Protein nanoparticles < 300 nm present peak around 5000 nm in PSD (see attached file). This peak represents a very low volume (%). It is important to remark that peak around 5000 nm appears in filtered samples (pore size: 450 nm).
Dear Osvald
I suppose you should check your equipment because above 2000 nm Brownian motion is neglectable and no pick is expected.
Best regards
Petar Dokić
Thank you Petar.
The instrument is a Zetasizer Nano ZS90 (Malvern).
Size measurement from 0.3nm (diameter) to 5 microns using 90 degree scattering optics (see link for more information)
Why the peak around 5000 nm appear?
Is it inherent to the intrument or the sample?
http://www.malvern.com/en/products/product-range/zetasizer-range/zetasizer-nano-range/zetasizer-nano-zs90/default.aspx
Ostwaldo
After looking at PSD you enclosed I would say a pick at 5000nm do not come from your proteins it is inherent to the instrument.
Also, Your curve is rather broad one. If you tell me more about your protein sample may be I get some idea to suggest you to deal with.
Petar
That peak might come from any dust impurity can be in your sample preparation. To get rid of this impurity, try to prepare your sample in a hood, or not use gloves, use cap for your sample and as well as water/solvent source. More importantly, use appropriate filter for your sample.
Dear Dr. Osvaldo Sponton,
I think the peak around 5000nm may be produced by baseline error of correlation function data for some reasons such as dust. The result may be improved by some inversion algorithm or stragies.
Second, the peak around 5000nm is intensity based PSD, if it is change into volume or number based PSD, the peak will become smaller and even be neglected, according to Mie scattering.
Best wishes,
Xinjun Zhu
I also think this peak is a "spurious peak". Particles of 5 µm show a very slow Brownian motion. A trace of dust may play a role, but the peak may also arise from problems with the deconvolution of the autocorrelation function due to noise in the baseline. Maybe the following publication will provide some help:
Weiner BB, Tscharnuter WW
Uses and abuses of photon correlation spectroscopy in particle sizing
ACS Symposium Series 332 (1987)48-61, Amer. Chem. Soc., 1155 16th ST, NW, Washington, ISSN: 0097-6156 DC 20036
Anyway, even if this is a "true" peak, it represents a really tiny trace of the material considering the strong dependency of scattered light intensity on particle radius.
Best,
Harald
I agree with next part of Harald Zänker's answer as most probable
A trace of dust may play a role, but the peak may also arise from problems with the deconvolution of the autocorrelation function due to noise in the baseline.
I realise that this is an old discussion but to add a little extra information from the answers above, such a peak is often associated with a baseline noise of the correlation function. This may be caused by dust as mentioned, but also by stray light scattering.
Switching to a glass cell with better optical quality instead of a disposable plastic cell may make this peak disappear.
If your sample is of a particularly low concentration and therefore the detected scattering intensity is quite low, you may also see baseline noise which may be corrected for by increasing the correlation time.
It is entirely possible to measure D of 3 micron spheres using light scattering spectroscopy. I have done it with a conventional correlator. The short form analysis is that your data analysis software is not very effective, no matter that it is widely used. The number of parameters is extractable from a DLS spectrum is small, say five. You may be looking at a very little dust contamination. As a practical matter, disposable plastic cells, carefully pre-cleaned, are likely to be better than glass cells if you are worried about dust.
Said differently, perhaps your data analysis scheme is being unreliable for some reason.
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Powders