Hi everyone,

I'm new to metasurface and simulation (e.g., CST) and working on simulating a metasurface placed between TX and RX coils to enhance magnetic coupling in a wireless power transfer system. The metasurface consists of a 3×3 array of unit cells (e.g., resonant loops or spirals, using same geometric from some papers), and I'm trying to achieve negative effective permeability (μ) near resonance — but I always get positive values, even near the intended resonant frequency.

I’ve tested two simulation setups:

1. Single unit cell with waveguide ports:

- Boundary conditions: X = PEC, Y = PMC, Z = Open

- Excitation: Waveguide ports on Z faces 

- Goal: Analyze normal-incidence response to extract effective parameters - Result: Only positive μ, no clear magnetic resonance.

I also tried Floquet port excitation with full periodic boundaries (X/Y = Unit Cell, Z = Open + Floquet ports), but the s-parameter and extracted μ was even less physical.

2. Full 3×3 metasurface placed between two coils: 

- Boundary conditions: All open

- Excitation: Only TX coil modeled as discrete port

- Goal: Evaluate WPT efficiency improvement

-Result: clear field enhancement can be observerd.

Is there something wrong with my simulation setup or approach? Should I consider different boundary conditions or retrieval methods? Any insights on possible mistakes in the settings would be greatly appreciated. Thank you very much for your time.