Can you describe in more detail what you mean by laser and photodiode getting tilted? Are you losing the laser focus or do you see actual deformations in your system? Is the system working properly when you try analyzing a pristine substrate? Does it return to a functioning state when you remove the sample and try something else?

With the information we have so far, it sounds less like a sample preparation and more like an AFM setting issue.

Sapphire? I would choose something cheaper, e.g. glass or steel or aluminium. Sticky tape can easily cause creep, i.e. more or less steady motion in x,y,z, on the scale of micrometres per hour. But none of that appears to be your problem.

No idea what you mean by "semicontact".

I would guess that you try to explain that your cantilever is bent so much that your deflection signal is min or max, so it cannot be used for feedback, and approach is impossible. My first idea is that your cantilever and also the polymer are both charged, so there is very strong electrostatic attraction or repulsion, bending the lever.

Alexander M Bittner A bit offtopic, but: sapphire isn't expensive anymore, you can get good SSP c-plane 2" wafers for 10-20€ and, as long as they are not put under stress, they can easily be reclaimed by acetone sonication.

Jürgen Weippert , sure, but compared to the simple glass slides I use....... I just made the remark because the sample was not directly placed on the sapphire, but on sticky tape.

1) use mica or glass, ensre laser is properly aligned;

2) ensure tip is under the sample;

3) use tape for sem - betrer than other as 3M;

4) if film is gel-like use long probes with smal k;

5) if afm system will not start properly try do some measurement on well known sample.

Hope it helps