HFSS CSRR simulation

Hi everyone. While trying to verify the article on CSRR based sensor , 'Complementary Split-Ring Resonator-Loaded Microfluidic Ethanol Chemical Sensor' by Salim et. al. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5134461/) , I am not getting the correct results.

I tried using both lumped ports and wave ports.

While using lumped ports, the dip in return loss is very low.

While using wave ports, the resonating frequency is itself changing with a low return loss. ( is it because of the boundary box being close to the structure? As the automatic boundary condition in HFSS is PEC if we don't mention anything. If then how to simulate using wave ports? Is it by using a coaxial cable structure to the feed line)

I have attached my simulation files. Kindly have a look and tell me where I am going wrong in both of them. Also what should be the correct boundary conditions and excitation in planar resonators like these?

Thanks

Try to simulate your design in Eigne Mode and you will get an idea of your resonant frequency and Q factor. If these values match (or close match) then go for wave port or lumped port solution.