PDF

CPW Resonator for Circuit Quantum Electrodynamics
Introduction
Developments in the last decade have made circuit quantum electrodynamics (cQED) the leading architecture for quantum computation. cQED is the solid-state version of cavity QED, which studies the basic light-matter interactions at the quantum level. This new architecture for quantum hardware has two main components: superconducting qubits and transmission line resonators. Superconducting qubits are the artificial meta-atoms that serve as a two-level quantum system and the transmission-line resonators are high-quality superconducting oscillators that play the role of cavities.
The energy difference between the quantum states of superconducting qubits is given by E01 = hf01, wherein a two-level quantum system E01 is the energy difference between the ground state and the excited state, h is Plank’s constant, and f01 is the operating frequency of quantum qubits. This frequency is typically in the range of 4–8 GHz and also equal to the frequency of the microwave pulse used to coherently excite the qubits. Just like atoms, superconducting quantum qubits interact with microwave photons at quanta levels.
Figure 1: CPW resonator coupled to a CPW transmission line. Air domains are removed for a better view.
In this model, one of the main component of cQEDs, transmission-line resonators, are demonstrated. This resonator can be built from CPW transmission lines terminated with a combination of open and short ends. Open-short ends form nodes or antinodes at the end of the CPW which results in standing wave patterns within the transmission line and the transmission line serves as a resonator. Figure 1 illustrates the CPW resonator terminated with an open and a shorted end, also called a quarter-wave resonator, coupled to a CPW feeding line used in the model.
Model Definition
Figure 2 shows the schematic cross section of a CPW line that is used for the resonator and the feeding line. The impedance of the CPW is basically related to the ratio between center conductor width and gap width and the dielectric constants of the substrate. The conductive regions are simulated as perfect conductors for the sake of simplicity to mimic superconductors. A more realistic temperature-dependent superconductor model can be employed easily. Silicon is used as a substrate with relative permittivity 11.7.
Figure 2: Schematic of the CPW cross section where w/d=7/4 and the characteristic impedance is 50 Ω.
Numeric ports are employed to excite and terminate the feeding CPW line. Therefore the boundary conditions on those surfaces are the corresponding mode fields. Radiation boundary conditions are used on the remaining boundaries. Since the loss is very small, the quality factor of the system is too high, and the resonance is very narrow. To make the simulation computationally efficient adaptive frequency sweep is employed.
In such a high-quality factor component, to capture the resonance behavior, a very fine frequency sweep is required in the vicinity of the resonance. By default, COMSOL stores field values for all the frequencies, within the 3D computational domain. If we proceed with the default settings, generated file which stores field distribution would be quite large. Since we are mainly interested in the field distribution on the CPW filter surface, to reduce the size of the file, we may only choose to store field distribution on that surface. For this purpose, geometric selections are created, and in the study settings, field values for that geometric selection are stored.
Results and Discussion
Figure 3 illustrates S parameters of the system which demonstrate resonance behavior. At the resonant frequency, all the energy is efficiently coupled to the resonator. At off resonances, electromagnetic energy does not interact with the resonator at all.
Figure 3: The S-parameters plot demonstrates a very narrow resonance behavior.
Figure 4: Illustration of standing wave pattern formed within the resonator. The height distribution corresponds to the total electric field. Antinode and node can be observed at the open and short ends.
Figure 4 demonstrates the standing wave formation in the resonator at the resonance frequency. Nodes and antinodes can be observed at the short and open ends of the resonator.
Notes About the COMSOL Implementation
Since the CPW resonator is a very high-quality factor system, it is a challenging structure to simulate. It is highly mesh-sensitive and a mesh-refinement study is necessary to make sure that the results are reliable. To show a simple modeling workflow, this model only focuses on the simulation of a CPW resonator and the mesh-refinement study is not included in the example. It requires ~20 GB of memory with the settings used here. To get accurate results, it is highly recommended to follow all the steps in this document; ignoring some of the settings may result in inaccuracies such as resonance shift. Even though this shift could be very small, since the system is very high Q, it could be challenging to find the resonance.
Application Library path: RF_Module/Filters/cpw_resonator
Modeling Instructions
From the File menu, choose New.
New
In the New window, click  Model Wizard.
