Omnidirectional magnetic imaging of magnetic anisotropy and phase transitions

Abstract

Micron scale imaging of magnetic fields is an important tool for understanding the evolution of magnetism through phase transitions and as a result of interactions inside of heterostructures. However, most imaging platforms, like the nitrogen-vacancy (NV) centre in diamond, are restricted to applying magnetic fields along the quantisation axis of the quantum sensor. This greatly restricts the utility of these systems for exploring materials that emit strong fields or exhibit variable response with respect to the applied field direction. Here we explore an alternative approach using weakly coupled spin-pairs in hBN that exhibit a spin-1/2-like behaviour and an isotropic response to magnetic field. We demonstrate that the spin-pair system can operate in the presence of strong fields from a thin film magnet which were incompatible with NV diamond imaging even with applied fields along the quantisation axis. Further, we demonstrate that using this platform allows for imaging with an arbitrary applied magnetic field direction, allowing us to probe the anisotropy and spin-reorientation transition in the ferrimagnet TbMn$_6$Sn$_6$. Finally, we propose an improved geometry for imaging small anisotropy contributions such as crystalline anisotropy. These results demonstrate how this or similar spin-1/2 systems might be used for imaging magnetic materials that are incompatible with other techniques despite the reduction in sensitivity compared with NV in diamond imaging.

Figures (7)

• Figure 1: Introduction to spin-pair imaging. (a, b) Illustration of the crystal structure of (a) the NV centre in diamond and (b) the optical spin-pair in hBN. (c) Comparison of the spin contrast of the NV and spin-pair for different magnetic field directions with a magnitude of $|B| = 75$ mT. (d) Illustration of the vector magnetic field components projection onto the NV quantisation axis and the reorientation of the spin-pair to match the combined field. (e) Magnetic image taken with an ensemble of NV centres in the presence of a strong magnetic material that quenches the spin contrast near it (top panel) with ODMR spectra from the marked positions (bottom panel). (f) Similar magnetic sample to the one imaged in (e) but now with the spin-pair in hBN which exhibits a much larger dynamic range to strong external fields. Taken at $T = 250$ K with an OOP field of $B=75$ mT.
• Figure 2: Imaging of a temperature induced spin-reversal transition. (a) Exemplary magnetic images of the inplane magnetic field component switching during the spin reversal transition. (b, c) Extracted temperature curves from calculating the average magnetic field around the magnetic material for both (b) IP and (c) OOP fields. Due to the strong magnetic field gradients the regions from directly above the thin-film were not included in the OOP case. (d) The mean full width half maximum of the hBN SP ODMR over the sample as a function of temperature, where the error bars represent the standard deviation.
• Figure 3: Magnetic imaging of in plane anisotropy Magnetic images of the inplane magnetic field from the TMS thinned single crystal for various magnetic fields within the plane (taken at $T=315$K). Insert: simulations of the expected magnetic field for a perfect well-ordered thin film with no in-plane crystalline anisotropy. The extracted magnetic field from around the TMS thinned crystal (blue) and above (orange) for the range of magnetic field angles with comparison to the simulation (solid lines). Grey regions around the x-axis correspond to regions were RF to spin coupling is expected to be bad and thus are less reliable.
• Figure 4: Proposal for accurate crystalline anisotropy imaging. (a) Illustration of RF delivery for in-plane measurements with a spin-pair material and an optimised material shape for minimising shape based artifacts and anisotropy. (b, c) Simulation of the spin-pair signal with $B=75$ mT and a signal of $B_M=1$ mT for (b) no anisotropy and (c) with a single vertical crystalline anisotropy. (d) Simulated magnetic field from around the material with different degrees of magneto-crystalline anisotropy, assuming instant snapping to the new axis with no dragging.
• Figure 5: (a) Scanning electron microscope image of TbMn$_6$Sn$_6$ film that was placed on the diamond in Fig. \ref{['fig:intro']} after the focused ion beam removal the surrounding material. (b) Optical image of the TbMn$_6$Sn$_6$ film encapsulated with the hBN MOVPE sensing layer used throughout the manuscript.