Bi2Se3 interlayer treatments affecting the Y3Fe5O12 (YIG) platinum spin Seebeck effect
Spin Seebeck effects (SSE) arise from spin current (magnon) generation from within ferri-, ferro-, or anti-ferromagnetic materials driven by an applied temperature gradient. *
Longitudinal spin Seebeck effect (LSSE) investigations, where the spin current and temperature gradient evolve along a common z axis, while the magnetic field is applied in the y axis and the voltage contacts are spaced along the x axis, have become the most popular spin Seebeck device architecture. *
In article “Bi2Se3 interlayer treatments affecting the Y3Fe5O12 (YIG) platinum spin Seebeck effect”, Yaoyang Hu, Michael P. Weir, H. Jessica Pereira, Oliver J. Amin, Jem Pitcairn, Matthew J. Cliffe, Andrew W. Rushforth, Gunta Kunakova, Kiryl Niherysh, Vladimir Korolkov, James Kertfoot, Oleg Makarovsky and Simon Woodward present a method to enhance the longitudinal spin Seebeck effect at platinum/yttrium iron garnet (Pt/YIG) interfaces. *
The introduction of a partial interlayer of bismuth selenide (Bi2Se3, 2.5% surface coverage) interfaces significantly increases (by ∼380%–690%) the spin Seebeck coefficient over equivalent Pt/YIG control devices. *
Optimal devices are prepared by transferring Bi2Se3 nanoribbons, prepared under anaerobic conditions, onto the YIG (111) chips followed by rapid over-coating with Pt. The deposited Pt/Bi2Se3 nanoribbon/YIG assembly is characterized by scanning electron microscope. The expected elemental compositions of Bi2Se3 and YIG are confirmed by energy dispersive x-ray analysis. *
A spin Seebeck coefficient of 0.34–0.62 μV/K for Pt/Bi2Se3/YIG is attained for the authors’ devices, compared to just 0.09 μV/K for Pt/YIG controls at a 12 K thermal gradient and a magnetic field swept from −50 to +50 mT. *
Superconducting quantum interference device magnetometer studies indicate that the magnetic moment of Pt/Bi2Se3/YIG treated chips is increased by ∼4% vs control Pt/YIG chips (i.e., a significant increase vs the ±0.06% chip mass reproducibility). *
Increased surface magnetization is also detected in magnetic force microscope studies of Pt/Bi2Se3/YIG, suggesting that the enhancement of spin injection is associated with the presence of Bi2Se3 nanoribbons. *
To understand the surface magnetization effects in sample BSYIG1-a further, magnetic force microscope (MFM) measurements were undertaken using a commercial atomic force microscope and magnetic NanoWorld Pointprobe® MFMR AFM probes. *
MFM differs from traditional atomic force microscopy in that the AFM probe, in addition to providing a surface height profile, is also able to detect the magnetic field gradient above the sample. *
MFM surface profiling of BSYIG1-a revealed that a typical ribbon is comprised of multilayers of Bi2Se3, providing thicker sections ca. 250 nm thick [e.g., the profile along vector 1 in Figs. 3(a) and 3(b) cited below] and additional thinner sections ca. 100 nm thick [e.g., the profile along vector 2 in Figs. 3(a) and 3(b)]. Re-running ribbon profiles 1 and 2 with the magnetic probe at a height of 100 nm above the topological surface provided data on the magnetic field gradient variation along the same line profiles. The MFM amplitude [Figs. 3(c) and 3(d) cited below] increases over the Bi2Se3 flake, and furthermore, the magnetic enhancement correlates with the thickness of the Bi2Se3, being larger for the thicker part of the sample. *
This amplitude enhancement suggests that the observed effect is magnetic rather than due to long-range electrostatics, supporting the inference that the surface magnetization is improved by the presence of Bi2Se3 flakes at the interlayer of a Pt/YIG device. However, it was not possible to extract quantitative information about surface magnetization from this study, but Yaoyang Hu et al. are hopeful that future experimental and theoretical work can provide further explanation. *
Please have a look at the @NanoWorld blog for the full citation and a direct link to the full article.