Velocity-selected molecular pulses produced by an electric guide

Electrostatic velocity filtering is a technique for the production of continuous guided beams of slow polar molecules from a thermal gas. We extended this technique to produce pulses of slow molecules with a narrow velocity distribution around a tunable velocity. The pulses are generated by sequentially switching the voltages on adjacent segments of an electric quadrupole guide synchronously with the molecules propagating at the desired velocity. This technique is demonstrated for deuterated ammonia (ND${}_3$), delivering pulses with a velocity in the range of $20-100$ m/s and a relative velocity spread of $(16\pm 2)\%$ at full width at half maximum. At velocities around $60$ m/s, the pulses contain up to ${10}^6$ molecules each. The data are well reproduced by Monte Carlo simulations, which provide useful insight into the mechanisms of velocity selection.

Figure 1

The experimental setup consists of three differentially pumped vacuum chambers. In the first vacuum chamber molecules are injected into the guide through a ceramic nozzle. A fraction of these molecules is guided via the four-segment electric guide to a quadrupole mass spectrometer (QMS) where they are detected. The inset in the lower right part of the figure depicts the transverse profile of the electric field. $r$ is the free inner radius of the guide, at which the maximum trapping field is reached. The lengths, $l$, and curvatures, $R$, of the four segments are detailed in the table in the upper left corner of the figure.

Figure 2

In (a) the switching sequence of the electrodes is presented. In the time intervals $[t_B^{\mathrm{on}},t_A^{\mathrm{off}}]$ and $[t_C^{\mathrm{on}},t_B^{\mathrm{off}}]$, respectively, two consecutive segments are in guiding configuration. The pulse-production process is sketched in (b). $z_1$, $z_2$, and $z_3$ are the distances from the exit of the ceramic tube to the entrance of segment $B$, to the entrance of segment $C$, and to the QMS, respectively. The transmission for certain velocities is schematically presented in the $z-t$ diagram in (c). The slopes of the black lines represent molecules from different starting times that can overlap each other at the QMS position. The slopes of the red dashed lines show the maximal and minimal velocities of the molecules in the pulse.

Figure 3

Time-of-flight signal for a pulse with a center velocity of 60 m/s. The detector response is plotted against the arrival time of the molecules. The middle of the time interval $[t_C^{\mathrm{on}},t_B^{\mathrm{off}}]$ corresponds to $t=0$ in the plot. The distance the molecules traverse during the time-of-flight measurement is the sum of the lengths of segments $C$ and $D$ and the gap of 2.2 cm between the end of the guide and the QMS. The depicted pulse has been obtained with the gradient off-configuration (see text). Background contributions have been subtracted from the raw data.

Figure 4

Normalized velocity distributions of pulses with an average velocity of 60 m/s measured in two modes of operation: with the grounded off-configuration (black filled dots) and with the gradient off-configuration (red open dots). For the guiding configuration, $+4\hspace{0.5em}\mathrm{kV}$ are applied to one pair of opposite electrodes while the other pair of opposite electrodes is at $0\hspace{0.5em}\mathrm{V}$, this leading to electric fields of up to $~$40 kV/cm. In the gradient off-configuration the deflection field reduces the amount of molecules that can still enter the successive guide segment. The asymmetric shape of the grounded off-configuration peak results from the bend geometry of segment $C$ of the guide.

Figure 5

Experimental velocity distributions obtained from the continuous beam (black dots with error bars from statistical uncertainties) and from the segmented guiding (red dots). Monte Carlo simulations of the velocity distributions are shown for continuous and pulsed operation of the molecular guide by black solid curves. Both experimental and simulated velocity distributions have been rebinned to 1 m/s.

Figure 6

FWHM of the velocity distributions are presented as a function of the mean velocity of the distribution for both the experimental data (red circles) and the Monte Carlo simulations for the bent guide (black inverted triangles) and the straight guide (gray triangles). The widths are obtained from fits of the velocity distributions with Voigt profiles. The solid curve represents the analytical model for $\tau =244\hspace{0.5em}\mu$s [see text and Eq. (3)]. There are no points for the bent guide for velocities above 100 m/s since these velocities approach the longitudinal cut-off velocity of our electric guide, and hence the molecular flux is strongly reduced (see Fig. 5).