the isolation performance of single-balanced mixer

the single-balanced mixer is applied in a 2.4GHz zero-IF receiver for low NF design with supply current of 2mA. Are there some serious disadvantages with the single-balanced mixer, except for the RF-IF leakage?
what about the dc-offset caused by the IM2 in the single-balanced structrue?

In double-balanced mixers even-order spurious are rejected when in single-balanced are not.
In double-balanced mixer (compared to single-balanced) the 1-dB compression point and third-order intercept point is almost 6dB higher.

thank you ,vfone!
the reason i choose the single-balanced structre because the LNA is single-ended output, and the noise figure of single-input double-balanced exceeds the tolorance of the LNA gain. that's to say , the NF charactoristic is the derect reason i choose the single-balanced structre.
what harms does the bad 2nd-order spurious performance of the mixer will do to the zero-IF receiver?
expecting for more advice!

For a superheterodine receiver the 2-nd order distortion is out-of-band, when for zero-IF the 2-nd order distortion is in-band.
So, zero-IF needs better 2-nd order performance.

Imbalance in the mixers will result in a non-time variant DC-offset, when 2-nd order distortion will result in a time variant DC-offset (due to mixing with other interferers). So, when DC-offset cancellation is implemented must take in consideration both cases.

thank you vphone
i determin to abort the single-balanced structre, because i have not find any single-balanced mixer applied in direct-conversion receiver.maybe it's too adventurous， especially when the correnlated offset is time-varient。
thank you again for your advice