I recently presented a webinar exploring the practical aspects of sieving method – the advantages and disadvantages.  I compared the method to other techniques such as laser diffraction indicating the increased wealth of information that can be obtained with these more modern instrumental techniques and the common pitfalls and assumptions of the screening technique.

More than 800 of you registered for the webinar and I received a large numbers of interesting questions which I have been working steadily through. I thought it would be a good idea to share the questions and answers via the blog, so here they are!

By the way, if you missed the webinar, you can view the recording on our website.

May you explain the second smallest dimension for sieve measurement?

This simply arises from considering any irregular particle that needs to be specified with more than 3 numbers.  The smallest dimension would allow the particle to penetrate the screen (but would just sit there). The second smallest dimension allows the particle to pass through the screen.  In many instances when people try to think of objects passing a screen they tend to think of solids in 3 dimensions and (in this case) the second smallest and second largest dimensions are the same.

We are working on switching from sieve PSD measurement to Laser diffraction for a dry powder. Obviously, results at first glance are different, what would be the best approach to take to make the switch over?

There are a bunch of features that makes laser diffraction superb for obtaining a particle size distribution (PSD) rather than having a small number of points at the extremes of the distribution. As you said ‘the results are different’ – and are expected to be different because different aspects of the particles are being measured.

In your case, I’d be running both techniques in parallel for some time and adding the diffraction results to the specification sheet. I would see the results as different as apples and pears and no need to spend time to bend one result to fit the other (ideally the screen results should be corrected to give a true volume). IMHO, there’s no need to start a project to try and convert the results from one technique to another – you’ll end up with a lot of Excel spreadsheets and little or no insight. Run the systems alongside each other and soon you’ll start to see the benefits of all that valuable particle size distribution information that diffraction provides. Quote both techniques or specifications in the appropriate place. The important thing is that both techniques must generate stable and reproducible results or you’re lost. Focus on that aspect – getting excellent measurements – and see the difficulties of taking 10 screen measurements in a row when it’s easy for diffraction….We need to note, though, the fact that many customers may sell their products based on specifications derived using sieving, so there may be a need to map the LD result to the sieve result. This requires use of the emulation function in the MS3000 software and the absolute requirement for stable data in developing this correlation. A later webinar by Sarennah Longworth-Cook addresses this requirement on “How to replicate a sieve analysis on the Mastersizer 3000“.

For method development in laser diffraction, see: “An acronym approach to laser diffraction method development

What are D[3:2] and D[4:3] mean?

These are moment means of a particle size distribution.  The D[3,2] is the surface area moment mean and the D[4,3] is the volume/mass moment mean.  For more information see the attached and the recent webinar: “Basic Principles of Particle Size Analysis

Accurate characterization of powder coating fines (<20µm) dependent on dependent air pressure on MS2000. What is the correct value?

You are correct and the pressure size titration for dry method development does not produce a plateau. The correct value is dependent on the type of powder coating (epoxy, polyester, hybrid) and is found by comparing the (median, x50) results against the wet reference method where a plateau of dispersed material is possible.  The wet-dry examples I show in the following webinar are all actually powder coatings (slides 42 onwards where 42 is the answer to life, the universe, and everything): “An acronym approach to laser diffraction method development“.

You can wet a powder coating with a little dilute surfactant in order to get it into suspension for measurement.

We have witnessed sieve blinding (our QC target being % passing 150 microns) when we saw an increase in particles <30 microns (measured by laser diffraction). Is this something that is very common with sieves? Is there a way to get around it other than moving to laser diffraction?

Sieve blinding generally results from small primary particle sizes causing attraction and bridging of particles (indeed this has spawned an old website called ‘Why we hate TiO2’) often combined with humidity or sticking issues.

Excessive fines are easily seen in laser diffraction and can be measured in this ‘sub-sieve’ size region where blinding is much more likely to happen.  A small cohesive material is a problem for screens and hence the < 38 µm region being referred to as the sub-sieve size region.  Yes, it’s pretty common with smaller and damper material.

In some instances, wet sieving may help, but screen blocking is still common here especially in the smaller sizes.  Sieves don’t allow you to disperse the particles as diffraction does,  Indeed following the dispersion with a wet method in diffraction allows you to get a handle on the degree of agglomeration present – something that screen could never really do.

Can you briefly go over “DRY” diffraction or how to measure soluble materials?

It’s a good question but one we weren’t tasked with dealing with from the scope/abstract of the webinar – perhaps we’ll tackle in a later webinar.  We do have a bunch of material that helps you here. I’d recommend the wonderful ‘Demo at your Desk’ series to see how a dry measurement is conducted. For example Mastersizer Demo at your desk.

How can we secure the sampling for laser granulometry measurements? I mean the sample size is much smaller with laser technique. How can we guarantee that the result is representative of the product?

Yours is a vital question inherent in any metrological technique. Obviously, any instrument will measure what it’s given and in the worst case ‘garbage in = garbage out’.  The 4Q’s apply here.

What is the AQL (acceptable quality level) of the organization?  Or alternatively, what is the required or desired standard error (SE) in the measurement?

This relates to any specification that is or will be set.  This is defined by the end user

Is a bulk size (‘as is’/with agglomerates) required or is a dispersed (primary) size desired? This is the answer to the question “What is the purpose of taking the measurement?

