Supporting Information for: Transformation in VO 2 Thin Films by

Transcription

1 Supporting Information for: Imaging Nanometer Phase Coexistence at Defects During the Insulator-Metal Phase Transformation in VO 2 Thin Films by Resonant Soft X-Ray Holography Luciana Vidas,, Christian M. Günther, Timothy A. Miller, Bastian Pfau, Daniel Perez-Salinas, Elías Martínez, Michael Schneider, Erik Gührs, Pierluigi Gargiani, Manuel Valvidares, Robert E. Marvel, Kent A. Hallman, Richard F. Haglund Jr, Stefan Eisebitt,, and Simon Wall, ICFO - Institut de Ciències Fotòniques, The Barcelona Institute of Science and Technology, Castelldefels (Barcelona), Spain Institut für Optik und Atomare Physik, Technische Universität Berlin, Berlin, Germany Max-Born-Institut, Berlin, Germany ALBA Synchrotron Light Source, E Cerdanyola del Vallès, Barcelona, Spain Department of Physics and Astronomy, Vanderbilt University, Nashville, Tennessee , USA luciana.vidas@icfo.eu; simon.wall@icfo.eu 1

2 Sample fabrication VO 2 samples, 75 nm thick, were deposited, using pulsed laser deposition (PLD), onto Si 3 N 4 membranes featuring a 1.2 µm thick Cr(5 nm)/au(55 nm) multilayer layer on the reversed side. Prior to deposition, the chamber of an Epion PLD-3000 system was pumped down to 90 µtorr and the vanadium metal target was ablated to remove surface contamination. The deposition was performed at pressure of 11 Torr while maintaining a 2 sccm ow of ultra-high purity oxygen gas. The sample was 8 cm from the target during deposition. The VO 2 thin lms were deposited by ablating the target with a KrF excimer laser (wavelength 248 nm). The beam (4 J cm 2 per pulse, 25 Hz repetition rate, and 25 ns pulse duration) was rastered across the rotating vanadium metal target during deposition. After deposition, the samples were annealed in a tube furnace at 450 C, while the O 2 ow rate (1922 sccm) was actively adjusted to maintain 250 mtorr. After deposition, the gold multilayer was used to dene the holography mask. A focused ion beam was used to mill an aperture of 2 µm in diameter through the multilayer to enable transmission of the X-rays. Additionally, three reference holes with exit diameters of 5090 nm, were drilled through the gold, Si 3 N 4 and VO 2 layers. The SEM image of the sample used in Figures 3 and 4 of the main text is shown in Figure S1. Figure S1: SEM image of the second sample used in Figures 3, 4 and Supplementary Figure S4. Field of view is 2 µm in diameter. 2

3 Optical measurements The samples were characterized by means of optical measurements. The hysteresis curves were measured on a witness sample grown under the same conditions as specied above, but without the gold masks used for holography. The optical transmission was measured with white-light from a tungsten-halogen lamp ( nm) and detected with an amplied InGaAs PIN. A hysteresis measurement is shown in Figure S2, which is typical of high quality VO 2 lms. In addition, the samples were characterized by Raman scattering in order to demonstrate that their initial state is the M 1 phase, as shown in Figure S3. Transmission (a.u.) Heating Cooling Temperature (K) Figure S2: Temperature-dependent optical transmission of a VO 2 sample. The hysteretic behaviour shows a width of 6 K. 350 X-ray measurements Measurements were carried out at the UE52-SGM beamline of the BESSY II synchrotron radiation source at Helmholtz Zentrum Berlin using the ALICE X-ray scattering instrument. 1 The instrument is equipped with a cryostat and resistive heaters which were used to control the sample temperature with a stability of 0.1 K. The charge-coupled device (CCD) camera was placed approximately 40 cm away from the sample. This geometry corresponds to a 3

4 Intensity (a.u.) M 1 (on Si) Raman Shift (cm -1 ) Figure S3: Raman spectrum of a VO 2 sample at room temperature that manifests the monoclinic M 1 structure. The saturated peak at 500 cm 1 corresponds to the Silicon of the substrate. maximum detectable in-plane momentum transfer of about 90 µm 1 and a maximum spatial resolution of 41 nm. XAS was measured via transmission of the thin lm in the areas not closed by the gold mask with the incident X-ray beam parallel to the sample normal. Fourier transform holograms were recorded in the same transmission geometry. All images shown are taken from the real part of the sample's exit wave as reconstructed by taking the Fourier transformation of the hologram. A beam block was used to block the highly intense central beam in order to adapt the dynamic range of the scattering pattern to the capabilities of the detector. The beam block acts as a low pass lter and essentially subtracts the average transmission of the entire sample. As a result, the absolute transmitted intensity for each phase depends on the average phase of the whole sample, which is changing as the phase fraction grows. However, although the absolute values change, the contrast between phases remains. As a result, the threshold used in Figure 4 of the main text is manually adjusted to ensure the contrast is not lost for each temperature. In addition, an upper threshold is applied to the blue and green channels in order to move the white regions (corresponding to defects and boundaries of the crystals seen in Figure 3) into the red channel in Figure 4. 4

5 c-axis Figure S4: Polarization resolved imaging on VO 2. X-ray dichroic images of VO 2 lms measured at the d peak (530.5 ev) at room temperature. The X-ray polarization was rotated by 180 degrees in 5 degree steps (not shown). When the X-rays are parallel/perpendicular to the rutile c-axis, the dichroic (dierence) image is strongest due to the anisotropy of the d state (top row). This dierence is lost when each polarization is rotated by 45 degrees with respect to the c-axis (bottom row). The dashed red line corresponds to the ROI used in Figures 3 and 4. Field of view is 22 µm in diameter. Polarization-dependent images and spectra were measured in the MaReS end station at BOREAS beamline at the ALBA synchrotron. 2 The full polarization dependence of the images is shown in Figure S4. References (1) Abrudan R., Brüssing F., Salikhov R., Meermann J., Radu I., Ryll H., Radu F., Zabel H.:`ALICE An advanced reectometer for static and dynamic experiments in magnetism at synchrotron radiation facilities', Rev. Sci. Instrum.., 2015, 86, pp (2) Barla A., Nicolás J., Cocco D., Valvidares M., Herrero-Martín J., Gargiani P., Moldes J., Ruget C., Pellegrin E., Ferrer S.: `Design and performance of BOREAS, the beamline for resonant X-ray absorption and scattering experi- 5

6 ments at the ALBA synchrotron light source', J. Synchrotron Radiat.., 2016, 23, pp