Both provide 2D images of a 3D system.

Hi Razieh,

what do you mean when talking about analysis? Grain size, chemical composition, crystallography, structural defects, interfaces, physical properties, ...? In general, for surface analysis SEM is obvious choice. TEM is suitable for internal structure visualisation. If you have opportunity take advantages of both options. Limiting factor for using of SEM is grain size of your nanoparticles. If you have less than 10-20 nm grains SEM has almost no sense.

Depends; The synthesized nanoparticles usually assessed by TEM, but the nanoparticles in a context mostly assessed by SEM imaging. Of course, it also depends on what instrument you have access to. 

If you will say something more about goal of your analysis (morphology, particle size, composition etc.) and material of nanoparticle (metal oxide, polymer...) it will be easier to point some direction

Elemental analysis can be also performed using TEM and with a much better resolution

Not sure why my earlier response was downvoted.  It would be helpful to have some constructive critique and not an anonymous 'downvote'.  Shape and size are 3D issues - not 2D problems - and any measurement technique that does not take care of and understand this fundamental issue is simply an artifact generator  An example is sample preparation in TEM - imagine a collection of identically sized spheres.  Microtoming these for TEM use will result in a collection of discs from essentially 0 in size to the largest disc which is equivalent to the true diameter of the spheres - an artificial  distribution results from a monodisperse system.  Imagine sectioning the human body to find the proportion of mass of the liver - a vertical section provides totally different answers to horizontal sections.  Note that small quantities of materials (pg/ng) are examined with both techniques and that representative sampling is a vital factor to consider.  In mass/volume/value terms it requires 1 million 1 nm particles to make the mass of a single 100 nm particle. Missing or ignoring that single larger particle is the same as missing or ignoring 1 million of the smaller ones in mass terms,

@Markus Thank you for your comments.  Much more useful.  I'm always careful ( = skeptical!) with any (black-box) technique as one needs to understand the question(s) and assumptions in order to generate a reasonable answer. The original question ('To analyze nanoparticles which one is better,  SEM or TEM? why?') is incomprehensive without context and thus my answer was entirely (IMHO!) in line with the question. It's like asking is an apple better than a pear?  As you'll know, in science, the answer is invariably 'it depends'.  We have no context in which to provide an answer in line with your statement "A good answer is a comprehensive and thoughtful response to the original question.” Was this really a 'good question' based on the good answer definition?

To answer your further question, Imaging and microscopy are absolutely vital in all aspects on size and shape analysis of nanoparticles.  We always have to look at the system to get some feeling for overall shape, size, crystallinity, aggregation, agglomeration etc.  But these are qualitative, not quantitative, features of the material system  To translate these into quantitative results needs much more care and, for example, a minimum number of features in line with the standard error requirements or specifications.  That's why I illustrated the examples of the slicing of spheres (NBS/1956!) and the liver mass.  There's a lecturer in University of Cambridge, UK that believes that everything under the TEM is an artifact!  I'm not that skeptical but the analysis of a few 2D images from a lot of material is extremely dangerous, IMHO. Shape parameters (e.g. aspect ratio) are often ratios of 2D features (length/breadth) and don't reflect the fact that nano-objects are 3-dimentsional. 

We can obviously continue this interesting debate but I note that there are a number of answers posted to the original but the original poster has not returned with any description of why the question was posed and the materials being examined. Fell free to continue your useful thoughts and comments - I appreciate them both off- and on-line.

To continue with more thoughts.  Electron microscopy probes the electron dense rich parts of a particle.  It may be very difficult to see a surfactant or stabilizing coating for example.  It's this surface that determines how the particle will interact with its environment.  Stability in suspension or colloidal form is dictated by this surface layer,  All in all, one needs a suite of tools in one's nano toolbox in order to understand more about our materials. Going to a material with only one technique (and microscopy is by far the most useful) is highly dangerous - it's akin to trying to fix a timing issue on one's car with a sledgehammer... I would consider specific surface area, surface composition, and surface charge to be very important techniques in nanoparticle characterization also.

"You should not ask questions without knowledge"  W Edwards Deming

I agree with the Markus Engenhorst Sir. Using SEM one can only see the shape and size of the nanoparticles but the exact size can not be determined. In addition to this, the surface coating of the nanoparticle is very hard to examine. The exact size and shape of the nanoparticle can only be determined using TEM analysis. The adhered material at the surface can also be seen using TEM. This study also facilitate the fringe width which shows the arrangement of the atoms. Using TEM one can also claim the crystallinity of the nanoparticles with the help of the SAED pattern which are obtained during TEM analysis.

Take a look, too, at this NIST posting:

https://www.nist.gov/news-events/news/2016/12/how-tell-when-nanoparticle-out-shape

Hi all

The technical difference between the SEM & the TEM can be described as  follows:

The SEM functions by bombarding a conductive particle surface with an electron beam from a Tungsten filament that under a KV potential range between 15-30KV.   As a result, an inelastic scatter of the secondary electron from the surface of the particle occurs and can be detected  by a special SEM detector.   Consequently, SEM images can be produced with a spatial  resolution up to a submicroscopic scale.   Indeed, the size of the  conductive particle surface  should be big enough to accommodate the electron beam bombarding in order to cause the inelastic scattering of the secondary electron.

In contrast, The TEM functions by transmission of an electron beam  through a transparent particle surface from a Tungsten filament that under a KV potential range up to  200 KV.   As a result, a shadow of the transparent particle surface  occurs and can be detected  by a special TEM detector.   Consequently, TEM images of the internal structure of the particle can be produced, i.e., grain boundaries, dislocations, second phase particles, and so on  with a spatial  resolution up to a nanometer scale.   In addition, not only the size of the  transparent particle surface  should be big enough to accommodate the transmitted electron beam but also, the thickness of the particle should be thin enough  in order for the electron beam to be transmitted through the particle thickness.

There is a number of techniques, i.e., ion milling,  of thinning the particle to reach the transparent stage.

The electron beam in both SEM & TEM is normally directed to the examined particle by a special lenses & electromagnetic field throughout the electron beam column. 

So, to answer your question;

Use SEM for imaging the surface of the nanoparticle in a submicroscopic scale .   In contrast, use TEM to image the internal structure of the nanoparticle in a nano-meter scale..       

simply TEM is MORE powerful than SEM. but if your particles are larger than 500 nm. it can be photographed by SEM as it gives BETTER figures (3D) than TEM. specially for crystals.

Dear Razieh / Hi all

You have triggered a thoughtful argument around your question.   You have managed to draw the attention of several  people into a useful discussion .   

So, for now; "say something...I am  giving up on you... ".

In the spirit of the holy season and the goodness of the coming year 2017, I wish you all a happy new year with much success, health, and prosperity.

@Alan F Rawle , Excellent answer!

In addition, You can use ImageJ software to determine the average size of nanoparticles in SEM, TEM image.

Link to download ImageJ: https://imagej.nih.gov/ij/download.html

Follow this video: http://bit.ly/2sbb5B5

Dear colleague Razieh Kheshtzar,

Have a nice time and good day.

Taking into account the aforementioned excellent discussion, you can use SEM to calculate grain size while TEM to calculate the crystallite size, there are a lot of differences between grain size and crystallite size, you can check at: https://www.researchgate.net/post/What_is_the_difference_between_particle_size_grain_size_and_crystallite_size

I wish this is useful and help you.

Yours

A. EL-Denglawey