† Corresponding author. E-mail:
‡ Corresponding author. E-mail:
To realize scale quantum processors, the surface-electrode ion trap is an effective scaling approach, including single-layer, double-layer, and quasi-double-layer traps. To calculate critical trap parameters such as the trap center and trap depth, the finite element method (FEM) simulation was widely used, however, it is always time consuming. Moreover, the FEM simulation is also incapable of exhibiting the direct relationship between the geometry dimension and these parameters. To eliminate the problems above, House and Madsen et al. have respectively provided analytic models for single-layer traps and double-layer traps. In this paper, we propose a semi-analytical model for quasi-double-layer traps. This model can be applied to calculate the important parameters above of the ion trap in the trap design process. With this model, we can quickly and precisely find the optimum geometry design for trap electrodes in various cases.
The ion trap[1] is currently a major candidate for quantum information process (QIP), serving as quantum memories, quantum gates, and nodes of quantum communication networks.[2] Quantum bits based on trapped ions have many advantages, including exquisite coherence properties and high efficiency to be prepared, entangled, and measured. However, the remaining challenge is the scaling of trapped ions to tremendous quantum bits, which is necessary for realizing scale quantum processors.[3]
The surface-electrode ion trap is an effective scaling approach due to its compatibility with the existing semiconductor process.[4] Several main surface-electrode trap geometries, single-layer, double-layer, and quasi-double-layer, have been experimentally realized. Single-layer surface-electrode (SSE) traps are much more suitable for the silicon-based micro-fabrication technology, where all electrodes are laid in the same plane. Since the first SSE trap was fabricated by MIT,[5] numbers of SSE traps[6–11] have been fabricated and tested, with high-fidelity quantum bits preparation, detection, transportation, and gate operations.[12–19] Double-layer surface-electrode (DSE) traps[20–23] have a deeper potential trap than SSE traps, which can extend the life time of trapped ions remarkably.[24,25] Therefore, although the fabrication process is complex, some researchers still attempt to develop DSE traps.[26,27] Another surface-electrode trap, the quasi-double-layer surface-electrode (QSE) trap, has the construction characteristic of that in between SSE and DSE traps.[28,29] After optimizing, the potential trap of QSE traps can be several times as deep as that of SSE traps, confirming that ions easily like a DSE trap. Moreover, the fabrication process of QSE traps is similar to that of SSE traps, which is much easier than that of DSE traps.
For all the traps above, to achieve a desired potential trap in the trap design process is critical for experiments. However, the widely used approach is always time consuming, which is to simulate the electric field with a finite element method (FEM) based software (i.e., Ansoft Maxwell or COMSOL). Moreover, the FEM simulation is also incapable of exhibiting the direct relationship between the geometry dimension and the critical trap parameters, such as the trap center and trap depth. To eliminate the problems above, House and Madsen et al. have respectively provided analytical models for SSE traps[30] and DSE traps.[31] Here, to provide insights in QSE traps design, this paper proposes a semi-analytical model for QSE traps. This model provides a fast and precise way in calculating trap parameters and designing the geometry of trap electrodes for desired fields. We have verified the proposed model against the FEM simulation. With this model, we have analyzed critical factors which determine the trap center and trap depth, and found the optimum geometry design for QSE traps. The remainder of this paper is organized as follows. In Section 2, we compare QSE traps with SSE and DSE traps and define geometry parameters for QSE traps. In Section 3, we propose the semi-analytical model for QSE traps and discuss insights provided by this model. Section 4 demonstrates that the model can describe QSE traps accurately through verification experiments. In Section 5, the model is applied to find the optimal geometries for the trap electrodes. Finally, Section 6 provides our conclusion.
In order to enhance the scalability of Paul traps, SSE and DSE traps have been implemented by mapping Paul traps to a flat surface as shown in Fig.
