculate the InelasticTransmissionSpectrum but get an error just when it gets to that very point (i.e. DynamicalMatrix & HamiltonianDerivatives calculations finish without errors).

Just when the output gets to this point:

Quote
+------------------------------------------------------------------------------+
| Calculating inelastic transmission spectrum |
+------------------------------------------------------------------------------+
InelasticTransmissionSpectrum: k-index (1 / 1), q-index (1 / 1)

Calculation stops with the following error written to stderr:

Quote
mypath/QuantumWise/VNL-ATK-2015.1/bin/atkpython: line 11: 2949 Illegal instruction $EXEC_DIR/atkpython_exec "$@"

Attached is the input file I'm using. I've tried using LOE instead of XLOE and using a DFT calculator. Also tried with an applied bias of 25mV. Same error every time. Can provide nc files if needed (too large to attach here)

Would someone please take a look see why I'm getting this? thanks in advance
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: Jess Wellendorff on January 22, 2016, 08:51: We will have a look at this. NC files would be very helpful. You can email them to [email protected], or use some online repository of your choice.
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: Anders Blom on January 22, 2016, 09:58: Did you try it with more than 1 k and q points? Just an idea...
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: asanchez on January 22, 2016, 13:01: I did not try more than 1 k/q-point as it's a NW structure (more like a chain, really). Should I?

I'll email you the NC files Jess, thanks you both for your help.
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: Anders Blom on January 22, 2016, 13:02: There is still dispersion along the wire if it's 1D, but ok not in A and B, no. I don't know, it was just a thought. Segfaults are rare in ATK and usually indicate a real problem, so having access to the file will be helpful for us to get to the bottom of it.
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: asanchez on January 22, 2016, 14:03: I've sent the NC files to Jess. (posted email address didn't work, added a dot between name and surname and seemed to work. hopefully you received the email correctly)
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: Jess Wellendorff on January 26, 2016, 10:34: I cannot reproduce the error you find. Because I do not have access to the DFTB parameters for Cu that you have used, I have changed all atoms in your device to Au and run the calculation with ATK 2015.1 and Hotbit parameters for Au. The calculation runs through just fine (script attached).

BTW: Any particular reason you choose a 100 Ha density cut-off instead of the default 10 Ha?
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: asanchez on January 26, 2016, 11:57: What about the DFT calculation? Can't reproduce it with that either? I suppose I could run it again with some freely available DFTB parameters and send you the NC file.

No idea why that 100*Ha is there. Will check how it got there.
Title: Re: Problem when calculating InelasticTransmissionSpectrum
Post by: asanchez on January 26, 2016, 16:13: It's apparently something to do with the cluster I'm using. I tried your Au script and got the same error as before. Ran it on a different cluster and it ran fine. No idea why it's doing that. All other calculations I've attempted run fine.

Thanks for your help, Jess.