S M Naimul Mamun

You don’t have particles, just a fused aggregated mass. Try BET or laser diffraction on the powder.

Dear S M Naimul, The SEM image that you have attached is unclear. But if you have proper SEM Image you can calculate the particle size using MATLAB. We can use image processing tool box and can calculate.

Here I have attached the ZIP file consists .m file and some example images.

Please go through the readme file before executing the .m file.

Another solution is follow the attached link.

https://www.youtube.com/watch?v=QbNsECyTXbQ

All the best

Dear S M Naimul Mamun,

The particles exhibited on your providing image are very tiny and aggrigated. You may use ImageJ software for measing your particle size. Albeit it is kind of a patient-work, it seems not impossible to accompinish. I have attached some videos that could be helpful for your study.

https://youtu.be/JQAZ--c9YfI

https://youtu.be/5UObaiiHplI

https://youtu.be/zrGm-gBhsUQ

https://youtu.be/H7D5R6wVw6A

Best Regards

Rakibul Hassan

S M Naimul Mamun

Grain size may be a better descriptor than particle size. You do not have any discrete, separate particles in your image.

Dear S M Naimul, the SEM gives an image for particles agglomeration, not for separated crystallites. So, to determine the particle size (crystallite size) you must use TEM where the size will be given directly by the program used.

Dear S M Naimul Mamun

Generally, it is preferred to measure particle size using the DLS technique. However, only in cases of monolayer arrangement of particles, SEM can be used. It is strongly suggested to go with the DLS technique.

Thanks

Dear @Gorthala Guruprasad, Thank you for the zip file. Your file contains similar SEM images to mine. I am not familiar with Matlab. I read the readme file but do not get it. Do you have any video tutorials where you describe how to calculate the particle size in similar SEM images using MATLAB?

Dear Kamilia Sedeek, I will go for TEM this week. But the problem is the technician here is not familiar with doing TEM for thin film samples; rather, he suggested me to make powder samples for the TEM. But my supervisor said that the TEM of the thin film and powder would not be similar. I am confused about how I should approach this.

Alan F Rawle

Here the research facilities are very limited, and we do not have any opportunities to perform BET or laser diffraction on the powder. Also, my samples are GA2O3 thin films.

S M Naimul Mamun By definition, you can't have 'particles' in a thin film...

Try using ImageJ for particle size, but image is unclear

To add my 5 cents:

> How can I determine whether the morphology is changing or not with doping?

The presented images is hard to interpret, because significant information is mission (beam energy, detector, pixel separation) are missing.

Looking at the image I would call it a SEM image of a rough polycrystalline material (later in the discussion it turned out to be Ga2O3 thin film), but the quality (pixel size) is too small for discussion and some experiments.

So, what morphology do you want to measure?

- The size of the “grains”, their shape (are they more or less circular, or quadratic, or elongated; there are several descriptors), the roughness (height differences in a certain area)?

- If you have different samples (doping) you should deduce the numbers and prove if the results are statistically different or not. I’ve seen a lot of papers where a set of let’s say 4 images is presented without quantitative interpretation and the figure captions reads “Effect of (doing/heating) on grain size…”.

TEM was mentioned and has a better resolution, but a much smaller field of view. Therefore, getting good statistics may be very time consuming. In TEM the information is along the beam in the sample. It doesn’t matter if your image is a thin layer in transmission or a cross section. Your lamella may contain several grains along the beam.

If we interpret the objects in your image as grains, how can we measure in SEM?

If was often discussed here:

https://www.researchgate.net/post/How-do-you-obtain-particle-size-distribution-from-SEM

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

https://www.researchgate.net/post/How-to-calculate-the-grain-size-from-a-SEM-image-or-from-microscope-optical-image

https://www.researchgate.net/post/How-to-calculate-the-grain-size-from-a-SEM-image-or-from-microscope-optical-image

A very simple and fast method is the Average Grain size method:

http://homepages.math.uic.edu/~jbaldwin/crystal%20task/agi_method.pdf

Particle analysis requires images of sufficient quality (contrast, signal-to-noise ratio), resolution (pixel separation), and image size (pixel number). The presented image is not sufficient.

https://forum.image.sc/t/enhancing-grain-boundaries/22399/3

For you, you should also consider XRD and AFM. AFM software can do such statistics (Gwyddion is free).