The answer governs the energy that will be required in the measurement and is related to the end use of the material and the fitness of purpose thereof.  This is defined by the end user

What is the top end (largest size) and the polydispersity (width/spread) of the particle size distribution and the density of the material?

This is set by the material and its mode of manufacture

What is the mass of sample that you utilize in your particle size experiment?

This is set by the accessory utilized or purchased by the user and the amount of sample added. You’re in control!

I run a very popular 1/2 day  ‘Sampling for Particle Size Analysis’ course every year at Pittcon that deals with this issue in much more detail.  In dry analyses, we can use many grams of sample akin to a screen. We may be able to divide the sample up using a rotary/spinning riffler.  In all cases we’re limited to a minimum mass of sample required for a prescribed specification.  Even with a correctly divided sample if the minimum mass requirement is not met then we will end up with variation (often attributed to instrument variability) inherent from the heterogeneity of the sample.  Pierre Gy spent many years trying to get these points across.  We can also back-calculate the best standard error based on the heterogeneity of the sample.

Estimation of fundamental sampling error in particle size analysis
Sampling for Particle Size Analysis

Can you discuss obscuration and is it important for obscuration to be “in range” during the entire time that sample is being introduced into the Mastersizer 3000

The obscuration is basically a measure of the optical concentration of a sample.  If the obscuration is changing during a wet measurement (for example) it tells us that we’re losing particles from the system (e.g. dissolution).  We look for stability in the value.  If we request 5 seconds of measurement then we require 5 seconds (50000 snaps in the MS3000) of delivered measurement – if we’ve not got these then it tells us that the material wasn’t within the desired obscuration limits for the duration of the measurement. In a dry measurement, we’d like to control it in the 0.5 – 5% range typically to keep dispersion conditions stable and minimize changes of particle recombination after separation in the venturi.  In the wet we’d want to control to say +/- 0.5% – and you’re in control of how much you add…..  For smaller systems (< 1 um typically) we may wish to undertake a concentration study as multiple scattering may be an issue that needs to be balanced against adequate signal-to-noise (which is always the defining factor).

Changes in obscuration, size (x90 and x10) tell us about the system and changes in the sample during the measurement that we need to understand. Typically for QA, the 3 objectives are stability, stability, and stability – any changes are a pointer to processes occurring in the sample measurement (could be desirable such as deagglomeration or undesirable such as dissolution). These changes need to be understood – unless, of course, a kinetic study is contemplated.

Sieve blinding and detection of the degree of blindness? Can the surface area really be accurately be measured by laser with correlation to BET?

Sieve blinding generally results from small primary particle size causing attraction and bridging of particles (indeed this has spawned an old website called ‘Why we hate TiO2’) often combined with humidity or sticking issues.   Excessive fines are easily seen in laser diffraction and can be measured in this ‘sub-sieve’ size region where blinding is much more likely to happen.

Laser diffraction does not measure surface area.  However, it can be indirectly derived with a bunch of (normally invalid!) assumptions. Indeed if you’d want to measure the surface area on a regular basis then I’m sure you’d be using BET N2 physisorption.  However a conversion to the specific surface area can be made as 6/[D3,2] = SSA and the D[3,2] comes from the D[4,3] – see attached Basic Principles.  As the N2 molecule can penetrate into all the pores of a solid and laser diffraction is (in the simplest analogy) looking at the contour of the particle then the BET value is usually much larger than that derived and displayed by laser diffraction. However, there’s usually a very good correlation between the techniques and I explored this in a paper a number of years ago.

Basic Principles of Particle Size Analysis (Application note)
Micron-sized nano-materials

I analyze my particles with laser diffraction.The d 0.5 is about 5µm but the PSD changes a lot from measure to measure. Sometimes, unimodal distribution, sometimes bimodal distributions coming from the same batch. We use Tween 20 as a dispersant, also, what is it which causes blinding of sieves?

If changes are occurring during the measurement then we need to understand these. As a general guide to method development takes a look at An acronym approach to laser diffraction method development.

For small material (typically < 20 µm) we need to control the following:
Dispersion – follow the BDAS regime as outlined above in line with the 4Q’s attached. Multiple consecutive measurements are essential
Optical concentration – needs to be controlled to better than 0.5% in order to balance signal to noise against possibilities of multiple scattering
Optical properties – a bimodal for a single component is very suspicious and needs investigation. If there’s a minimum at a harmonic of the laser (1.2 µm is a classic) then this is a pointer to the extinction coefficient being mismatched

You’ll only need a surfactant if the material does not wet in the fluid you use. Avoid foaming. If you are able to send on the raw data file (.mea in Mastersizer 2000; .mmes in Mastersizer 3000) then we can provide more specific assistance.

If I have a large particle, like a cylinder, using laser diffraction the result will be incorrect because of the shape?

You have outlined the basic conundrum for non-spherical particles.  More than one number is needed to describe them correctly.  Stated in USP <776>: ‘For irregularly shaped particles, characterization of particle size must also include information on particle shape’.  All particle size techniques rely on some form of equivalent.  I refer you to the Basic Principles of Particle Size Analysis.

How can I and other colleagues access this presentation in the future?