-
Ion traps are important devices for confining charged atomic and molecular ions in a well-defined volume. Over the years, two main types of ion traps, e.g., Penning trap and radiofrequency (RF) trap, have been commonly utilized for such a purpose based on the employed physical fields for ion trapping [1]. Specifically, Penning trap employs magnetic and static electric fields, while RF trap employs static and RF electric fields, as solely static fields cannot maintain charged ions in a stable stationary equilibrium configuration according to the Earnshaw principle [2, 3]. Compared to a Penning trap, the construction of a RF trap is simpler and the motion of ions within the trap is also simpler, allowing it to be capable of storing relatively large number density of ions and of forming pulsed ion clouds that is eaisly compatible with advanced laser spectroscopy and mass spectrometry [4].
In 1953, Paul developed the technique for trapping charged particles in mass spectrometry by quadrupole electric fields, and the core part is now widely known as quadrupole ion trap, or Paul trap, which won him the Nobel prize in physics in 1989 [5, 6]. Ever since, the RF ion trap technology has undergone extensive development, with a variety of RF ion traps of different structures, such as blade traps, printed circuit board traps, ring traps, and multipole traps evolved from quadrupole traps, being developed and fabricated [4, 7−11]. They have been playing increasingly important roles in multiple research fields, in particular in precise measurement, quantum computing, atomic and molecular physics, chemical reaction dynamics, and high-resolution molecular spectroscopy [11−22]. To ease manufacturing and enable miniaturization of the devices, simplified trap structures have been developed. For example, Cooks and co-workers detailed a miniature cylindrical ion trap for mass spectrometry, which consists of a cylindrical ring electrode and two flat end-cap electrodes that are much easier to be fabricated compared to the conventional hyperbolic-shaped electrodes [23]. They later further designed a rectilinear geometry ion trap, exhibiting enhanced ion trapping capacity with a simple trap structure of only 3.5 cm3 in internal volume, as demonstrated in a mass spectrometer [24].
During the past decades, integration of ion trap technology with spectroscopic techniques, such as negative ion photoelectron spectroscopy [25, 26] and action spectroscopy in the spectral range across from microwave to infrared to optical, has revolutionized the relevant researches [13, 19, 21, 22, 27−29]. For example, Wang et al. realized, for the first time, the combination of a continuous electrospray ionization source with a pulsed time-of-flight mass spectrometer and a magnetic bottle photoelectron analyzer with the help of a 3D Paul trap [25]. The trap is critical for that development as it enables to store and enrich the ions within the central volume of the trap, and to eject them in pulsed ion clouds for the following pulsed-mode experiments. This innovative advancement allowed them to measure the repulsive Coulomb barrier in multiply charge anions [30, 31]. Further development by attaching the ion trap to the cold head of a cryocooler allows to cool the ions trapped inside the device to a low temperature via collisional energy transfer among the ions, buffer gas, and the wall of the trap [26, 28, 32]. This has greatly advanced the spectral resolution of the spectroscopy and generated new opportunities for relevant research disciplines, since low temperature allows the ions to mainly populate into their ground states thus eliminating hot bands [33]. In addition, low temperature also allows the formation of inert gas-tagged ions that are superior for action spectroscopy via photon absorption induced action on the molecules, for example, dissociation, reaction, or electron ejection [21, 28, 29]. For instance, with a 22-pole RF ion trap attached to the cold head of a helium cryocooler, Maier and co-workers measured the absorption spectrum of C60+ in the gas phase through measuring the wavelength-dependent depletion of C60+-He complex via He loss [13]. They identified C60+ to be a carrier of diffuse interstellar bands (DIBs) at 9577 and 9632 Å, representing a breakthrough in understanding the carriers of DIBs almost 100 years after it was discovered in 1920s [13, 34] and demonstrating the significance of ion trap technology in advancing fundamental research.
