Multiple Frequencies in

Transverse Field

FFT exercise for TSI 2011 Data Analysis workshops

More often than not, when a material with large paramagnetic moments is studied in high transverse field (HTF) with µSR, several frequencies of µ+ precession are observed. This is conventionally interpreted as evidence for several muon sites in a magnetically ordered crystal lattice; however, this interpretation is hard to accept when the sample is not magnetically ordered. Then the usual explanation is multiple magnetically inequivalent sites with different (frequently quite large) Knight shifts, another "stretch" of plausibility. Some think the µ+ may drive a region around the muon into a magnetically ordered state (an alarming proposition for those who like to think of the µ+ as a "gentle probe"); others claim that the µ+ is bound to a pre-existing "magnetic polaron" (a species prdicted by de Gennes over half a century ago). This is a controversial topic, so it behooves anyone wishing to study magnetic materials with µSR to be familiar with using the fast Fourier transform (FFT) to display these frequency spectra as a function of T.

A good first attempt at producing the desired FFTs for TRIUMF µSR runs can be achieved on any Web browser using the MUD Runs Database at http://musr.ca/mud/, but a publication-quality 3D-like plot is best produced using the fft3d program on bnqrexp.triumf.ca (unfortunately this version is not available on the Hennings workstations). This program is based on the same msrfft program used by the Web utility, but interfaced to the TRIUMF GPLOT graphics library, which can make editable plots in EDGR format.

The Experiment

Muons were stopped in a spherical ball pressed from a powder of the magnetic semiconductor samarium sulphide (SmS) in the HiTime spectrometer running in Sample/Reference (S/R) mode in various high transverse fields (HTF) at various temperatures (up to 875 K). Two distinct oscillation frequencies are observed in the Sample Spectrum.

The Exercise

Visit the TRIUMF µSR runs database at http://musr.ca/mud/runSel.html and search for all runs with "SmS%3T" in the title.

Click on the run-number button for the high-temperature run 1152 and investigate the run header information, especially the names and ordering of the data histograms. Select various Spectrum types and look at the range from 0 to 0.1 µs with 0.2 ns bins. Eventually choose Spectrum type 4 (Histograms 1,3,2,4) with a RRF frequency of 406 MHz and a packed bin size of 0.05 µs to display the whole time range in the RRF.

Now click on "Plot FFT" and adjust the FFT parameters and options until you get a sharp, symmetric line in the FFT, plotting the Real part of the amplitude. (Allow pop-ups in your browser!) Do this for type 4 spectra which histograms 1,3,2,4 for the Sample signal; then repeat with histograms 5,7,6,8 to see the Reference signal. Then click on a few other runs at different Ts and see if their FFTs look different.

If you like, you can move on to the Advanced Fitting Exercise on this dataset.


Jess H. Brewer
Last modified: Sun Aug 14 22:56:47 PDT 2011