Problem when altering temperature

Joshua Gray

unread,

Aug 15, 2022, 8:53:13 PM8/15/22

to DMFTwDFT

Dear developers,

I have successfully managed to run the SrVO3 example, but I am having issues when I try to raise the temperature of the system to that seen in the paper (~110 K). When I set "beta" to 0.01 in the Input.py file in order to set the system at 100 K the code no longer runs. I believe it may be due to the way the number of Matsubara frequency points are generated in sigzero.py.

The if/else statement in lines 102 - 104 in sigzero.py define the integer for the number of frequency points in line 178, but for temperatures greater than ~70 K this if/else statement gives a value of 1 (and temperatures lower than this only give a handful of frequency points). If I manually alter the integer for the number of frequency points to get around this issue I get problems further down the line in other files such as the Delta.out file, where after a value of ~10,000 in the first column I get a line of asterisks (although this may be due to how many decimals are used to write the file).

How did you model the system at 110 K for SrVO3? Any assistance would be greatly appreciated.

Best,

Josh

Yuan Yin

unread,

Dec 13, 2022, 4:37:47 AM12/13/22

to DMFTwDFT

Hello, I guess you need beta=105.5 for a run at 110 K. Because the T (K) =11605/beta. Maybe it is a late response. Hope it helps.

Uthpala Herath

unread,

Dec 15, 2022, 3:24:41 PM12/15/22

to Yuan Yin, DMFTwDFT

Sorry Josua, I missed this message.

Thanks for responding Yuan! That is correct, 110 K corresponds to beta=105.5.

Best,

Uthpala

--
You received this message because you are subscribed to the Google Groups "DMFTwDFT" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dmftwdft+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dmftwdft/2350342c-fa20-485c-846f-15d97b428022n%40googlegroups.com.

--

Uthpala Herath

Postdoctoral Associate

Department of Mechanical Engineering and Materials Science

Duke University

Durham, NC

Tel. (304) 216-2535

Email: uthpala...@duke.edu

Website: https://uthpalaherath.github.io/

Joshua Gray

unread,

Jan 17, 2023, 8:05:20 PM1/17/23

to DMFTwDFT

Hi Uthpala,

I had meant to respond to my own query not long after posting as I realised I was using incorrect units to get the correct beta value. My apologies for not clearing up my query before you both looked into it.