Williamson-Hall analysis of tungsten nanoparticles for two different experimental conditions shows vast variation in average crystallite size, though both have same strain value of the order of 10^-3. One shows avg value of 123 nm and the other shows only 30 nm from the intercept of the plot. However, from the Scherrer formula former shows 32-36 nm and the latter shows 16-19 nm. Why is this so?
When both strain and size broadening are present, results of 'size' can be ambiguous, see our article. You don't explain the difference in experimental conditions- is this differences in the samples or in the measuring? If the former it is possibly a real effect, if the latter there can be issues with removing instrumental broadening. For both there can be errors that cause these
Article An evaluation of Diffraction Peak Profile Analysis (DPPA) me...
Dear Mr. Sarmah,
Using Debye Scherrer formula you can calculate crystallite size. Using Williamson-Hall plot you can calculate crystallite size as well as strain. From FESEM, you can calculate particle size, strain and agglomeration. The variation in size is because of the above mentioned factors. For finding particle size from FESEM micrographs manually, you can use the software "ZEN". I have done so and it gave good results. The published paper of Ru doped ZnO nanocrystals has indicated that.
Best of luck and regards.
Doing an analysis where the results are comparable is important and I beliv you are facing difficulty in relating the results. The vast differences which is arising when calculated by DS formula and WH plot can be mainly attributed to the synthesis process. In case the strain is large an appreciable difference in results is possible. As you might be knowing the reciprocal of the intercept gives the crystallite size. Also the slope will give a rough estimate of the strain. Incase the slope is almost nil then the variation in the result will be minimum.
Now question arises what is the solution to this. In case you have a powder solution, before you put it under test using XRD do a proper crushing and elevated heating of the sample. Experimentally it has been observed that ball milling is a process which help in reducing the strain and then the crystallite size will be comparable to DS result when observed using the WH plot. In case of thin films possibly further optimization of the system is needed before deposition.
My response might not be complete in all aspects but I hope it helps.
Regards
it will be better if you able to take TEM of the nano particles and compare the average size of the particles with XRD data.
Hi Babak,
You are correct, but it is worth noting that WH and other peak profile techniques are bulk measurements that use info from many grains. So unless one expects differences in size and strain in different directions or parts of the sample, the results of one sample should be characteristic. If there are significant texture effects, such that not many grains for a particular diffraction peak are sampled, this could be a cause for repeats. This can be checked by comparing the peak intensities with 'expected' values.
And as Pranab and Arandam say secondary measurements are always a good idea with these type of techniques.
Tom
- Badges
- Science topic
- Similar topics
- Chemistry
- Nanotechnology
- Nanostructured Materials
- Nanoparticles