Lost in translation: the pitfalls of method transfer

One of the most important parts of my job is aiding multinationals with transferring methods between sites. You may think that, with the SOP function present in Malvern’s products, saving methods and emailing them to the other sites guarantees easy method transfer. However this is not always the case as sometimes the ability to send the method electronically is forgotten about and paper instructions are not ideal. Often this leads to analysis conditions being subtly different between sites.

Sampling and dispersion are the key parts of a method that need to be thoroughly detailed, including any sample preparation which occurs before the sample is added to the particle size analysis system.

Sampling control

If the particles size is greater than 75 microns then sampling needs to be considered. Simply scoop sampling from the top of a container can give a 17% error (see T. Allen, Particle Size Measurement, Chapman and Hall 4th Edition (1993), Page 39). Spinning rifflers and other mechanical sample extraction devices will help here. If the variation on repeat measurements of the same sample is poor (with random variation for parameters like the Dv90), simply increase the measurement time so a consistent amount of large material is measured.

If you make a suspension in order to disperse your material, is all of it added to your sizing instrument? If it is sub–sampled, how are you doing this? Is one person’s pipetting technique the same as another person’s? It may well be, but you’ll need to explain how to sub sample the sample in the method, so everyone can do it like you do.

Dispersion and measurement

A recent method I encountered simply stated ultrasonicate the sample for 3 minutes in an external bath before measurement. There was no mention of the make and model number on the ultrasound bath, the power applied, the amount of liquid in the bath, or where in the bath the sample should be placed. These are all factors that should be considered.

Environmental factors also need to be considered when designing a method especially when ultrasound is involved. Any ultrasonic dispersion process will cause a temperature rise as well as dispersing the material. In organic solvents this can provide some light scattering (caused by the refractive index of the liquid changing during heating). This will dissipate over time but how long this takes will depend on laboratory temperature. This means that any equilibration time should be locally set at each site and not be set in stone in a method.

The same is also true of air and vacuum for dry measurements, air should be equivalently clean and dry and all vacuums should be equivalent. I once encountered a customer who had in all seriousness attached a handheld domestic vacuum cleaner (more at home in sucking up toast crumbs in the kitchen) to their equipment, and they were surprised that I wasn’t impressed (though I was more impressed than their janitor who’s vacuum they then “borrowed” for their Mastersizer!).

To conclude, always provide as much detail as you can on the key parts of a method:

• How should users take a sample?
• How should users disperse the sample?
• How should users analyse the sample?

And providing a typical result (including a plot of the scattering data expected from the sample) is time well spent, and one that the QC analysts in your organization will anonymously thank you for at some point in the future!