Logo: to the web site of Uppsala University

Graphene with ultrathin metal-oxide layers has emerged as a scalable platform for integrated electronic, spintronic, and neuromorphic devices. The in-plane electric fields in such devices are typically inferred indirectly from transport measurements, which do not provide microscopic details of the interface potential profile. Here, we use photoemission microscopy to uncover the precise electrostatic landscape of a complete device comprising diverse, heterogeneous ultrathin metal-oxide-covered graphene junctions. Using in operando x-ray photoelectron spectroscopy with submicron resolution, we directly image the real potential profile across an entire graphene device, including regions covered by AlO_x and TiO_x nanometer-thick overlayers. Tracking the valence band edge under current bias and gate voltage enables quantitative analysis of internal field evolution via a spectroscopic screening factor, revealing an up to more than 50% suppression of the electric field in graphene covered with oxides, in agreement with transport measurements. This demonstrates local reconfiguration of electric fields in graphene for advanced two-dimensional (2D) device engineering and circuitry, while our operando circuit spectroscopy measurements provide a generic means to uncover intricate potential landscapes and carrier redistribution in 2D electronic and spintronic circuits.

We demonstrate ultrafast graphene spin-field-effect junctions, where gate-tunable superdiffusive spin currents across graphene-ferromagnet interfaces enable electric field control of magnetization dynamics in the ferromagnet. By electrostatically tuning the Fermi level in graphene underlying a cobalt thin film, we modulate the ultrafast spin transport across graphene-cobalt interfaces, reducing femtosecond laser-induced demagnetization time from 203 fs in bare cobalt thin films to 93 fs, a more than 100% increase in the rate of magnetization quenching. Supported by superdiffusive spin transport calculations, our findings unlock field-tunable magnetic speeds in devices, paving the way for innovations in subpicosecond spintronic memory-logic operations. Furthermore, this work creates new possibilities for electrical modulation of spin dynamics and ultrafast spin injection into two-dimensional quantum materials, with potential for nextgeneration quantum sensors and faster magnetic technologies.