What can’t you do with atomic force microscopy (AFM)?
These days it seems sometimes as if AFM would be the universal technique applicable to simply everything: from geological samples, (human) tissue, entire bacteria and cells to (membrane) proteins, in biology, and down to single molecules on the lower, physical limit. More exotic application examples are the skin of the mummy ötzi or the AFM on mars.
Moreover, force spectroscopy is possible as well as topography plots, each with a variety of modes suitable for different samples.
But, is this actual AFM hype justified?
Everybody who has worked at an AFM will agree that it sometimes makes you believe in voodoo. Therefore, it is not surprising that a scientific fraction has assembled, which doubts its significance as a scientific method. The still existent difficulty to reproduce AFM data nourishes such doubts constantly.
So, what is it?
Is AFM still in its childhood?
Or is this unreliability just simply the price we pay for the insight into „nano-world“ provided by this technique?
And what is its true significance compared to other measurement methods?
First of all, I’d like to know what/who you mean by a certain „scientific fraction“ that doubts the value of AFM? Do you have any pointers to where we can see their criticisms?
It’s true and I’ve heard it myself that individuals criticise the AFM as unreliable and over-hyped. But I’d claim that a big chunk of those people think AFM is something that it isn’t (and thus have too high or the wrong expectations) or simply haven’t understood the tool fully. Because we always see these nice AFM height images, many not so familiar with the AFM think that the AFM is an imaging tool similar to e.g. optical microscopes. But this is simply wrong. First and foremost, AFM is a force measurement and manipulation tool. And with this premise comes a wholly different set of things to consider when analysing AFM results or when designing AFM experiments.
Any serious researcher working in the AFM field will gladly tell you about its limitations, but also its strengths. And I think AFM is not different from other methods in that poorly designed experiments lead to unreliable results.
Regarding your statement about „still existent difficulty to reproduce AFM data“, you should consider that the AFM has advanced to a key quality control tool in huge industrial sectors like the semiconductor and chemical engineering industries. Chip makers wouldn’t check their wafers by AFM if the results weren’t reproducible.
What might be a valid question and really a problem for the AFM field, is if the use of AFM somehow leads to a larger number of poorly designed experiments, e.g. by being a technique that is easier to misunderstand (see above). This is not a problem of the AFM technique/tool itself but could still be a problem for AFM researchers if this reflects poorly on them. I do not know if this is the case. Solutions here could be having more AFM in science education, since it is being increasingly used in any (natural) scientific field.
Lastly, I actually am not really sure that AFM is so much hyped. AFM is ubiquitous in the nanoscience research, this is true. But outside our field, AFM has been ignored or only a part of its potential has been used. But slowly more and more scientists from all kinds of fields are becoming interested in this tool. I see this in our lab, where more and more people come to us and ask if AFM could help in solving their scientific questions.
(I apologize for the long posting ;))