Multistage Zeeman deceleration of atomic and molecular oxygen

Multistage Zeeman deceleration is a technique used to reduce the velocity of neutral molecules with a magnetic-dipole moment. Here we present a Zeeman decelerator that consists of 100 solenoids and 100 magnetic hexapoles that is based on a short prototype design presented recently [Phys. Rev. A 95, 043415 (2017)]. The decelerator features a modular design with excellent thermal and vacuum properties and is robustly operated at a 10 Hz repetition rate. We use this decelerator to demonstrate the state-selective deceleration of atomic oxygen to final mean velocities in the range 500–125 m/s. We further characterize our decelerator with molecular oxygen, which, despite its heavier mass, is velocity tuned in the range 350–150 m/s. This corresponds to a maximum kinetic-energy reduction of 95% and 80% for atomic and molecular oxygen, respectively. The long multistage Zeeman decelerator presented here demonstrates that the concept of using alternating hexapoles and solenoids is truly phase stable. This Zeeman decelerator is ideally suited for applications in crossed-beam scattering experiments; the state-selected and velocity-controlled samples of O atoms and ${\mathrm{O}}_2$ molecules are particularly relevant for studies of inelastic and reactive processes.

Figure 1

Schematic representation of experimental setup. The main components are the pulsed valve with optional discharge, the decelerator modules with turbo-molecular pump (TMP) and the REMPI detection setup. A second source chamber (not shown here) is connected directly to the detection chamber without the five decelerator modules in between. The sequence of alternating hexapoles and solenoids is represented by the (gray-orange) dashed lines inside the decelerator modules.

Figure 2

Two photographs and a schematic cross section of the multistage Zeeman decelerator used in this experiment. In the photographs, the lid is removed from the first vacuum chamber. The zoomed-in image shows the connection between two modules, which enables an uninterrupted sequence of solenoids and hexapoles. Solenoids are encased in white epoxy resin (Torr Seal) to resist flexing forces in the solenoids due to self-inductance. The cross section shows the close proximity of the solenoids and the circuit boards that drive the solenoid currents.

Figure 3

The Zeeman energy-level diagram of (a) O (${}^{3}P_2$) and (b) ${\mathrm{O}}_2$ ($X{}^3\mathrm{\Sigma }_g^{-}$, $N=1$). The low-field-seeking states with the largest Zeeman shift are indicated by the red lines.

Figure 4

(a) TOF profiles of atomic oxygen ($2p^4{}^{3}P_2$) detected with the MCP. Experimental traces are shown in the top half of the graph, simulated profiles in the bottom half. The intensity is normalized to the nondecelerated beam (black). For the decelerated peaks, only the selected part of the beam was measured around the final mean velocity of the deceleration sequence. The intensity of the two slowest packets has been magnified by a factor of ten for clarity. (b) Full TOF profiles after deceleration to a final mean velocity of 300 m/s. (c) Full TOF profiles of the atomic oxygen guided at 510 m/s.

Figure 5

(a) TOF profiles of oxygen molecules ($X{}^3\mathrm{\Sigma }_g^{-}$) exiting the decelerator. Experimental traces are shown in the top half and simulated profiles in the bottom half. The intensity is normalized to the nondecelerated beam (black), which was produced with a mean velocity of 350 m/s. For the decelerated peaks, only the selected part of the beam is shown with corresponding mean velocity. (b) Full TOF profiles of molecular oxygen guided at 350 m/s.

Figure 6

The experimental REMPI transitions of atomic oxygen from the three spin-orbit levels in the electronic ground state ($2p^4{}^{3}P_J$) to the excited state ($2p^{3}3p^1{}^{3}P_J$). The peaks in the top half correspond to the beam that was manipulated by the Zeeman decelerator, while the spectrum in the bottom half was measured from the ELV in the secondary source chamber connected directly to the detection chamber. The intensity of the peaks for the $J=1$ and $J=0$ levels are magnified 50 times for clarity.

Figure 7

The spectrum of molecular oxygen in the ground state ($X{}^3\mathrm{\Sigma }_g^{-}$) ionized via the $3d$ Rydberg levels in a $2+1$ REMPI transition. The top half corresponds to the beam that was manipulated by the Zeeman decelerator, while the spectrum in the bottom half was again measured from the ELV in the secondary source chamber. The signal intensity was scaled such that the peaks at 225 nm were equal in intensity. Initial $J$ states and transitions were assigned by using a simulated spectrum.

Figure 8

Simulated TOF profiles of oxygen atoms (${}^{3}P_2$) of (a) deceleration and (b) guiding sequences. The TOF profiles for the low-field-seeking states $M=1$ and $M=2$ are shown separately as the solid and dashed lines, respectively. Atomic oxygen in the $M=0$ state is not shown, because it is not focused by the hexapole array and therefore negligible in intensity. High-field-seeking $M=-1$, $-2$ states are defocused by the hexapole array, and therefore do not reach the end of the decelerator.