Most likely due to lower resolution in SEM analysis, you observe the paticles with two or three or ... number as an individual particle with larger size. So, when you are observing the microstructure via TEM, you behold the particles with smaller size!

Because of the magnification of the instruments.

SEM may not be adequate to give the needed resolution to make assessments of nano-particle sizes. Thus, the particles are not well resolved and magnified. This fact makes it impossible to make accurate estimates of the particle size distribution of the SEM image of nano-samples.

Dear Mohammad is it possible that organic layer on synthesized nanoparticles is dissolved in solvent (DMF) during sample preparation for TEM. that's why it give very less size value than SEM

All images taken by TEM are artifacts of sample preparation (e.g. dispersion in epoxy and microtoming) and selection of particles. Consider microtoming a cube. See attached from 'A Primer of Higher Space' by Claude Bragdon The Manas Press Rochester and New York (1913)

In both cases (TEM & SEM) the imaging and calculation of 10000 particles would be needed to specify the mean to 1%. If the x90 is required then 10000 particles would be needed in the region above x90. Thus, in both cases it's highly unlikely that a representative sample is taken. You can measure particle size but not reliable particle size distribution with both EM techniques.

Last, but not least, the harder you look then the more you see. The same effect can be shown with optical microscope measurements where, when the objective lens is increased in magnification, then the measured sizes decrease. The comment by Madhukara Naik is succinct and true.

I am grateful my dear Alan for your worthy and meaningful comment. Now the answer is cleared after reading your comment

Most likely a sampling issue. I think you can try embedding the nanoparticles in resin, obtain a thin section on a TEM grid and locate the same nanoparticles under SEM and TEM. Alternatively, you could simply disperse your nanoparticles in DI water and place a drop on carbon coated grid. This would be a direct size measurement comparison. If the difference is huge, then your technician needs to check the calibration.

Most general differences have been discussed in this thread: https://www.researchgate.net/post/What_is_the_difference_between_SEM_and_TEM_techniques2/1

So for nanoparticles, I'd like to point out that SEM has lower resolution (nm-range) and is able to scan larger areas than TEM. It's not suitable for very small particles (

Enoch Victor SW Wong You indicate precisely where TEM goes wrong:

• A tiny amount of sample (< pg usually) is taken which is not representative of the whole
• Enough particles are not measured for statistical validity (10000 random particles to get the standard error on the mean at 1%; tails of distributions need the same constraints)
• The sample preparation stage produces a size and shape distribution even if there wasn't one to begin with. Look at the single cube above in the Blagden pictures. Consider slicing monodisperse spheres - a set of discs from effectively 0 up to the true diameter of the sphere. Average will be 1/2 the true value in this situation

I would only ever use TEM for lattice and atomic spacings in crystalline material. It is not capable of delivering a true size or shape distribution

SEM is used for imaging the surface of the nanoparticle in a submicroscopic scale . In contrast, use of TEM is to image the internal structure of the nanoparticle in a nano-meter scale.. simply TEM has a higher resolution than SEM by a factor of 10 or more. ... In a TEM, a nearly parallel beam of electrons travels through a thin specimen, and the resulting image is magnified electron-optically by a series of electromagnetic lenses, the main one of which is the objective lens. That why Particles appear ∼3% larger in the SEM than in the TEM microscope.

Low resolution of SEM cannot tell the agglomerations but TEM can, resulting in more small size.

Most likely due to lower resolution in SEM analysis, you observe the paticles with two or three or ... number as an individual particle with larger size. So, when you are observing the microstructure via TEM, you behold the particles with smaller size!

Mostafa Dadashi Firouzjaei So we change the sample and its particle size distribution in TEM which we don't in SEM?

I do not know which SEM microscope do you use and how conductive is the sample. Howewer, if both SEM and sample is ok, you shloud be able to see particles of diameter 4 nm... at least you can feel that they are there.

Many thanks for this interesting question. As you know, in the SEM and TEM instruments, the desired materials are illuminated with high energy electron beams. In a SEM instrument, the beam is scanned over the sample, while in a TEM instrument, the beam passes through the sample. For this reason, a SEM image shows surface morphology and surface characteristics of the samples in form of powder or film, while a TEM image displays internal structure and real particle sizes because of operating at high working potential. The origin of the differences in the particles sizes obtained by SEM and TEM images can be understood from the step of sample supplying for these analyses. For an SEM analysis of powdered materials, some powders are usually utilized. The powder is stuck onto a SEM stub and it is coated by Au or other conductors for nonconductive materials. In the case of a TEM analysis, the desired material is usually suspended in a solvent by a high power ultrasonic instrument to spread the sample well. Then, the suspension put onto a TEM grid and let the solution air dry to obtain a thinly sliced sample. Therefore, the images obtained from TEM analysis has higher resolution and the agglomerations could be distinguished with this instrument, while in a SEM image, the agglomerated particles are usually observed with larger sizes. Hence, TEM analysis is the best technique to obtain the real sizes of materials in comparison with SEM analysis, which is difficult to study real particle sizes. It is noticed that to obtain the best size distribution using TEM analysis, it is recommended to take more images of the samples from different points.

Aziz Habibi-Yangjeh

I refer readers of this post to your discussion:

https://www.researchgate.net/post/The_differences_between_particle_sizes_from_SEM_and_TEM

SEM with low resolution compared to TEM, hence it is not true size of one particle, maybe several particles aggregated together.