Electric-field sensing near the surface microstructure of an atom chip using cold Rydberg atoms

The electric fields near the heterogeneous metal-dielectric surface of an atom chip were measured using cold atoms. The atomic sensitivity to electric fields was enhanced by exciting the atoms to Rydberg states that are 10${}^8$ times more polarizable than the ground state. We attribute the measured fields to charging of the insulators between the atom chip wires. Surprisingly, it is found that although the chip wire currents were turned off before Rydberg excitation, the measured fields were strongly influenced by how the wire currents had been applied. These fields may be dramatically lowered with appropriate voltage biasing, suggesting configurations for the future development of hybrid quantum systems.

Figure 1

(a) Experimental apparatus. (b) Scanning electron microscope image of the atom chip at one end of trapping region, showing wires and insulating gaps. (c) ${}^{87}$Rb Rydberg excitation scheme (see for example Ref. 20). (d) Experimental sequence timing. A single cycle takes $\approx$$15\mathrm{s}$.

Figure 2

Rydberg excitation spectra after release from the (a) MOT (both with and without a magnetic field present), and (b) and (c) microtrap, with compensating electric fields applied (see text for details of compensation and wire biasing). (d) Measurement of the Rydberg signal as a function of applied electric field (by varying plate voltages, corrected for MCP fringing field and ac line interference; see Sec. 2e), with Gaussian fits. Positive wire bias (see text) was used for the spectra at $150\mu \mathrm{m}$ and $600\mu \mathrm{m}$, whereas the result for the larger distance $4200\mu \mathrm{m}$ was obtained by release from the MOT. We refer to the center of the fitted Gaussian (the applied field needed to null out the average electric field present at the atoms) as the “compensating field.”

Figure 3

(a) Distance dependence of compensating field (corrected for MCP fringing fields and ac line interference; see Sec. 2e) measured after microtrap release and MOT release, with power-law fits. The horizontal error bars indicate the excitation beam waist $\pm w$ (see Sec. 2d). (b) Compensating field for positive wire bias, measured at various distances from the chip, taken over 8 different days (represented by different point styles) in a two-month period. In all microtrap measurements in (a) and (b), the atoms were held in the trap for 225 ms prior to release. Inset: Histogram of 14 compensating field measurements $3\mathrm{mm}$ from the surface, taken after release from the MOT, on 13 days in a two-month period. (c) Charge accumulation in the dielectric gaps near a negatively biased wire (left) and positively biased wire (right).

Figure 4

Compensating field as a function of hold time in microtrap together with exponential fits, (a) positive wire bias, Rydberg excitation $210\mu \mathrm{m}$ from surface; (b) negative wire bias, Rydberg excitation $105\mu \mathrm{m}$ from surface.

Figure 5

A small change $dF$ in the electric field shifts the transition by some amount $d{f}_o$, via the Stark shift. This changes the measured signal level by $dS$.