Since Gerlich and co-workers first introduced the design of 22-pole ion trap around 1995 [4, 7], it has been widely employed for high-resolution spectroscopy experiment due to its great cooling capability [13, 32, 35]. However, a 22-pole ion trap has a much larger flat region in its potential field and thus a much larger ion motion area, compared to a quadrupole ion trap [28, 36]. That could potentially distort the mass resolution of a time-of-flight mass spectrometer when they combine together. A quadrupole ion trap has a much smaller ion motion area, but due to the typically large size of the electrodes, RF heating effect is normally severe, reducing its cooling capability [6, 26]. In addition, the precision requirement of the electrodes of a quadrupole ion trap is high, posting quite some fabrication challenge.
Inspired by the good performance of ion traps constructed by replacing the large size rod electrodes with small wire electrodes [28, 37], we describe in this work a detailed design of a wire quadrupole trap (w4PT), in which six wire electrodes represent one rod electrode in a conventional quadrupole ion trap. This design makes the advantages of both a quadrupole trap and a multipole trap, for instance a 22-pole trap. Our simulations show that it can effectively mitigate the RF heating effect and tightly confine the trapped ions in a small volume. The cooling ability of w4PT employing buffer gas cooling has been investigated, and its construction and connection to the cold head of a Sumitomo cryocooler are presented as well.
-
The non-uniform and time-varying nature of RF electric fields is a distinctive characteristic [4, 6]. The driving force applied to ions is often subject to nonlinear variations, resulting in approximate analytical solutions for describing the ion motion only under specific conditions. In 1967, Motz and Watson introduced to use the effective potential to describe the ion motion in a RF field [38]. This model simplifies complex systems into a potential energy surface, making the analysis of ion motion straightforward. Under the assumption of adiabatic approximation, the motion equation for an ion is given by,
where
$ Q $ is the ion charge,$ m $ is the ion mass,${R}_{0}$ is the position of the ion,$ {\ddot{R}}_{0} $ is the acceleration of the ion,$ \varOmega $ is the frequency of the RF field, and$ {E}_{0} $ is the strength of the RF field. This leads to the effective potential of the ion trap,The static field
${\phi }_{\rm{s}}$ should be included for the ion motion within the trap,Thus, the general formula of the equation of motion is:
Different types of ion traps have different forms of
$ {V}^{\mathrm{*}} $ , which will affect the motions of ions inside the trap and thus the shape of ion clouds. This is an important factor to evaluate the performance of an ion trap. In the next, we will be focusing on linear ion traps that have been widely employed.In general, a linear ion trap can be characterized using the parameter n, with n being the number of RF electrodes and 2n the total number of electrodes. For example, n=2 corresponds to a quadrupole trap, and n = 4 an octopole trap [39]. In linear ion traps, analytical solutions for the ion motion equation are only available for n=2, i.e., a quadrupole ion trap, for which the effective potential is harmonic. By separating the electric field within the trap into static part
$ U $ (from the end cap electrodes and the direct current of the RF field) and RF part$ V $ (from the alternating current of the RF field), thenwhere
$ \mathrm{\kappa } $ is the geometrical factor of the trap. The$ z $ axis is defined as the axial direction that ions are confined by the direct current field of the endcap electrodes, and the radial directions x and y are confined by the RF field. Then, the ion motion can be described by the Mathieu equation [4],where
$ {Z}_{0} $ is half of the trap length,${R}^{\prime}$ is the equivalent distance from the ion to the electrode,$\ddot{R}_{i}\mathrm{ }$ is the acceleration of the ion.${a}_{i}\quad\mathrm{and}\quad{q}_{i}$ are specific parameters in the Mathieu equation under the conditions of an ion trap. The geometrical$ \mathrm{\kappa } $ and${R}^{\prime}$ originate from the difference between axial electrodes and parallel plate electrodes and the difference between the cylindrical electrodes and ideal hyperbolic electrodes. These two parameters normally need to be fitted using experimental data. The solutions of the Mathieu equation in$ x $ ,$ y $ , and$ z $ directions can all be written as:with
$ {\omega }_{i} $ the oscillation frequency of ion motion,In practical situations, the RF ion trap potential field is not entirely equivalent to an ideal RF field. The energy of ion motion is not conserved, and the ion energy steadily increases with the accumulation of storage time. This is the so-called RF heating effect. In the radial direction, the motion of ions is the superposition of the low frequency
$ {\mathrm{\omega }}_{i} $ long-term secular motion and the high-frequency ($ \varOmega \pm {\omega }_{i} $ ) intrinsic micromotion. When ions deviate from the axis with zero RF field, they will be driven by RF field, generating excess micromotion at a frequency of$ \varOmega $ . In principle, the potential in the axial direction generated by the endcap electrodes is harmonic, but deviations from ideal conditions, for instance the poor fabrication and installation precision of the electrodes, could also induce heating effect.For traps with more than 4 electrodes (n > 2), the motion of the ions within the trap no longer has analytical solutions. Instead, numerical solutions have to be adopted, and the effective potential in the multipole traps can be written as,
The equation tells that at the same voltages, traps with a greater number of electrodes exhibit larger potential variations near the electrodes, thus the range of the central flat region is larger. This implies that ions are more tightly confined in a quadrupole trap than in a multipole trap with electrodes more than 4.
