by nanocomposite, you are talking about polymer nanocomposite, right? Both SEM and TEM can do the job, but TEM is better since it gives more details for size distribution. I never worked with graphene, but I think that its same as silica work.

Khan U, O'Neill A, Porwal H, May P, Nawaz K, Coleman JN. Size selection of dispersed, exfoliated graphene flakes by controlled centrifugation. Carbon. 2012;50(2):470-5

At this range, SEM should work otherwise TEM for sure can give you clear image. Then you can use some graphic software such as photoshop and try to calculate the number of particles. finally by using Excel you can calculate frequency of particle size or their distribution.

using SEM for 200 nm is difficult because you can't see nanoparticles clearly in that size but for 1000 nm I think it is OK. and using TEM is OK for both sizes.

Besides TEM, DLS could also be a better choice in determining size distribution of nanoparticles including graphene

I would take TEM - you will be able to differentiate graphene from the rest of the composite if there is difference in atomic composition.

@ Milad Esfahani - photoshop is not the best choice. I am using ImageJ - it`s free and have multiple options including particles analysis.

It is possible to determine the distribution with software like imageJ. This is very well suitable for spherical particles as nanosilica. Unfortunately in TEM pictures you see a “projection” of your particle on the plane of the sample, for this reason a coplanar layer will show all its shape whereas the same layer in an orthogonal orientation will show an area much lower corresponding to the thickness of the layer. For this reason I think is not possible to measure directly the distribution of layered particles like graphene by this way. At least you can assume that you layer are equally oriented in all direction and then elaborate the distribution from imajeJ. But this assumption is rather to be true either for the propension of lamellar filler to be oriented and the sample thickness needed for TEM (50-100nm) that could be lower than the lamellar length.

In addition to the useful info from Marco Zanetti:

also, please, try the VERY quick introduction guide from Matlab Image Processing Toolbox (http://www.mathworks.com/help/images/image-enhancement-and-analysis.html).

It would be useful, if you have Matlab, of course.

If not, it would give you the general view on the state of the arts in image processing and related work on size distributions.

Please, look how we used this toolbox to get size distribution of elongated structures in some of our works:

http://www.scientific.net/KEM.465.306

http://edges-grid.eu/c/document_library/get_file?folderId=80175&name=DLFE-1627.pdf

From my perspective you can disperse the graphene particles in a solvent by using ultra sonic or stirrer and take a DLS test (dynamic light scattering) from them. It is more precise. because DLS is made to measure the particle sizes and it will give you a mean diameter too. In addition it will cost less.

I have to tell you that I saw great difference between the mean size of particles that were measured by SEM and DLS.

Use Image J analysis!

How to use image j for particle size analysis

Hi,

TEM works fine, be aware that the handeling of the probes may affect your results. The samples are dried on a Cu grid and the electron beam can distroy your sample, you must be careful. DLS could be another option. Read my paper

"99mTc-labeled superparamagnetic iron oxide nanoparticles for multimodality SPECT/MRI of sentinel lymph nodes.", J Nucl Med. 2012 Mar;53(3):459-63.

Dear Abubaker,

Minimum resolution of SEM is around 2 nm. How are you telling "using SEM, 200 nm is difficult"?

For reference you can go through the following link:

http://en.wikipedia.org/wiki/Scanning_electron_microscope

http://web.utk.edu/~prack/MSE%20300/SEM.pdf