Wet etching GaN is not the greatest way to pattern GaN. Depending upon the particular material you have (N type/P-type and nitrogen face versus gallium face) the etch responds differently. It is likely you have C-plane, gallium face which is the hardest to wet etch. Assuming that, you can only use KOH and UV light exposure or KOH and electrochemical enhancement (if it is N or P-type and not insulating). Neither give great uniformity unfortunately and tend to be difficult to do (defects tend to etch faster). You can use a silicon oxide or silicon nitride mask for the etch with the nitride mask having greater selectivity. If at all possible, consider an RIE or ideally an ICP instead, using Cl, BCl3 and Ar chemistry.

Hi, I second Quentin's suggestion, In the past I have tried the procedure he has mentioned for Ga-face layers on Sapphire with good success. I dont exactly remember what I used for N-face, but I remember following one review article for N-face. But it depends on why you want to use wet-etching.

Quentin and Dhamodran summarized it. 1.5 um is too thick to go for wet etch.

You have to go for Cl based RIE

In Chlorine plasmas & For the etching of 1,5um; a simple photoresist of 2 to 3um thiok can be sufficient '

Regards

The mask nature has a direct impact on GaN etch profile. As a consequence. ..if you need almost anisotropic profiles, an oxide hardmask Will be prefered to a simple photoresist. Furthermore...selectivity is better with a hardmask than with photoresist.

Regards