Upgrade of a low-temperature scanning tunneling microscope for electron-spin resonance

TL;DR

This paper details the upgrade of a helium-3 STM to enable electron spin resonance measurements with high-frequency RF capabilities, pulsed operation, and synchronization for advanced spin spectroscopy at the atomic scale.

Contribution

The upgrade introduces RF delivery up to 30 GHz, pulsed operation modes, and synchronization techniques, enhancing the capabilities of ESR-STM for atomic-scale spin studies.

Findings

Successful magnetic field sweep ESR on TiH/MgO/Ag(100)

Achieved RF power delivery up to 30 GHz

Demonstrated synchronization of RF and DC pulses

Abstract

Electron spin resonance with a scanning tunneling microscope (ESR-STM) combines the high energy resolution of spin resonance spectroscopy with the atomic scale control and spatial resolution of STM. Here we describe the upgrade of a helium-3 STM with a 2D vector-field magnet ($B_z$ = 8.0 T, $B_x$ = 0.8 T) to an ESR-STM. The system is capable of delivering RF power to the tunnel junction at frequencies up to 30 GHz. We demonstrate magnetic field sweep ESR for the model system TiH/MgO/Ag(100) and find a magnetic moment of $(1.004 \pm 0.001)$ $\mu_B$. Our upgrade enables to toggle between a DC mode, where the STM is operated with the regular control electronics, and an ultrafast-pulsed mode that uses an arbitrary waveform generator for pump-probe spectroscopy or reading of spin-states. Both modes allow for simultaneous radiofrequency excitation, which we add via a resistive pick-off tee to the bias voltage path. The RF cabling from room temperature to the 350 mK stage has an average attenuation of 18 dB between 5 and 25 GHz. The cable segment between the 350 mK stage and the STM tip presently attenuates an additional $34_{-3}^{+5}$ dB from 10 to 26 GHz and $38_{-2}^{+3}$ dB between 20 and 30 GHz. We discuss our transmission losses and indicate ways to reduce this attenuation. We finally demonstrate how to synchronize the arrival times of RF and DC pulses coming from different paths to the STM junction, a prerequisite for future pulsed ESR experiments.