Mssbauer Polarimetry with Laboratory Sources and at Synchrotrons

Mössbauer Polarimetry with Laboratory Sources and at Synchrotrons: Applications to Thin-Film Magnetism Dénes Lajos Nagy Wigner Research Centre for Physics, Hungarian Academy of Sciences, Budapest, Hungary 第十二届全国穆斯堡尔谱学会议 12 th Chinese National Symposium on Mössbauer Spectroscopy Changchun, 12 -13 July 2014

Outline · Magnetisation (hyperfine field) direction: definition and significance. · Conventional approach of hyperfine-field- alignment determination in the Mössbauer laboratory and at synchrotrons. · Linear and circular Mössbauer polarimetry in the laboratory; CEMS polarimetry, multilayer applications. · Summary.

Magnetisation alignment and direction In thin films, the alignment/direction of layer magnetisation gives information about: · Basic interactions (coupling, anisotropy) · Spin dynamics (e. g. , magnetisation reversal) Direction (alignment + sign) Alignment The direction of magnetisation is reflected in x the direction of hyperfine field. H All examples will bef taken q from Mössbauer spectroscopy. z y 57 Fe

57 Fe Mössbauer spectroscopy using an unpolarised source x Hhf f = 0 o q f z q Hhf y source 1 sample 1 2 3 4 5 detector 6 Hhf

Line intensities using an unpolarised laboratory source Hhf

Hyperfine splitting of nuclear levels G » 5 ne. V Ehf » 100 ne. V Eg » 14. 4 ke. V Ehf » 100 ne. V 57 Fe

Nuclear resonant scattering of SR: Mössbauer effect with SR · In a laboratory experiment, the bandwidth of the source (≈ 1 ne. V) is much less than the hyperfine splitting (≈ 1 μe. V). Þ Transitions in the sample are excited independently. The resultant spectrum is the incoherent sum of the individual transitions (the intensities are added). · The bandwidth of SR is much larger (≈ 1 me. V) than the hyperfine splitting. Þ All transitions are excited at the same time. The resultant time response is the coherent sum of the individual transitions (the amplitudes are added).

Energy- and time-domain Mössbauer spectra 57 Fe 14. 4 ke. V Temporal beats R. Röhlsberger

Orientation of the hyperfine field (the ”Smirnov figures”) z y E x k B z By E x k E z y B x k 1 2 3 4 5 6

Antiferromagnetically coupled Fe/Cr multilayer Fe Layer magnetisations: Cr Fe Cr Giant magnetoresistance

Antiferromagnetically coupled Fe/Cr multilayer Giant magnetoresistance A. Fert and P. Grünberg Nobel Prize in physics, 2007

Arrangement of an SMR experiment from the high-resolution monochromator z or w Hext Off-specular Specular reflectometry w Q/2 Q-scan: Qx-scan Qz-scan x =d 1/D = 2 p/Q Qx z y x APD E 2 k

AF reflections in time-integral SMR experiments ·For uniaxial antiferromagnets with magnetisation in the x -y plane the quantum beat pattern is always B || y-type! z y ·Can we still see the magnetisation alignment of an antiferromagnet with SMR? ·Yes: the time-integral SMR reflection appears only for B || k! B z. By z B y E x k

Specular SMR: Bulk spin flop in Mg. O(001)[57 Fe(26Å)/Cr(13Å)]20 0 1/2 250 200 150 1 100 Counts 3/2 8 0. 0 0. 2 0. 4 0. 6 0. 8 Q [deg] 1. 0 1. 2 50 12 16 20 4 0 H [m. T] 5 4 1 2 3 ESRF ID 18

Polarised photons: tools for obtaining information on field alignment and sign · Linearly polarised photons may only see the alignment of Hhf. 1 2 3 4 5 6 · Circularly polarised photons may also see the sign of Hhf — possibility for measuring the direction. 1 2 3 4 5 6

57 Fe Mössbauer polarimetry using a magnetised 57 Co( -Fe) source x x j H j q z y q H’ y source 1 2 3 4 5 detector sample 6 1 2 3 4 5 z 6

Linear polarisation (θ = θ’ = 90°) x x j = 0 o j’ = 0 o H’ H q’ q z z y y source 3 4 1 1 4 3 detector sample 3 4 1 1 4 3 9 6 1 1 22 || ^ || 6 22 52 9

