How much sample do I need?

Written by: Sarennah Longworth-Cook

One of the questions I am regularly asked by my Mastersizer customers is how much sample do I need to measure?

I get the feeling people are hoping for a simple answer, such as 1g, 10ml etc. Unfortunately, the science of sampling is not so straightforward…

An expert in sampling, Francis F Pitard, says “a sample is equitable when it is reproducible and accurate enough to satisfy both the supplier and the customer”. I take this to mean the results need to be reproducible AND have RSDs for the percentiles that satisfy the guidelines in the ISO13320:2009 standard of less than 3% for Dv50 and less than 5% for Dv10 and Dv90.

But this doesn’t help my customer, who is still scratching her head and wondering how much material that’ll be.



One of my colleagues here at Malvern has done some calculations on a typical material with a density of 1.5gcm-3 which we can use as a rough guide.

Fortunately it seems that the dumper truck is not required!

For more information about the calculation and to look into sampling in more detail, please see this document and this webinar by Alan Rawle (who created the graph).

For more about reproducibility in laser diffraction systems see this document and for a guide to method development, which includes sampling recommendations as well as other considerations, please see this document on wet dispersion method development and this document on dry powder