-
The effective potential approximation assumes that ions engage in energy-conservation motion within a static potential field. However, in reality the electric field within a trap is not ideally static, and the ions will continuously gain energy along with the trapping time. This is the above mentioned RF heating effect, which will make the kinetic energy of ions inside the trap increase. Ions of higher kinetic energy will move closer to the RF electrodes, resulting in a more severe RF heating. Finally, the ions could even escape from the trap when they gain enough energy, hence proper methods have to be adopted to cool the ions in order to counteract the RF heating effect [40].
Many methods have been employed to cool ions, such as active-feedback cooling [41], collisional buffer gas cooling [42], evaporative cooling [43, 44], laser cooling [45], sisyphus cooling [46], among which collisional buffer gas cooling is one of the most commonly utilized methods to cool the ions trapped within an ion trap. It offers the advantage of being relatively simple to implement, good cooling effect, and the ability to cool both the translational and internal degree of freedoms of the desired species [22, 26]. In combination with a helium cryocooler, it is possible to achieve a cooling temperature below 5 K using buffer gas cooling, as demonstrated by several research groups [13, 28].
-
The electric field of a trap is determined by the number and shape of electrodes. In a RF ion trap, as represented by Paul trap, the electric field distributions and its performance is highly sensitive to the phase of the RF electrodes. If there is a phase difference between the two RF electrodes, the ions will feel additional driving forces [47]. RF ion traps require precise control over the electronics to ensure high-quality RF signal, and the transmission distances from the signal source to the two electrodes should be ideally identical to prevent generating additional relative phases. With perfect hyperbolic electrodes, the potential inside the trap is exactly a quadrupolar field, but due to the high precision requirement of manufacturing hyperbolic electrodes, cylindrical electrodes are more practically employed. This may weaken the field strength, but it greatly reduces the difficulty in fabricating the electrodes and the central region is still approximately a quadrupolar field.
To find a superior trap design that can tightly confine the ions in a small volume with minimum RF heating effect, three types of traps including quadrupole trap, 22-pole trap, and wire quadrupole trap were considered [28]. We first modeled the electric fields using the two-dimensional finite element method and cross-validation was conducted in conjunction with the boundary element method of the electrostatic module in COMSOL Multiphysics [48]. The effective potential energy models mentioned above were employed. FIG. 1 shows the geometric structures and the simulated electric fields of quadrupole trap (Paul trap), 22-pole trap, and wire quadrupole trap (w4PT).
The modelled Paul trap as shown in FIG. 1(a) is composed of 4 cylindrical electrodes of 13.70 mm diameter. The distance between the electrode surface and the trap center is designed to be 5.98 mm according to Fanghänel et al. [49]. The 22-pole trap is based on the design of Gerlich et al. in 1995 [7]. The 22-pole trap consists of 22 cylindrical electrodes of 1 mm diameter, with RF voltage being applied to every alternate electrode and the rest grounded electrode (FIG. 1(b)). The electrodes of the 22-pole trap are arranged in a circular configuration, allowing for variable sizes of the trapping region. From the simulation, it can be seen that the center of the 22-pole trap features a large flat field-free region, providing an extensive area for ion trapping and cooling due to collisional energy transfer between ions and buffer gas. The w4PT is derived from Paul trap [37], but instead of using cylindrical electrodes as described above, w4PT achieves the required electric field by carefully positioning 24 wire electrodes with 6 wires on each side simulating the hyperboloids (FIG. 1(c)). The field distributions of the Paul trap and the w4PT are basically the same in the central region, with differences only at the boundaries. They have a relatively smaller flat region in the center region compared to the 22-pole trap, and the spatial distribution of ions is smaller.
