Dear all, I performed xrd diffraction on a sintered sample of alumina-mullite and after that I crashed the sample and performed the xrd on the powder.
The patterns that I obtained are different regard the mullite peaks: in sample pattern the relative height of mullite peaks are in agree with the theorical peak list; in powder pattern instead the mullite peaks highlights a wrong distribution of the heights. Do you have any idea of the reason? thanks a lot
I assume you used a Bragg-Brentano type diffractometer ? Please be more specific on the experimental settings. Instrument type, sample preparation Wave length, slits, sample size etc. Is the intensity discrepancy a simple function of 2Theta or does it change with HKL? Best provide plots of the two measurements.
There are several differences that can come to mind.
-Assumption one, the size and thickness of the samples in the two experiments is similar.
A likely source of intensity variation is preferred orientation. You might be luck that the sintered sample has all grains in a random orientation distribution. The crushed sample might easily undergo preferred orientation if you pressed the sample into a very nice flat surface within the sample holder. Almost impossible to avoid. Try to prepare the sample several times and carry out individual measurements. If the intensities vary from measurement to measurement preferred orientation is the likely culprit.
- Sample size / thickness differ
Make sure that the slit setting of the instrument are such that the footprint of the primary radiation is always smaller that the sample size. This is especially relevant at smaller 2Theta. With slits that are too large, you should likely have 2Theta dependent intensity variations.
In addition to Prof. Neder's answer I recommend checking several mullite-specific aspects related to composition and unit cell parameters.
How do you compare your measurement with the "theoretical peak list"?
# On what model is the theoretical peak list based?
Note that mullite forms a solid solution and the composition depends a lot on the synthesis. If you treated your sample at ~1200° to 1400° C you would expect the so-called 3/2-mullite with approximately 60 mol% Al2O3 and 40 mol% SiO2. Make sure that your unit cell is chosen in such a way that a is smaller than b. Also make sure that as a theoretical reference model you choose an alumina mullite reference with a similar composition. Useful references for 3/2-mullite are Balzar & Ledbetter (American Mineralogist (1993) vol 78, pages 1192–1196) and for 2/1-mullite Angel, McMullan & Prewitt (American Mineralogist (1991) vol 76, pages 332-342).
# Note on lattice parameters:
The length of the unit cell parameter a is very close to b. Reported lattice parameters are rougly in the range 7.5 Å < a < 7.8 Å, 7.6 < b < 7.7 Å, c is about 2.9 Å. For mullite with an alumina content of less than ~75 mol%, the usual setting of mullite described in space group Pbam is so that a < b. For high-alumina mullite (not expected in your case, requires special synthesis conditions like sol-gel or chemical vapor deposition) b > a. There are several reviews on the crystal structure of mullite, e.g. Schneider et al (2015) Journal of the Americam Ceramic Society, vol 98, pages 2948-2967.
# Note on intensities:
Assuming that the unit cell is correctly set, the intensities still vary quite a lot as a function of the composition. If all these mullite-specific aspects are clarified and there is still a deviation between the observed and expected intensities, then you should again check in detail the aspects mentioned by Prof. Neder.
Hi Emma Barbareschi
GIXRD system operates based on transmitting a collimated X-ray beam onto the surface of the sample; the beam is scarred accordingly to the structure of the material(s) within the film or powder. The scattered beam(s) is detected by a detector. The collimated beam is at an angle θ with respect to the plane of the sample. This incident beam undergoes diffraction when interacting with the crystalline / micro-crystalline segments of the sample thin layer. This phenomenon follows Bragg’s law.
So when you crashed your samples to the powder, the beams are going to be diffracted by the staggering molecules in a different conformation and physical planar position which implicating the scattered angle θ respectively so probably peaks would have a number of red or blue shifts with sharper (lower FWHM) intensity instead of one unique peak with compare to relatively theoretical mullite peaks.
I've seen this observation in our study on OSCs thin-film structural analysis in a comparison between powder and film samples.
I hope that would be helpful.
Dear prof. Neder, thanks for you reply.
Here the experimental setting that I used.
- Bragg-Brentano type diffractometer (Empyrean)
- Cu radiation (detector HR)
- Tube condition: 40KV 40mA
- Samples mounted on spinner holder
- Sample holder: powder crashed manually, mounted in ring shaped holder with inner diameter 12mm by back mount technique; bulk sample mounted in a holder for front loading sample (10x10mm)
- Slits used: mask 5mm, soller ½ antiscatter ¼
- Detector Pixcel 1D
This morning I prepared several times the powder of a sample and carried out the xrd pattern. I tried also to mount the sample with the front mount technique. Every time I mounted the sample, I obtained a different diffraction pattern.
Attached two examples: as you can see, I obtained different peaks distribution, and I don't find a relationship with a specific preferred orientation.
I am in trouble: it's quite difficult to perform xrd analysis on the bulk sample due to the samples are not completely flat and the dimension in some cases are too small. For those reason I prefer, if it's possible, to perform xrd analysis on the powder.
In addition I should quantify the amount of mullite in several samples but if I don't obtain a "real" xrd pattern, I can't perform a real quantification by Rietveld.
I'm open to suggestions if you've got any idea to solve this problem.
thanks a lot for any help,
Emma
Hi
Dear Emma Barbareschi
I didn't understand the difference between bulk & powder XRD patterns!
Are they have the same 2teta with different intensities?
(Can you share patterns?)
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science
- Reasoning