KKR workshop

Canterbury, 4-6 July 2008

The purpose of this workshop is to bring together the groups who are developing KKR codes. One aim is to get an overview of the codes which are available, and compare their functionalities and also technical implementations. A particular issue will be how to obtain reliable total energies from KKR. It is planned to collect some benchmark results for selected systems from all codes before the workshop which shall be used to identify the best methods. Other topics, such as the treatment of correlations in KKR, calculation of susceptibilities, etc., will also be discussed.

Programme
The KKR forum
Benchmark calculations
Travel information

Preliminary Programme:

Friday:

14:00 - 14:30 Registration and Welcome
14:30 - 15:00 Martin Lueders: Overview of burning KKR issues
15:00 - 15:30 Hubert Ebert: The Munich SPR-KKR package - recent developments and applications

15:30 - 16:00 Coffee

16:00 - 16:30 Rudi Zeller: Iterative solution of the TBKKR Dyson equation for large systems
16:30 - 17:00 Pawel Buczek: HUTSEPOT environment & recent developments in multiple scattering theory
17:00 - 17:30 P. Weinberger: tba

Saturday

9:30 - 10:00 Budapest: Balazs Ujfalussy or Laszlo Szunyogh
10:00 - 10:30 Malcolm Stocks: tba

10:30 - 11:00 Coffee

11:30 - 12:00 Julie Staunton: tba
12:00 - 12:30 Hisazumi Akai: Calculation of spin transport using the KKR Kubo-Greenwood formula

14:30 - 15:15 5 minute poster presentations
15:15 - 16:30 Posters
16:30 - 17:00 Peter Zahn: SPIN-FLIP SCATTERING IN METALS: IMPURITIES IN CU
17:30 - 18:00 Robert Hammerling: Near field corrections to electrostatic potential in KKR

Sunday

9:30 - 10:00 Walter Temmerman: Summary
10:00 - 11:15 round table discussion
11:15 - 12:00 Balazs Gyorffy: The future and the KKR

The KKR forum

We set up a "KKR forum" on the ccpforge webserver. Ccpforge is aimed to provide a sourceforge like environment for the codes, developed within the collaborative computational projects (CCPs) in the UK. The server provides file space, discussions forums, a wiki, a mailing list, etc. This technology might be useful for collecting and distributing experiences within the KKR community. Also we would like to collect some benchmark results from different codes on this server, so that they become accessible to other KKR developers.

If you are interested in using this service to discuss with other KKR developers and/or share information such as benchmark results, then simply register on CCPforge (clock on "New account" in the top right corner) and, once registered, request to join the KKR-forum project. Please feel free to contact me (m.lueders_AT_dl.ac.uk) if you have any problems with the site.

Over the next days I will provide more information on the forums wiki and describe how to upload the benchmark results there.

Benchmark calculations

One aim of the workshop is to establish that different KKR codes agree in their results, and also how techniques beyond muffin-tin or ASA improve the results.

1. Consistency check:

To establish whether different codes produce consistant results when run in the same mode, we would like to gather total energies for Cu and Ce as a function of the lattice constant for muffin-tin geometry and ASA both with l-max=3 (up to f)

2. Best possible calculation for Cu and Ce:

Please try to get the best possible answer for the equilibrium lattice constant for Cu and (alpha-) Ce from your code.

3. l-max convergence:

Please calculate the total energies and magnetic moments of MnO (lattice constant a=8.4 a.u.) for the AF II structure and a ferromagnetic structure as a function of the angular momentum cut-off.

Additional suggestions are welcome.

Travel information

The workshop takes place in the Canterbury Cathedral Lodge. Canterbury can be reached by air via: (information taken from Kent university web pages)

London Heathrow:
either London Underground to London Victoria railway station (50 mins), train to Canterbury East (85 mins)
or London Underground to London Charing Cross railway station or London Waterloo East railway station (50 mins), train to Canterbury West (90 mins).
London Gatwick:
Gatwick Express to London Victoria (30 mins four times each hour), train to Canterbury East (85 mins).
London Stansted:
either Train to Tottenham Hale station (30mins), London Underground to London Victoria (25 mins), train to Canterbury East (85 mins)
or Stansted Express to London Liverpool Street railway station (45 mins), London Underground to London Charing Cross railway station or London Waterloo East railway station (20 mins), Train to Canterbury West (90 mins)
London Luton; Train to London Bridge railway station (50 mins), train from London Bridge to Canterbury West (90 mins).

London Underground - One Railway - Stansted Express -Gatwick Express - Thameslink Trains - National Rail Enquiries - South Eastern Trains

Click here for an map of the Canterbury city centre.

Hotel information