Model Wizard
1
In the Model Wizard window, click  3D.
2
In the Select Physics tree, select Radio Frequency>Electromagnetic Waves, Frequency Domain (emw).
3
Click Add.
4
Click  Study.
5
In the Select Study tree, select Empty Study.
6
For the sake of simplicity, import geometry feature is used since the geometry is complicated and COMSOL kernel is used for Geometry representation. The CPW resonator contains high-aspect-ratio features, and it’s also highly mesh-dependent due to its very narrow bandwidth. It is important to use the COMSOL kernel to reproduce the results shown in this example. Otherwise, minor variations in the mesh structure may result in discrepancies. The CAD kernel could be used, however, a mesh refinement process should be performed accordingly. In general, the choice of the kernel does not matter. But, this is a very high-quality factor device, and special care should be taken.
Geometry 1
1
In the Model Builder window, under Component 1 (comp1) click Geometry 1.
2
In the Settings window for Geometry, locate the Units section.
3
From the Length unit list, choose mm.
Import 1 (imp1)
1
In the Home toolbar, click  Import.
2
In the Settings window for Import, locate the Import section.
3
Click  Browse.
4
5
Click  Import.
6
Click the  Wireframe Rendering button in the Graphics toolbar.
Electromagnetic Waves, Frequency Domain (emw)
Perfect Electric Conductor 2
1
In the Model Builder window, under Component 1 (comp1) right-click Electromagnetic Waves, Frequency Domain (emw) and choose Perfect Electric Conductor.
2
Scattering Boundary Condition 1
1
In the Physics toolbar, click  Boundaries and choose Scattering Boundary Condition.
2
Port 1
1
In the Physics toolbar, click  Boundaries and choose Port.
There is no analytical solution to define mode field of the CPW. Use numeric ports and perform boundary mode analyses. The field distribution obtained from the boundary mode analysis will be used for the eigenfrequency and frequency domain analysis. Since a quasi-TEM wave is propagating on a CPW, use the Analyze as a TEM field option and define Integration Line for Voltage.
2
3
In the Settings window for Port, locate the Port Properties section.
4
From the Type of port list, choose Numeric.
5
Select the Analyze as a TEM field check box.
Integration Line for Voltage 1
1
In the Physics toolbar, click  Attributes and choose Integration Line for Voltage.
2
In the Settings window for Integration Line for Voltage, locate the Edge Selection section.
3
Click  Clear Selection.
4
Port 2
1
In the Physics toolbar, click  Boundaries and choose Port.
2
3
In the Settings window for Port, locate the Port Properties section.
4
From the Type of port list, choose Numeric.
5
Select the Analyze as a TEM field check box.
Integration Line for Voltage 1
1
In the Physics toolbar, click  Attributes and choose Integration Line for Voltage.
2
In the Settings window for Integration Line for Voltage, locate the Edge Selection section.
3
Click  Clear Selection.
4
5
Locate the Settings section. Click Toggle Voltage Drop Direction.
Materials
In the Home toolbar, click  Windows and choose Add Material from Library.
Add Material
1
Go to the Add Material window.
2
3
Click Add to Component in the window toolbar.
4
In the tree, select Built-in>Silicon.
5
Click Add to Component in the window toolbar.
6
In the Home toolbar, click  Add Material to close the Add Material window.
Materials
Silicon (mat2)
Select Domain 2 only.
Field is confined in the close vicinity of the CPW gaps. Use Refine conductive edges to refine the mesh in the vicinity of CPW gap.
Mesh 1
1
In the Model Builder window, under Component 1 (comp1) click Mesh 1.
2
In the Settings window for Mesh, locate the Electromagnetic Waves, Frequency Domain (emw) section.
3
Select the Refine conductive edges check box.
4
From the Size type list, choose User defined.
5
In the Size text field, type 5[um].
6
Click  Build All.
To see the mesh structure on the CPW surface, Use Hide for Physics.
Definitions
Hide for Physics 1
1
In the Model Builder window, expand the Component 1 (comp1)>Definitions node.
2
Right-click View 1 and choose Hide for Physics.
3
In the Settings window for Hide for Physics, locate the Geometric Entity Selection section.
4
From the Geometric entity level list, choose Boundary.
