Dear all,
I have a sample with almost spherical particles from 20 to 400 nanometers. To get a good estimate of the number of both small and large particles, I performed the particle size analysis on secondary electron (SE) images acquired at different magnifications, i.e. 5000X and 10000X.
Now, to get a global (and more representative) particle size distribution, I would to combine the particle size analyses performed on the 5000X images with that carried out on the images acquired at 10000X.
However, I am not sure that this is scientifically correct, since the "global" particle size distribution is calculated from images acquired at differnt magnifications.
Do you have any suggestion, please?
You can do that if you work smart on it. Of course, you can only calculate the correct particle size distribution from the image where all particles are sufficiently representative. This means that the picture with the lowest magnification must be your basis. Let's say it's x5000. For example, you can divide this image into 4 squares (magnification x10000) and analyze each square in terms of particle size. The sum of 4 analyzes then results in the real particle size distribution.
Thank you for your aswer Vadim but maybe the original question was unclear. I need to reduce the time of image acquisition. For this reason, I do the particle analysis on images at 5000X. However, at 5000X I can't see the small particles (e.g. 20-40 nm), thus the particle size analysis would under-estimate the number of small grains. Thus, to detect the small particles, I need to perform a particle size analysis also on a limited number of 10000X images. Then, I woul to combine the two particle size analysis (that's the problem).
In order to get a result you have to see all the particles, there is no other way. This means that you are not allowed to analyze an image 5000x. The 10000x image shows all particles but is not representative enough. It's not critical, because to get better statistics on image analysis 10000x, you just need some more images (5-10) from different parts of the sample. I don't know how you can reduce the analysis time when it comes to good results.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators