Here we explore the interactions between mechanical, magnetic resonators and gaseous neutral atoms.  We study two different regimes: in the first, the atoms are confined in a buffer gas vapor cell, and in the second we use trapped, laser-cooled atoms in a UHV chamber.  In the long term, the novel coupled atom-resonator system could be used in experiments involving degenerate quantum gases and even ground-state mechanical resonators.   Such a hybrid quantum system could be useful for testing quantum mechanics at unexplored scales or for quantum information processing.

This experiment constitutes the first experimental observation of a direct resonant coupling between the motion of a mechanical resonator and atomic spins. This coupling, mediated by a magnetic particle attached to the tip of a micrometer-scale cantilever, excites a coherent precession of the atomic spins about a static magnetic field.

A commercial atomic force microscope (AFM) cantilever with magnetic tip (upper left, Fig. 1) is mounted on a piezoelectric transducer (PZT) and positioned near a ∼1 mm3 volume micro-fabricated Rb vapor.  The atoms in the magnetically shielded vapor cell experience the combined magnetic field of an external bias field Bext and the magnetic field from the cantilever tip (see Fig. 2).

When the PZT drives the cantilever at its resonance frequency, its motion creates a time-varying component of the magnetic field.  This causes a coherent precession of atomic spins for those atoms in the “resonant slice,” which happen to have a Larmor frequency matching the cantilever resonance frequency.  The diagram at left in Fig. 2 shows one such atom in this resonant slice.  The spin precession is detected by observing the resulting absorption modulation of a circularly polarized probe laser beam (which also serves to optically pump the atoms) as it propagates through the atomic vapor.  A photograph of the experimental setup is shown at right in Fig. 2.

Fig. 3 shows direct evidence of the coupling of the cantilever motion to the atom spins.   The plot shows the amplitude of the probe laser modulation derived from a lock-in amplifier as a function of the driving frequency of the PZT.  The narrow (9 Hz width) resonance is due to the cantilever motion, while the broader background resonance from the oscillating field generated by the current driving the PZT.  The narrow peak is centered at the cantilever resonance frequency (14.3 kHz), showing that the cantilever is exciting the desired magnetic resonance.

In this experiment, we move forward into the regime of trapped, laser-cooled atoms. Instead of a vapor cell, we magnetically trap the atoms near the tip of the cantilever.  This allows greater flexibility in tuning the interactions between the atoms and cantilever and takes advantage of the highly localized interaction with the cantilever tip.  The cantilever can be used to excite spin resonances in the atoms, but there are other possibilities as well.  One could do the reverse experiment and observe a motional resonance of the cantilever due to an external modulation of the atom spins.  Microresonators potentially could be useful for localized atomic internal state control and detection on an atom chip.  The magnetic interaction with the atom cloud could also serve as a way to cool the cantilever motion.  

Figure 4 shows the experimental arrangement for the cold atom system.  Atoms are collected in an upper MOT chamber and transported vertically into a differentially pumped lower chamber containing the magnetic cantilever.  After transport, the atoms are initially confined in a large magnetic quadrupole trap and then will be transferred to a small trap formed near the end of the cantilever.  This final quadrupole trap is formed by applying a uniform bias field opposing the tip magnetization (see field lines in Fig. 4). Figure 4 shows a magnet coil coaxial with the cantilever, which is used to create the large initial magnetic trap.

   

Magnetic resonance will be excited by capacitively driving the cantilever at its resonance frequency, leading to a time-varying component of the magnetic field at the Larmor frequency of those atoms at the resonant slice in the trap. Fig. 5 (left) shows a SEM micrograph of the silicon cantilever with an electroplated CoNiMnP magnet glued on its tip. The cantilever is positioned approximately 20 micrometers above the gold electrode used for driving its motion. The photograph on the right of Fig. 5 shows the cantilever mounted on its base-plate. There is a gold coating on the underside of the cantilever forming the second electrode for the capacitive drive. An AC voltage applied across these two electrodes drives the cantilever when the AC frequency matches the resonance frequency of the cantilever. 

Fig. 6 is a photo of the flange that holds the cantilever and the coil for the initial trap inside the vacuum chamber. The cantilever base-plate is fastened to the bottom of the rectangular hole in the cylinder.

Contact: Andy Geraci

References:

Y. Wang, M. Eardley, S. Knappe, J.M. Moreland, L. Hollberg, and J. Kitching, Magnetic resonance in atomic vapor excited by a mechanical resonators, Phys. Rev. Lett. 97, 227602 (2006).