Linear polarimetry with an in-plane magnetised 57 Co( -Fe) source H’ f’-f H

Linear (CEMS) polarimeter CEMS Detector B 57 Co( -Fe) k B(s) source Polariser

Linear CEMS polarimetry ·Spectra measured for general values of φ = φ’- φ are linear combinations of parallel- ( φ = 0º) and perpendicular-field ( φ = 90º) spectra. ·By decomposing experimental spectra into parallel- and perpendicularfield components, the inplane alignment of the layer magnetisation can be established with an accuracy of a few degrees. H H’

Linear CEMS polarimetry: bulk spin flop in Mg. O(001)[57 Fe(26Å)/Cr(13Å)]20 F. Tanczikó et al. , NIM B 226 (2004) 461.

SMR AF-peak intensity and linear CEMS polarimetry : coupling and anisotropy in Mg. O(001)[57 Fe(26Å)/Cr(13Å)]20 Easy axis B Nearly hard axis Analyser alignments

Nuclear resonant magnetometry using /4 phase-retarder plate and constant-velocity singleline reference 56 Fe(50 Å)/Cr(11 Å)/57 Fe(50 Å)/Cr(11 Å)/56 Fe(50 Å) C. L’abbé et al. , PRL 93 (2004) 037201.

Circular transmission polarimetry with a 57 Co( -Fe) source magnetised along the optical axis 11 3 3 k H 1 11 H 2 3 4 4 24 24 36 II↑↓ ↑↑ 24 24 80 36 3

Circular (elliptic) Mössbauer polarimetry x x j = 0 o q = 0 o j’ = 0 o q’ = 0 o H H’ z z y y source 3 0 1 1 0 detector sample 3 3 0 1 1 0 3

Determination of q’ by using elliptically polarised source = 0, φ = 0 H q’ k H’

Circular (elliptic) CEMS polarimetry Side window Alternative solution: single-line, circularly polarised source: W. Olszewski et al. , Nucleonica 52, (2007) S 17. F. Tanczikó et al. , Hyp. Int. , 188 (2009) 79. 20 m 57 Fe foil

Elliptical polarisation l l General case: the photons are elliptically polarized and the line intensities depend on q, q ’ and df = f’-f. Only four intensities, i. e. , three intensity ratios are independent. 9 I 3 3 I 2 I 1(q, q’, df) 6 I 3 I 2(q, q’, df) 4 I 2 I 3(q, q’, df) I 4(q, q’, df)

Elliptical polarisation The basic spectra I||, I┴, I↑↑, I↑↓ of linear and circular polarimetry form an alternative linearly independent base for the line intensities:

Elliptical polarisation Conversely, the angles can be expressed with normalized intensities (I||+I┴+I↑↑+I↑↓ = 1 ): or F. Tanczikó et al. , Hyp. Int. 188 (2009) 79.

The direction of the hyperfine field: Experimental parameters H 1 q, q’ @ 10° df @ 0° H 2 Fit results q, q’ @ 9° df @ 0° H 1 H 2

The direction of the layer magnetisation Parallel fields The direction of the hyperfine field: Experimental parameters q @ 10° q’ @ 5° Antiparallel fields df @ 0° Fit results q @ 10° q’ @ 4. 9° df @ 0°

Magnetometry by circular CEMS polarimetry AF-coupled asymmetric Fe/Cr trilayer Fe(20 nm)/Cr(1. 2 nm)/57 Fe(5 nm)/Cr(2. 4 nm) After. Spin-flop field reversal, in remanence Magnetisation reversal Cr 57 Fe Cr Bpolariser Fe Mg. O BB extext F. Tanczikó et al. , to be published B = 20 m. T Remanence Bext = 100 m. T

Summary · Full information on the direction of the hyperfine magnetic field can be obtained both from laboratory and from synchrotron Mössbauer experiments when using (preferably fully) polarised photons. · Full (elliptic) polarisation in laboratory experiments may be best achieved by using a magnetised, hyperfine-split ferromagnetic source (e. g. , 57 Co( -Fe)) linear and circular polarimetry can be realised. · For thin-film studies efficient conversion-electron Mössbauer polarimeters were built both in linear and in circular regime and were used for studying spin-flop and magnetization reversal phenomena in AF-coupled Fe/Cr multilayers, resp.

感谢您的关注！ Thank you for your attention!