† Corresponding author. E-mail:
We theoretically investigate three-dimensional (3D) focusing of pulsed molecular beam using a series of hexapoles with different orientations. Transversely oriented hexapoles provide both the transverse and longitudinal focusing force and a longitudinally oriented one provides only the transverse force. The hexapole focusing position are designed to realize the simultaneous focusing in three directions. The additional longitudinal focusing compared with the conventional hexapole can suppress the effect of chromatic aberration induced by the molecular longitudinal velocity spread, thus improving the state-selection purity as well as the beam density. Performance comparison of state selection between this 3D focusing hexapole and a conventional one is made using numerical trajectory simulations, choosing CHF3 molecules as a tester. It is confirmed that our proposal can improve the state-selection purity from 68.2% to 96.1% and the beam density by a factor of 2.3.
Since the technique of electrostatic hexapole was introduced half a century ago,[1–4] it is demonstrated to be a quite powerful tool to focus molecular beams and realize the state selection.[5,6] The state-selected molecular beams are ideal starting points for experiments of photodissociation,[7–9] photoionization,[10,11] surface scattering,[12–14] inelastic or elastic scattering,[15–18] and further laser-induced alignment and orientation,[19–21] etc. In these experiments, preparing initial molecular beam with high state purity and beam density is usually a key point. For example, in scattering experiments, preparing molecules in a pure quantum state is necessary to measure the state-to-state cross sections. In reactive scattering experiments, particular those with small cross sections, high beam density would be crucial.
The molecules with first-order Stark effect in the inhomogeneous electric field inside the hexapole will experience a transverse harmonic force towards to the hexapole axis while there is no longitudinal force. The molecular trajectories of a particular quantum state will be focused to the exit aperture to realize the state selection. This process is analogous to the optical imaging. The typical 10%∼20% longitudinal velocity spread of the molecular beam will lead to the phenomenon of chromatic aberration, i.e., molecules with different longitudinal velocities will be focused at different positions. The chromatic aberration effect will broaden the focusing curve and reduce the state purity. Meanwhile, since the molecules fly freely in the longitudinal direction, the molecular packet will spread out, thus reducing the beam density.
Previously we have proposed a useful low chromatic aberration hexapole scheme to suppress the chromatic aberration effect.[22] By switching off the hexapole voltage rapidly at an appropriate time, the focusing positions of all molecules are very close. However, in that scheme, the molecules fly freely in the longitudinal direction and the focusing moments are not the same. Thus the beam density is low at the focusing point. In this work, we present a three-dimensional (3D) focusing hexapole scheme to further focus the molecular beam in the longitudinal direction compared with the conventional hexapole. The experimental device consists of a series of transversely oriented hexapoles and a longitudinally oriented one. The longitudinal focusing force is provided by the transversely oriented hexapoles and the longitudinally oriented hexapole is used to adjust the transverse focusing time to realize the simultaneous focusing in three directions, i.e., two transverse directions and one longitudinal direction. The hexapole voltages are all operated in a pulse mode so that the molecules experience a force during an equal time and the focusing times are independent of the longitudinal velocity. This scheme can improve the state-selection purity as well as the beam density at the focusing point. This paper is a theoretical paper without experiments.
Longitudinal focusing of a pulsed molecular beam after Stark deceleration has been experimentally demonstrated using a buncher whose operation principle is equivalent to that of the Stark decelerator operated at a phase angle of 0°.[23] Very recently, quadrupole lenses combined with bunching elements are used to focus decelerated and cooled molecules at the detection region in a molecular fountain.[24] Here we design the hexapole voltages and pulse sequence to realize a simultaneous 3D spatial focusing of a pulsed molecular beam.
The layout of this paper is as follows. First, we present the details of the 3D hexapole focusing scheme and the principle of parameter design. Then, performance of this 3D focusing hexapole is compared with that of a conventional hexapole using numerical trajectory simulations, choosing CHF3 molecules as a tester. Finally, the conclusions are presented.
The hexapole consists of an arrangement of six rod electrodes, which are alternately charged to voltage
The focusing equation of a conventional hexapole can be found in Ref. [26]. In our previous work, we also derived it by tracing the deformations of molecular phase-space distribution in free flights and the harmonic potential. The result is given by[22]
The conventional hexapole is oriented along the longitudinal direction (molecular beam axis direction). Molecules in the hexapole are focused in two transverse directions and free in the longitudinal direction. The molecular longitudinal velocity spread will induce a spread of the molecular focusing positions, thus reducing the state-selection purity. Also the spread of the molecular packet will reduce the beam density. In order to reduce the longitudinal spread of the focusing positions and the packet, molecules should be longitudinally focused. The longitudinal focusing force can also be provided by the hexapole by means of orienting the hexapole transversely. The focusing region length of a transversely oriented hexapole is limited by the diameter of the hexapole, which is much shorter than the focusing region length of a conventional hexapole. The short focusing region length requires high hexapole voltage and thus strong electric field strength to realize the molecular longitudinal focusing. In order to realize the molecular longitudinal focusing under an experimentally available hexapole voltage and electric field strength, more than one transversely oriented hexapoles are needed.
