looks like you see the reflowing of photoresist. Do you do hardbake on the resist? Like oven 1h 120C or similar..

Dear Ziao Lu , I imagine you want to get a 90º step by applying a black and white litho mask. If you applied the right developtment solvent and time and your resist is not too thick or too thin you should not have problems to get a fairly good step.

As A. Bagolini pointed, some resists can experiment reflowing by heating after the developtment, and that could cause softening of the step´s edge and sometimes a kind of round edge due to surface tension, but in general that would not cause a sharp spike in your AFM plot, but a wide, soft one.

I think you are observing a sort of artefact. When you are scanning your sample with the AFM tip is goes smoothly over a flat surface and suddenly it feels the step, over-reacting (a bit like us when we are walking and suddenly an unseen step appears). What happens there with the AFM is due to the cantilever that holds the tip. When it is scanning the flat region the tip is in contact with the surface (in contact mode) or close to it (in non contact mode), so the cantilever is exerting a pressure over the tip and it over the sample. When suddenly the surface dissapears under the tip, the cantilever tries to push the tip to find the surface again, but if the step is deep it could or not find it, but in any case the cantilever is now really bended, and then it has to go back (action-reaction Newton´s law) so you see a sharp spike in your plot.

The best way to test this hypothesis would be to check your sample with a SEM. In this way you could observe the real shape of your step. There are other means to obseve the profile of your step, like using a profilometer, specially an optical one. Also interferometry could be useful. But SEM is handy in many universities central services and fast and many new equipment comes with advanced software that can retrieve 3D info about your sample.

Hope it helps. Good luck with your resarch.

Best, Manuel

Manuel Gómez Thank you very much for your suggestion. I did not consider SEM for observation before, and I will use this characterization method in the future. Good luck with your research, too

Ziao Lu , you´re welcome. Keep in mind that non conductive samples need - in general (*)- some sort of coating with a very thin layer of conductive material to properly observe on a SEM. Gold or other metals are sputtered over the sample before the observation.

* Sometimes it is possible to observe an isulating sample without conductive coating, but you need a method to get rid off the charge accumulated on the sample´s surface, something that depends on the SEM equipment available.

Good luck!

Manuel