February_EDFA_Digital

edfas.org ELECTRONIC DEV ICE FA I LURE ANALYSIS | VOLUME 24 NO . 1 30 optimally assembled in the first commercially available Qnami ProteusQ microscope. DETECTION PRINCIPLE Samples are measured by scanning the diamond tip over a surface while recording the fluorescence of the NV center (Fig. 2a). Thus, information on the topography of the device and its surface magnetic fields are measured simultaneously. This is how the optical detection scheme operates. [2,3] The NV center is optically excited with a green laser (515 nm) and emits red fluorescence light (600 − 850 nm). Amicrowave frequency is applied in order to drive a spin transition of the NV center. The NV fluo- rescence versus applied microwave frequency (Fig. 2b) is recorded. The resulting optically detected magnetic resonance (ODMR) spectrum exhibits a dip at 2.87 GHz if no magnetic field is applied to the NV center. Under the influence of amagnetic field, the dip splits. This splitting is linear with magnetic field. Measuring the splitting locally thus resolves the magnetic field B NV at the surface of a sample. This results in a very high sensitivity, [4] allowing to measure minute fields in the µT range. UNVEILING THE CURRENT FLOW This section describes an example of how to lever- age this excellent magnetic field sensitivity to measure currents with high spatial resolution. An example where current is passed in opposite direction through two par- allel wires is sketched in Fig. 3a. The current generates a magnetic stray field pattern, indicated with gray lines. During the measurement, the Quantilever records simul- taneously the topography (dashed line) and themagnetic field B NV . From B NV the local current density can be readily calculated. Figure 3b shows the resulting data for such a struc- ture. The image shows the measured topography and the induced surface magnetic field B NV , plotted as surface color, while a DC current of I = 250 µA (current density | j | = 2.5 × 10 5 A/cm 2 ) is applied. A rather homogeneous magnetic field inside the loop is detected. The data is used to calculate the local current density | j |, shown in Fig. 3c. The arrows depict the calculated x- and y- component of j . Beside the sanity check that the current actually flows in the wire and that the applied magnitude for | j | is recovered (indicated with an arrow on the colorbar), two interesting details are observed. First, defects in thewire lead toanenhancedcurrent flow in their vicinity, see e.g., the defect at position (x,y) = (4.2,4) µm. Second, the current flow is enhanced in the inner edges of the U-shape and reduced at the outer edges. Such information is obtained due to the high spatial resolution of the technique (down to 50 nmwith state-of-the-art tips) Fig. 3 Operandomeasurement results: DC current. (a) Magnetic stray field pattern of two parallel wires with opposite current flow. (b) A current I = 250 µA (|j| = 2.5 × 10 5 A/cm 2 ) is applied to a Cr/Au (thicknesses 10/90 nm) loop. The plot shows the measured topography and surface magnetic field B NV (x, y). (c) Inferred local current density |j calculated | (x, y) from the data in (b). Vectors depict the spatial components of |j|. (d) Perpendicular component of the magnetic field (|j|, d = 0) at position ♣ (left) and ê (middle) and ( d , |j| = 1 × 10 6 A/cm 2 ) at position ê (right).