5
Mesh 1
As mentioned in the port section, there is no analytical solution to define mode field of the CPW. Perform Boundary Mode Analysis. The field distribution obtained from the boundary mode analysis will be used for the Eigenfrequency and Adaptive Frequency Sweep.
Study 1
Boundary Mode Analysis
1
In the Study toolbar, click  Study Steps and choose Other>Boundary Mode Analysis.
2
In the Settings window for Boundary Mode Analysis, locate the Study Settings section.
3
In the Mode analysis frequency text field, type 5[GHz].
4
In the Search for modes around text field, type 2.5217.
Step 2: Boundary Mode Analysis 1
1
Right-click Study 1>Step 1: Boundary Mode Analysis and choose Duplicate.
2
In the Settings window for Boundary Mode Analysis, locate the Study Settings section.
3
In the Port name text field, type 2.
Eigenfrequency
1
In the Study toolbar, click  Study Steps and choose Eigenfrequency>Eigenfrequency.
2
In the Settings window for Eigenfrequency, locate the Study Settings section.
3
In the Search for eigenfrequencies around text field, type 4.94[GHz].
4
Select the Desired number of eigenfrequencies check box.
5
6
From the Eigenfrequency search method around shift list, choose Larger real part.
By default, COMSOL stores the field values within the computational domain, for each frequency in the study step. For a densely meshed problem with a fine frequency sweep, the size of the automatically generated result file could be extremely large. To reduce the file size, we can omit the field values that we are not interested. For this purpose, a geometric selection could be generated and field values only within the geometric selection could be saved.
Definitions
Explicit 1
1
In the Definitions toolbar, click  Explicit.
2
In the Settings window for Explicit, locate the Input Entities section.
3
From the Geometric entity level list, choose Boundary.
4
Click  Paste Selection.
5
In the Paste Selection dialog box, type 2, 5, 8, 9, 11-18 in the Selection text field.
6
Study 1
Step 3: Eigenfrequency
1
In the Model Builder window, under Study 1 click Step 3: Eigenfrequency.
2
In the Settings window for Eigenfrequency, click to expand the Values of Dependent Variables section.
3
Find the Store fields in output subsection. From the Settings list, choose For selections.
4
Under Selections, click  Add.
5
In the Add dialog box, select Explicit 1 in the Selections list.
6
Solution 1 (sol1)
1
In the Study toolbar, click  Show Default Solver.
For this specific example using the combination of boundary mode analysis and eigenfrequency, one can take the advantage Vanka presmoother in the settings of Eigenvalue Solver to get a faster convergence and reduce computational time. The usage is limited in this model.
2
In the Model Builder window, expand the Solution 1 (sol1) node, then click Eigenvalue Solver 3.
3
In the Settings window for Eigenvalue Solver, locate the General section.
4
In the Relative tolerance text field, type 1.0E-5.
5
In the Model Builder window, expand the Study 1>Solver Configurations>Solution 1 (sol1)>Eigenvalue Solver 3 node.
6
Right-click Study 1>Solver Configurations>Solution 1 (sol1)>Eigenvalue Solver 3>Suggested Iterative Solver (emw) and choose Enable.
7
In the Model Builder window, expand the Study 1>Solver Configurations>Solution 1 (sol1)>Eigenvalue Solver 3>Suggested Iterative Solver (emw)>Multigrid 1 node.
8
Right-click Study 1>Solver Configurations>Solution 1 (sol1)>Eigenvalue Solver 3>Suggested Iterative Solver (emw)>Multigrid 1>Presmoother and choose Vanka.
9
In the Settings window for Vanka, locate the Main section.
10
In the Number of iterations text field, type 1.
11
Under Variables, click  Add.
12
In the Add dialog box, select Electric field (comp1.E) in the Variables list.
13
14
In the Settings window for Vanka, locate the Main section.
15
From the Block solver list, choose Direct, stored factorization.
16
In the Relaxation factor text field, type 1.
17
In the Model Builder window, expand the Study 1>Solver Configurations>Solution 1 (sol1)>Eigenvalue Solver 3>Suggested Iterative Solver (emw)>Multigrid 1>Postsmoother node, then click SOR Vector 1.
18
In the Settings window for SOR Vector, locate the Main section.