Dear, Mamun ji

An image should be clear. A lot of software are available to calculate the particle size. one software is J image software which is free for download and lots of videos are available for this software on youtube.

Firstly the image is not very clear and teh diemsion or scale of the image is also not mentioned. Since the morphology is particle-like so i think magnified scale SEM image would be a better option to choose. Further ImageJ can be used for the particle size estimation.

I am totally agree with Shiv dutta and Drashya that the SEM image should be very clear and the surface and particle brightness and contrast should be adjusted so that it will be easy to create colour threshold. After that using image J software, you can determine the particle size .

For all of those that advocate Image J above, then please show how this can be used for the posted SEM...

Try using ImageJ for particle size, but first use an image optimization program like Topaz Gigapexil

You can use Imagej for area distribution analysis, but I think you need to have better image at higher magnification scale and/or higher resolution, the particle shapes and sizes are not clear.

Plz repeat you sem characterization for better and clear image, and then use image j software.

First, your SEM image should be clear. And then you can use imageJ software for estimation of particle size.

True that the image is not clear. You can repeat the test and ask the SEM operator to do coating for your sample, after that you can distinct the morphology. As mentioned before, you can use Image J software to measure the size of NPS. Just search in Youtub you will find helpful videos

Compare the pure and dopant of your SEM images then you can identify whether there is any difference in the morphology. According to me SEM images were looks like spherical in shape.

Dear at what scale you magnified this

If it is 200nm then go ahead with image J software

Nanomeasurer or imageJ software can work. It is preferable to provide a clear fig.

S M Naimul Mamun

Looking keenly its spherical shaped particles...we can calculate its size by image j...now also...but need patience.

But, for measuring...the scale is visible...

Better to take sem again to calculate exact size ..

You need a scale bar and ideally a reference specimen to calibrate the magnification more accurately as most service technicians do not bother to do this to high precision for all acceleration voltages set up.

Then see if you can get an image with a better focus - this looks a little fuzzy.

After thresholding a lot of programs will give you a size distribution but check that the choice of threshold and magnification themselves do not influence your result. Finally, remember you will not see anything smaller than your point resolution or twice the pixel size chosen (Nyquist limit), whatever is the bigger, so your histogram will be cut off there.

SEM images does not give the exact particle size because there are some cases in which partcle aggregation takes place.

Hi

Do you have product`s XRD ?

using Image J you can find the size, but the image is very unclear.

The image is blurred. You might need to redo the SEM imaging and increase the focus this time before taking snapshots. Another option is to use TEM which has a better resolution. I hope this helps.

By using Image J, BTW, the SEM images are not Clear.

Dear Mamun,

You should take clear image. You can calculate particle size during the taking SEM images. Either you can use particle size analyzer that is very easy.

your image is good enough to measure the size.

And also you may try to use nano measurer to do this measurement.

Image J is another option.

You can test the morphology by using the atomic force microscope, and you can get the 1 micrometer x 1micrometer window in 2D and 3D views. For measuring the size of nanoparticles you can divide each photo on several fragments (small) and equals to each other. Then you can calculate the amount of particles in each. And their sizes you can estimate by means of scale of your device and size of all photo. Furthermore, you can determine the average size in each fragment, the average of size in window of your photo, the Gauss distribution of particles in their sizes. Later, you can compare the size distribution (by using Gauss statistic N(m,D)) with other distributions on the other photos with other particles. This method can help you to collect the information about sizes of particles much more than by measuring only single crystal sizes. Because, the procedure of doping is spatial distributed in all area of film or surface, too. The doping concentration can be spatial differs from one fragment of film to another. The application of statistical methods can detect the characteristics of film which are non transparent by measuring the sizes of a single particles. Only following by this way you can detect the dependence between the chemical changes of film and its morphological properties.

If your final product is powder, better to dilute it in a proper solvent and use the diluted sample for SEM imaging,

Shengjie Zhai OK, if it's so simple, let's see you demonstrate this and provide a 'particle size distribution' for this system where there are no separate, discrete, independent, particles...

You can use ImageJ software to determine the size of the particle. You will get several tutorials on YouTube, and by following the tricks shown there, you can easily do this.

Md. Somel Islam See my comment immediately above yours. I look forward to seeing your Image J report on the posted picture. As you say, it's 'easily' done...

Md. Somel Islam I see you have not posted your report of the 'easy' Image J analysis of the posted picture. Is it harder than you think?