Port to Port Isolation for Mixer

I was doing simulations on a fully balanced Gilbert Gain mixer - the LO signal is of amplitude 500mV and wanted to see how much of the LO signal leaks on to the RF port.

What I found was that there was very little contribution at LO on the RF port but there was a huge component at 2LO on the RF port. Is this normal?

Thank you in advance.

If you drive LO port with a sinusoidal signal, the second harmonic will be appeared at RF port because of capacitive and other couplings.Normally, you should drive LO port with square wave that has very low second harmonic content.Therefore, the second harmonic will be appeared as low as possible.But don't forget, the square wave amplitude should be very high as much as possible to prevent from this event.
Thus, mixer noise will be minimized because of hard transistion of square wave.