19
In the Number of iterations text field, type 1.
20
In the Relaxation factor text field, type 0.5.
21
In the Study toolbar, click  Compute.
Results
Multislice
1
In the Model Builder window, expand the Electric Field (emw) node, then click Multislice.
2
In the Settings window for Multislice, locate the Multiplane Data section.
3
Find the X-planes subsection. In the Planes text field, type 0.
4
Find the Y-planes subsection. In the Planes text field, type 0.
5
Locate the Coloring and Style section. From the Color table list, choose ThermalWaveDark.
Deformation 1
1
Right-click Multislice and choose Deformation.
2
In the Settings window for Deformation, locate the Expression section.
3
In the X component text field, type 0.
4
In the Y component text field, type 0.
5
In the Z component text field, type emw.normE.
6
In the Electric Field (emw) toolbar, click  Plot.
Surface 1
1
In the Model Builder window, expand the Electric Mode Field, Port 2 (emw) node, then click Surface 1.
2
In the Settings window for Surface, locate the Coloring and Style section.
3
From the Scale list, choose Logarithmic.
Add Study
1
In the Home toolbar, click  Add Study to open the Add Study window.
2
Go to the Add Study window.
3
Find the Studies subsection. In the Select Study tree, select Empty Study.
4
Click Add Study in the window toolbar.
5
In the Home toolbar, click  Add Study to close the Add Study window.
Study 1
In the Model Builder window, expand the Study 1 node.
Step 1: Boundary Mode Analysis, Step 2: Boundary Mode Analysis 1
1
In the Model Builder window, under Study 1, Ctrl-click to select Step 1: Boundary Mode Analysis and Step 2: Boundary Mode Analysis 1.
2
Study 2
Step 1: Boundary Mode Analysis
1
In the Model Builder window, right-click Study 2 and choose Paste Multiple Items.
2
In the Settings window for Study, locate the Study Settings section.
3
Clear the Generate default plots check box.
Adaptive Frequency Sweep
1
In the Study toolbar, click  Study Steps and choose Frequency Domain>Adaptive Frequency Sweep.
Since the CPW resonator has a very sharp resonance, the Adaptive Frequency Sweep can be utilized to reduce computational time. A good choice for the Asymptotic Waveform Evaluation (AWE) expressions increases the efficiency of adaptive frequency sweep. Magnitude of S11 is a suitable choice for this problem to decrease computational cost.
2
In the Settings window for Adaptive Frequency Sweep, locate the Study Settings section.
3
In the Frequencies text field, type range(4.9972[GHz]-3[MHz],0.1[MHz],4.9972[GHz]+3[MHz]).
4
From the AWE expression type list, choose User controlled.
5
Definitions
Explicit 2
1
In the Definitions toolbar, click  Explicit.
2
In the Settings window for Explicit, locate the Input Entities section.
3
From the Geometric entity level list, choose Boundary.
4
Study 2
Step 3: Adaptive Frequency Sweep
1
In the Model Builder window, under Study 2 click Step 3: Adaptive Frequency Sweep.
2
In the Settings window for Adaptive Frequency Sweep, locate the Values of Dependent Variables section.
3
Find the Store fields in output subsection. From the Settings list, choose For selections.
4
Under Selections, click  Add.
5
In the Add dialog box, select Explicit 2 in the Selections list.
6
7
In the Home toolbar, click  Compute.
Results
S-Parameters
1
In the Home toolbar, click  Add Plot Group and choose 1D Plot Group.
2
In the Settings window for 1D Plot Group, type S-Parameters in the Label text field.
3
Locate the Data section. From the Dataset list, choose Study 2/Solution 4 (sol4).
Global 1
1
Right-click S-Parameters and choose Global.
2
In the Settings window for Global, click Replace Expression in the upper-right corner of the y-Axis Data section. From the menu, choose Component 1 (comp1)>Electromagnetic Waves, Frequency Domain>Ports>S-parameter, dB>emw.S11dB - S11.
3
Click Add Expression in the upper-right corner of the y-Axis Data section. From the menu, choose Component 1 (comp1)>Electromagnetic Waves, Frequency Domain>Ports>S-parameter, dB>emw.S21dB - S21.
4
In the S-Parameters toolbar, click  Plot.