In HFSS, why the airbox should be set up with free space as lambda/4?

Hi, I'm trying to understand basic simulation on HFSS.

I've been told that when using boundary RADIATION, the airbox should be set up with lambda/4 air space, being lambda= c / freq (this freq is the one that is set up on "solution frequency" on Analysis options)
My question is why lambda/4 should be the minimum? I'm sure there is based on theory aspects but I haven't been able to find it out.

Thanks

Depends on what you are simulating. Lambda is often the rule of thumb for antennas since they usually do not radiate further than that. If you have prior knowledge to how far the fields radiate you can better estimate how far/close your air box has to be to your device. If you are to close to the radiate fields they could reflect off your air box and give you erroneous solutions.

Does it have to be an AIR box? Can it be some other material?

Thanks.

The quarter wavelength condition comes from the formulation of the radiation boundaries. These boundaries assume that the incident field is uniform, and doesn't have much phase variation across the incidence plane - like a plane wave. The quarter wavelength is the usual rule of thumb for the distance after which this condition is *close enough*. If you need the ABCs to be closer, use a PML instead.

In HFSS, radiation boundaries are programmed to work with the default material parameters. If you leave the default material as a vacuum, and put the radiation boundaries along a dielectric, they will not work properly, and reflect much of the incident power. You can modify this material in the HFSS -> Boundaries -> Edit Global Material Environment (HFSS 13).

Thank you for the previous clarification.

I have one more question.

Lambda refers to what frequency? The "solution frequency" on Analysis options or the lowest frequency of my sweep?

Thank you.