![](cpb_25_11_113701/cpb161198ieqn1.gif)
There is a large difference in the trap depth among Paul, SSE, and DSE traps. Figure
The QSE trap has a similar structure to the SSE trap, but RF and GND electrodes located in different planes. Compared with the SSE trap, fabricating a QSE trap is only required to add the process of etching, which is simpler than fabricating a DSE trap. Figure
Because the precise analytical model is hard to derive directly, we simplify the structure of the QSE trap with some assumptions as follows.
In fact, ion traps always have long electrodes, so assumption (I) is acceptable. Only the potential in the space y > 0 is considered in the semi-analytical model. In this space, the boundary condition is similar to that described in assumption (II). Assumption (III) is used to simplify the calculation of electric fields. We will add several additional factors to correct the error caused by assumption (III) in the following text. Some of the assumptions above have also been used in analyzing SSE traps[30] and DSE traps,[31] deducing analytical models whose accuracy has been confirmed. Moreover, the influences of assumptions (I) and (II) for surface-electrode ion traps have been found to be very small in realistic situations.[33]
According to assumptions (I) and (III), the RF electrode can be considered as a wire with an infinite length. After conformal mapping as follows:
The linear charge density q can be obtained by the following method. Regarding the RF electrode and the y = −d plane as a parallel plate capacitor with distance d and different plate area, the charge on the surface of the y = −d plane mainly distributes in the corresponding region with width c. This parallel plate capacitor is equivalent to lots of small series capacitors with distance dy and the same plate area. The capacitance value of each small capacitor can be calculated by the formula of C(y) = ε0S(y)/dy, where S(y) is the plate area of the parallel plate capacitor at y. By evaluating the integral of
Equations (
Insight 1 is useful in designing laser paths. When cooling ions for initialization or controlling ions for quantum computing or reading out the final state of ions, the trap center is the target location of applied lasers. For a given QSE trap, we can derive the target location from the model determined by the geometry factors including a, b, and d. Insight 2 indicates how to make the effort to improve the trap. One way to achieve a deep trap is to apply large voltage to electrodes. Thus, finding a way to improve the withstand capability of electrodes is of concern. Another way is to implement electrodes as small as possible, so long as technology allows. Insight 3 tells us that, for a determined characteristic size a, we can further improve the trap by finding the optimum design for the relative dimension α and β.
In the following verification experiments, the trapped ion is 40Ca+ and the frequency of voltage Ω is 10 MHz. The relationship between the trap depth and factors such as U, a, α, and β is important for trap design. One part of the verification is to test the consistence of that relationship between the semi-analytical model and the FEM simulation. Figure
![]() | Fig. 4. The trap depth of the QSE trap versus U and characteristic size a. U varies from 20 V to 180 V, and for each U, a is set to 50 μm, 100 μm or 150 μm. |
Figure
![]() | Fig. 5. The trap depth of the QSE trap versus (a) α (= b/a) in the range of 0.05–2 and (b) β (= d/a) in the range of 0.1–3. |
The semi-analytical model also provides the way to find the locations of the trap center and escape point as described in Eqs. (
One purpose in the design for QSE traps is to find the geometry dimension of electrodes that makes the trap depth as deep as possible. According to Insight 3, when the characteristic size a is determined, the factors that influent the trap depth are relative sizes α and β, which determine values of width b and height d of RF electrodes. In order to find the optimum design of RF electrodes, we calculate the trap depth when both α and β are varying. Figure
Another goal is to find the best dimension for the width w of the DC electrode, which should be chosen so that the field it creates has significant curvature everywhere along the ion transport path in the z dimension.[34] Figure
In this paper, we have proposed a semi-analytical model for QSE traps, providing a fast and precise way in designing the geometry of trap electrodes to achieve desired fields. We make several reasonable assumptions to simplify the deduction of the model. Through conformal mapping, we calculate the electric field of the QSE trap, and then obtain locations of the trap center and escape point, and the expression of the trap depth, which compose the full semi-analytical model of QSE traps. To correct the error caused by the assumption, we add two factors to the model. The verification experiments prove that our model can describe the QSE trap accurately. With the model, we develop some insights in trap design and find the optimum geometry design for QSE traps as α = 0.15, β = 1.2, and w = a.
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 |