Regarding the magnitude of electric field strength, under the electrical potential of 1 V on the electrodes, the electric field near the electrodes in the 22-pole trap reaches up to 2.6×103 V/m, which makes it a highly effective potential trap as determined by Eq.(12). In the central region, the electric field is zero, allowing the ions collide with buffer gas free from the effect of RF electrodes. For Paul trap and w4PT, the same electrical potential yields a maximum electric field on the order of 102 V/m. The high electric field strength within the 22-pole trap indicates that it can form a deeper potential field, and higher electrode voltages are required for Paul trap and w4PT to achieve the same trap depth as the 22-pole trap.
In this work, the main aim is to design an ion trap that is suitable for ion trapping, cooling, and the generation of an ion beam for following spectroscopic experiments based on time-of-flight mass spectrometry. Hence, two aspects need to be taken into account when choosing the trap design. First, the size of the ion beam directly affects the following mass resolution and the interaction area with the laser beam, hence tight confinement of the ions within the trap would benefit. Second, to achieve as low temperature as possible via collisional buffer gas cooling, RF heating should be minimized and also a simple trap assembly could help its shielding from blackbody radiation.
From the above analysis, it can be seen among the three types of design, 22-pole trap has minimal RF heating effect, but compared to Paul trap and w4PT, it has larger flat potential field region that will result in much larger spatial distribution of the ions in the formed ion beam. In contrast, both Paul trap and w4PT have similar electric field distributions that can tightly confine ions in a small volume near the central region of the trap. That would give a smaller spatial and kinetic energy distributions for the ions in the formed ion beam. Considering that the large cylindrical electrodes employed in Paul trap may post fabrication difficulty and induce ununiformly flat electrode surface, stray electric fields, for example additional axial RF electric field, may occur. That might produce additional RF heating effect and influence the ion motion in the trap. Instead, the smaller wire electrodes of w4PT makes them easier to maintain flatness, and the smaller wire electrodes also will not introduce too much stray fields, ensuring the uniformness of the field distribution within the trap. Therefore, choosing a wire quadrupole as the electrode structure for the ion trap is more advantageous.
-
This section details the overall structure of w4PT, including the ion trap cavity, electrodes, and envelope of the ion trap, as well as the simulated performance of w4PT. The designed w4PT trap is depicted in FIG. 2. The electrodes utilize a set of 24 wires (1 mm diameter) in total, with 6 wires in each group as arranged in a hyperbolic shape. A pair of circular endcap electrodes is positioned on both sides along the axial direction. Two groups of wire electrodes are applied with RF voltage, and the other two and endcap electrodes with DC voltage, to achieve the ion confinement. FIG. 2(a) shows an overview of the trap assembly. The cavity of the trap is rectangular in shape with a cylindrical ion confinement area at the center, housing the 24 wire electrodes. To avoid possible hydrogen embrittlement effect under high vacuum conditions, the cavity is constructed using oxygen-free high thermal conductivity (OFHC) copper. As mentioned above, the wires are arranged with six on each side to simulate the hyperbolic electric field. Each wire needs to be precisely positioned as designated, and these wires are supported by printed circuit boards (PCBs), on which the RF power and DC power circuits are printed. The PCBs are made from Rogers 3003 material, a composite of polytetrafluoroethylene (PTFE) and ceramic that has excellent electrical properties including low RF signal loss and good insulation at low temperatures. Two endcap electrodes provide voltages for the axial ion confinement in the trap; a pulsed voltage can be applied to one electrode with the other one being applied with reversed polar pulsed voltage to expel ions from the trap. The gas inlet on the side wall of the cavity has an external PTFE tube for attaching a needle valve or pulsed valve to inject buffer gas. There are also temperature probe holes that penetrate through the cavity wall into the ion trap cavity, enabling temperature monitoring.