The proposed experimental scheme is shown in Fig.
The molecular focusing process using a harmonic potential well can be described by the transformation of the molecular phase-space distribution. The description of the transverse focusing process can be found in Ref. [22] and can be extended to the longitudinal focusing case. In our scheme, the focusing time of each direction is calculated based on the focusing time equation (
The focusing time of molecules in each direction depends on the molecular Stark effect and forward velocity, the geometry of the hexapoles, the switching moments of the voltage pulses that are applied to the hexapoles, and the magnitude of the hexapole voltages. In the parameter design of the device, the molecular properties and the geometry of the hexapoles are first set. We define the molecule in the target state to be selected out which has no transverse velocity and the longitudinal center velocity of the molecular beam as the synchronous molecule. We determine the switch-on and switch-off moments of each hexapole when the synchronous molecule arrives at a certain position inside the hexapole. After that, the focusing times only depend on the hexapole voltages.
For simplicity, the voltages of the hexapoles with the same orientation can be set to be the same. A hexapole voltage corresponds to an angular frequency of the molecular harmonic motions inside the hexapole. We use
To compare the performance of the 3D focusing hexapole with that of the conventional hexapole in state selection, the method of numerical trajectory simulations is used. The oblate symmetric top molecule CHF3 is used in the simulations, as in our previous work.[22]
The electric dipole moment μ of the CHF3 molecule is 1.65 Debye and the rotational constants are A = B = 10.35 GHz and C = 5.67 GHz.[28] In the simulations, the average forward velocity of the pulsed CHF3 molecular beam is set to be 300 m/s, which can be realized via a low-temperature pulsed valve[23,29,30] This relatively low forward velocity helps to reduce the voltage requirements of the transversely oriented hexapoles since molecules spend more time in the focusing region. The velocity distributions of the initial molecular beam in all directions are independent Gaussian ones. The full width at half maximum (FWHM) in the transverse direction is 5 m/s and 30 m/s in the longitudinal direction. The initial transverse spatial distribution at the valve is also assumed to be Gaussian with an FWHM of 0.1 mm. The molecular pulse also has an initial duration at the valve. We assume the pulse has a Gaussian temporal function with a duration of
The parameters of the 3D focusing hexapole are designed as follows. The number of the transversely oriented hexapoles is set to be six. The distance between the nozzle and the center axis of the first transversely oriented hexapole is set to be 17 cm. The radii of the transversely oriented hexapoles are all set to be 10 mm. The radius of the cylindrical rods of the hexapole is set to be 0.565 of the hexapole radius, as recommended by Anderson.[25] The distance between the center axes of nearby transversely oriented hexapole is set to be twice the hexapole diameter, i.e., 40 mm. The flight region of the synchronous molecule when the voltage of the transversely oriented hexapole is on is set to be symmetrical about the hexapole axis with a length of hexapole radius, i.e., 10 mm. The value of
In the simulations, the initial CHF3 molecular beam contains
The parameters of the 3D focusing hexapole designed above is based on Eq. (
Due to the high electric field strength in the transversely oriented hexapoles, the nonlinear Stark effect should not be neglected. The first-order Stark interaction energy given by the first-order perturbation calculation is the diagonal element of the interaction Hamiltonian[34]
We calculate the Stark effect to the first order, second order and using the exact Stark interactions. Figure
The simulation results are given in Table
To assess the ability of the 3D focusing hexapole to increase the beam density on the focusing point, we calculate the molecular number in the
In summary, we have presented a proposal to improve both the purity and the density of the state-selected molecular beam by using a combination of longitudinal and transverse harmonic potential wells provided by a series of transversely oriented hexapoles and a longitudinal oriented one. Performance comparison between this proposed 3D focusing hexapole and a conventional one is made using numerical trajectory simulations. In the simulations, the ideal electric field distribution inside the hexapole is used and the exact Stark energy of the CHF3 molecule is taken into consideration. Electric field deviations from the ideal hexapole field is neglected. It is confirmed that our proposal can improve the state purity from 68.2% to 96.1% and the beam density by a factor of 2.3. For more complex polar molecules, the improvements of the purity and density of the state-selected beam of our scheme need to be further evaluated. We demonstrate theoretically the improvements of the purity and density of the state-selected beam, and the molecule loss due to collisions to the hexapole components is considered. In a realistic experiment, the extension of the molecular trajectories may induce other loss mechanisms of molecular number that have not been considered. Our proposed hexapole has promising prospects in some molecular beam experiments, e.g., molecular scattering experiments in which high state purity and beam density are desirable. The improvement of the molecular beam density is limited by the imperfect molecular source. It looks forward to the technological development of the molecular source with a narrower transverse distribution and a shorter duration.
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] | |
[18] | |
[19] | |
[20] | |
[21] | |
[22] | |
[23] | |
[24] | |
[25] | |
[26] | |
[27] | |
[28] | |
[29] | |
[30] | |
[31] | |
[32] | |
[33] | |
[34] |