It is very difficult to completely destroy an expert witness. Here are some of the strategies that one might ordinarily expect to see deployed:
1. Attack the expert's credentials: Difficult and very dangerous to attempt here, given academic status and the fact that the Judge himself appointed these experts.
2. Go after typos and minor mistakes in the report to score a couple of quick and easy points: Possible here, I suppose, but it never gets you very far.
3. Find prior inconsistent statements: Not sure if any are available here, e.g., published papers, prior reports, etc.
4. Suggest bias, if you can: Very dangerous here, given the fact that the judge appointed the experts.
5. Define the expert's scope of work as narrowly as possible, pointing out what they are not addressing/attacking: Not very effective here, since the court has already defined the scope.
6. Ellicit areas of agreement: Possible here, I suppose, to get the experts that agree that whatever came out of the machine does not exclude Sollecito/Kercher. Not sure if you can get more than that, though.
7. Nibble around the edges of the main opinion/findings: I don't think they will be able to do too much damage here, certainly not enough to fully undermine the experts.
Notwithstanding these tactics, it's very difficult to totally destroy an expert on cross-examination. Usually what you have to do, to win an expert argument, is to put on your own expert who is smarter and better and more persuasive. In this case, the prosecution's only real hope will be to put Stefanoni back on the stand to rebut what the experts have said. If this is even possible, and if there is only one day of testimony left, the only way to do this will be to cut short the cross-examination of C&V in order to get Stefanoni up there.
We'll see where it all goes.