We employed a 442 mm×190 mm×190 mm electric potential array to simulate the ion motion within the trap using SIMION 8.0 [50], which has a built-in microsecond-level timer and a fast-adjusting feature for the electrode voltages, allowing real-time ion motion simulation. The simulation describes the process in which ions experience rapid oscillations in the electric fields and are continuously cooled via collisional buffer gas cooling simultaneously, leading to their accumulation in the central region of the trap.
FIG. 3 presents the simulated trajectories of ions in the trap. As a demonstration, totally 20 ions with 200 amu molecular weight and 1e charge are employed in the simulation, and the total simulation time is 40 μs. These ions are initially released with a kinetic energy of 4 eV at the central region of the endcap electrode, and the RF voltage amplitude is set as 100 V with 1 MHz frequency and the endcap voltage is 5 V. Table I lists the detailed parameters employed for the simulations. The stable ion motion region can be reached after 10 μs of ion entry during the simulations.
The simulation shows that the ions initially exhibit a wide range of motion, and with the effect of buffer gas cooling, ions gradually approach the center of the ion trap. The ions can be confined within a range of approximately ±5 mm axially and ±0.7 mm radially within the ion trap, behaving a cylindrical distribution as expected from the design. The effect of buffer gas temperature and endcap electrode voltage on the ion confinement are shown in FIG. 4. It can be seen that with higher buffer gas temperature, both the radial and axial motion regions of the ions increase, implying that as the buffer gas temperature rises, ions gain kinetic energy and the cooling efficiency reduces. The endcap voltages determine the axial ion confinement, and the simulation shows that when the endcap voltage reaches 30 V, the ions distribute within a much smaller range of 2 mm along the axial direction, only about 40% of the initial axial distribution region when the voltage is 5 V. The side effect of using higher endcap voltages is that the initial kinetic energy of ions upon entering the trap will be higher, and thus the higher speed, which may result incomplete collisional buffer gas cooling and the escape of ions from the trapping region. Hence, the endcap voltages should be carefully optimized considering both the size of the ion cloud and the resultant cooling effect.
FIG. 5 shows the ejected ion beams from the w4PT at the entrance endcap voltage of 5 V and the exit endcap voltage of −5 V. It can be seen that the ion beam seems converge radially first and then diverge, and at a distance of 4 mm after the exit endcap, the radial distribution is ±0.7 mm. This radial divergence will become more with increasing distance and einzel lens may be employed to re-focus the ions radially to improve the mass resolution of the following time-of-flight experiments.
During the machining and assembling of the trap, care must be taken to ensure good thermal and electric contact. The RF signals of the wire electrodes should be stable and their phases are consistent. Cables connecting different RF wire electrodes should have the same length and the PCB design should ensure basic phase constancy and avoid the RF signal passing through the PCB and affecting the DC signals. The wire electrodes must be tightly connected to the PCBs, and ceramic components can be added to the electrodes for precise fixation, reducing heating effect caused by electrode tilting.
-
In the above sections, we describe the detailed design of the trap and the ion manipulation within the trap. Since the temperature within the trap is critical for its performance, we next will discuss about the connection of the w4PT with the cold head of a cryocooler, more specifically, a two-stage Gifford-McMahon cryocooler, RDK-408D2, from Sumitomo Heavy Industries. Special attention has been paid to the thermal contact between the ion trap and the second stage of the cold head, and its isolation from the surrounding environment to reduce the potential heating due to blackbody radiation.
FIG. 6 shows the designed peripheral structure of the ion trap, including the connection between the trap and the cold head and its isolation from the environment. The second stage of the cold head, connected to the OFHC copper base of trap cavity via an indium foil for good thermal contact at low temperatures, has a factory-designed cooling capacity of 1 W at 4.2 K. The first stage of the cold head, with a designed cooling capacity of 40 W at 43 K, is connected to the external part of the trap cavity with grooves to precool the gas lines or electric wires to around 43 K before entering the trap. An aluminum shield is added outside the trap cavity and connected to an extension of the first-stage of the cold head, preventing blackbody radiation from surroundings. The cryocooler is securely mounted between the external vacuum chamber and the ion trap using a bellow connection between two CF flanges, ensuring a collimated flight path.
