I have analyzed surface area using BET measurements, and another technique AFM image (is that correct). There is a significant difference between both values. What does mean?
I agree with Jurgen that more information is needed; specifically about the nature of your material. Is it porous? What is its chemical composition? The attached paper might be helpful.
Without more information from your side, from the moment I can tell you the limitations of AFM. First, you have to take into account the convolution with the tip, i.e. the tip apex has a certain radius R and if you have motifs on the surface smaller than 2R they will have an apparent shape that is not the real one. Second, the tip can not properly penetrate into vertical trenches or holes due to its pyramidal shape: a vertical wall will always appear with a slope given by that pyramid.
If you liked my answer please click "Recommend" so it can help other researchers.
My question is what are the differences between the surface area analyzed via BET and AFM. Simply, BET provides surface area and pore size of the nanomaterials, while AFM provides morphology. The AFM inbuilt software also can analyze surface area, which one is most suitable, and what kind of surface area are?.
As I explained above, you will never get the real morphology of the surface with AFM. Convolution with the tip appex and limitations due to the pyramidal shape of the tip will always produce and apparent morphology that it is not the real one. As a result, the value of the surface area with AFM caculated by the software will be the value for that apparent surface, instead of the one for the real surface.
Indeed, AFM topography images are a convolution of the real topography (surface morphology) with the shape of the AFM tip used. Though it should be mentioned that the convolution effect is negligible if the AFM tip radius is much smaller than the surface features to be imaged. Furthermore, micromachined AFM tips are available from several manufacturers that allow imaging of high aspect ratio structures, owing to a tip opening angle close to zero. Another approach would be to attach a linear nanostructure, such as a carbon nanotube (CNT), to the very end of the tip (for a related study, see Surf. Interface Anal. 2011, 43, 1382–1391, DOI 10.1002/sia.3727).
Nanopores or nanocrevices that are not directly accessible by an AFM tip, ie (partly) buried, may contribute to the overall response of a BET measurement. In these terms, I would expect that BET analysis gives a larger surface area than AFM, depending of the porosity characteristics of the surface under investigation.