To evaluate the cooling effect of the ion trap after its connection with the cryocooler, we conducted heat transfer simulations to study the temperature distribution of the ion trap using COMSOL Multiphysics. As described above, both the trap cavity and base are made of OFHC copper, the wire electrodes and screws are made of 1008 stand steel, the PCBs are made of Rogers 3003, and the tubing for the buffer gas is made of PTFE. In the simulations, the thermal conductivity coefficient of PTFE is set as a constant of 0.24 W·m–1·K–1, and its heat capacity at constant pressure (Cp) is 1050 J·kg–1·K–1. For the other materials, the built-in parameters are utilized. In the trap, the main heat-producing components are the wire electrodes and the PCBs on both sides due to the electric current heating, and the total current is assumed to be 1 A during the simulation. FIG. 7 presents the temperature distributions of the trap from heat transfer simulations, and it can be seen that the PCBs holding the wire electrodes are the parts limiting the cooling effect. That is most likely due to the not superior thermal conductivity of Rogers 3003 material used for PCBs, hence other materials having similar electronic property but better thermal conductivity at low temperatures than Rogers 3003 may benefit the cooling. Nonetheless, the simulations indicate that with the current design and material selection, the trap still can reach a temperature below 5 K, low enough to cool ions to their ground states and to tag helium atoms onto the ions for high resolution spectroscopy.
-
In summary, we described a detailed design of a wire quadrupole trap (w4PT) based on the initial idea of Gerlich. In the design, six wire electrodes are employed to represent one rod electrode conventionally used in a quadrupole ion trap, thus it can effectively mitigate the RF heating effect relative to a conventional quadrupole trap. Our simulations on the electric fields within the trap suggest that w4PT can tightly confine the trapped ions in a smaller volume compared to a multipole trap, for example a 22-pole trap. In addition, we also investigated the cooling ability of w4PT employing buffer gas cooling and detailed its connection to the cold head of a Sumitomo cryogenic cooler. Heat transfer simulations show that with careful selection of the materials, w4PT is able to reach a low temperature of below 5 K, satisfying the requirements for many low-temperature experimental environments.
-
This work was supported by the National Natural Science Foundation of China (No.22273070 and No.92261101) and the Innovation Capability Support Program of Shaanxi Province (2023-CX-TD-49). Gao-Lei Hou acknowledges the support from the “Young Talent Support Plan” of Xi'an Jiaotong University, the “Excellent Overseas Young Scientist Program” of Shaanxi Province, and the Fundamental Research Funds for Central Universities, China.
Design and Simulation of a Wire Quadrupole Ion Trap for Ion Trapping and Cooling†
- Received Date: 29/11/2023
- Available Online: 27/04/2024
-
Key words:
- Radiofrequency ion trap /
- Wire quadrupole trap /
- Buffer gas cooling /
- Cryogenic cooler
Abstract: Serving as the core device of an ion trapping system, radiofrequency ion traps are widely employed in chemical reaction dynamics, atomic and molecular physics, precise measurement, and ion spectroscopy. They can store, enrich, and cool ions down to very low temperatures in combination with a cryogenic cooler, thus greatly facilitating the following experiments, such as, high-resolution spectroscopy. In this work, we describe in detail the structural design of a radiofrequency wire quadrupole trap (w4PT) that utilizes six wire electrodes to act as one cylindrical electrode of the conventional quadrupole ion trap to simulate the hyperboloids. We simulated the electric fields and ion motion trajectories within the trap to investigate the performance of the designed w4PT, and showed that it is capable of reaching a tight confinement of ions in a small volume. We also conducted the heat transfer simulations based on the designed supporting platform for the trap, and the chosen materials for the electrodes, trap cavity, and printed circuit boards for installing the electrodes. The simulated temperature distributions showed that the current trap could cool ions to a low temperature below 5 K using a helium cryogenic cooler. This work shall be helpful for the development of ion trap technology to be employed in high-precision spectroscopy.
首页
登录
注册


DownLoad: