One of the more recently developed methods of particle size analysis is based on the physics of laser light scattering.  The author's first introduction to this type of instrument was in the early 1980s on the coal/water slurry project at Alfred University.  Laser scattering instruments then were relatively new, relatively large, rather expensive, and not particularly capable.  Today's instruments, although based on the same principles, are vastly improved and much more capable than the earlier instruments.

The purpose of this article is not to explain all of the details of operation of this type of analyzer, nor is it to go into great detail concerning the physics of laser light scattering.  The purpose is to highlight points of consideration that apply to any engineer, technician, or manager who runs this type of instrument, or who measures, presents, and/or interprets data collected by laser scattering particle size analyzers.

General Principle of Operation

It is fairly easy to give a simple explanation how laser light scattering instruments work.  Detailed explanations are typically not available because most such details of operation are proprietary to the particular instrument's manufacturer.  The author once asked a laser scattering company for more details on the operation of the unit.  In addition to being told that the details were proprietary, the company's representative followed that with, "Trust us, it works."  So much for detailed explanations.

Following that exchange, this author learned two things:  (1) ... not to worry about the physics of operation of laser scattering instruments -- because the information IS generally proprietary, and if the system is working, that's what counts;  and (2) ... to always maintain a healthy and appropriate skepticism about the validity of results from laser scattering particle size analyzers.  Murphy's Law applies to all such instruments, procedures, and results -- so if something can go wrong, it will go wrong. 

Similar to other particle size analysis techniques, errors frequently occur during particle size analysis measurements due to poor sampling and poor sample preparation techniques.  In this case, the analysis technique may be working fine -- but it is accurately reporting the data from poor samples.

The simple explanation for the operation of this type of instrument is:  laser scattering instruments pump dilute sample solutions through an analysis cell.  The laser beam also passes through the analysis cell (frequently at right angles to the flow of solution through the cell.)  As particles pass through the cell and the laser beam, photons of laser light are scattering at angles that are related to the sizes of the particles.  Sensor arrays are located at appropriate angles around the analysis cell to collect these scattered photons.  The instrument's computer then collects the intensities of the scattered photons, as functions of scattering angle, and calculates the particle size distribution of the particles in the sample stream. 

Beyond this, how exactly do they do it?  Each company has its own designs, geometries, procedures, methods, calculation routines, etc., which produce the differences between laser scattering instruments.  Which instrument is best?  There is no single, unique, correct answer to this question.  Each production company must evaluate appropriate particle size analyzers and make their own decisions to answer this question.

Two Different Scattering Phenomena

For instruments that measure particle sizes down into the sub-micron range (i.e., all of today's laser scattering instruments), a second analysis method is also incorporated.  The laser scattering described above requires the particles to be larger than the wavelength of the laser light.  As particle diameters approach the wavelength of the laser light, the scattering method described above no longer applies.  In the sub-micron size range, most instruments then use white light sources and filters to generate particular frequencies of light (0.4 to 0.76 micrometers) which are absorbed as functions of the diameters of particles.  Where this changeover takes place is proprietary to each instrument, but it mostly occurs in the vicinity of (and finer than) 1 micrometer. 

Since this is not a physics lesson, we will not go into the details of the different types of scattering phenomena that occur.  (The author is not a physicist anyway.)  That information simply isn't necessary if one wants to run, use, and interpret laser scattering results.  Suffice it to say that two different physics principles dealing with light scattering are applied to perform most laser scattering analyses from coarse particle sizes down into the sub-micron range.  Each different instrument manufacturer uses and applies these principles in their instruments in their own special way to produce accurate, useful particle size distribution analysis results.

The reason for pointing this out is that sometimes, data may appear unusual where changes in methodology (such as this one) occur.  When particle size analysis results look weird in the vicinity of and/or less than 1 micrometer, it could be that the particle size distribution is actually weird in that size range, or it could have something to do with the changeover in detection method that occurs in that size range.  It's more likely the first (results look weird because the sample IS weird) than the second, but it's nice to know that instruments generally switch from one technique to another in that size range.

Particle Shape Assumptions

The physics of laser scattering, upon which most such instruments are based, assume spherical particles.  Most powders are anything but spherical, but on average, this is an acceptable assumption.  As a computer modeller, the author knows first-hand that the spherical assumption is certainly the easiest assumption to use.  When particles are defined by flat surfaces (such as cubical or platy particles), however, it's possible that light reflection will occur in addition to diffraction and scattering.  This could cause increases in photon intensities reaching the sensors.  It's also possible that light will scatter off sharp corners of non-spherical particles differently than would occur if the particles were spherical and the surface radii were larger (consistent with spherical radii.)  When such phenomena occur, particles may report in non-characteristic size classes.  For systems of grossly non-spherical particles, these types of questions should be posed to instrument sales reps.  "Can your instrument accurately handle non-spherical particles?  How does it handle them?  etc."

Current Analysis Size Ranges

This author's first encounter with laser scattering particle size analyzers occurred about 25 years ago.  At that time, we purchased not one, but two instruments.  One covered the range from about 600 micrometers down to about 10 micrometers.  The other covered the range from about 20 micrometers down to about 0.1 micrometers.  With these two instruments, plus sieve analysis, we attempted to analyze the whole range of particle sizes from ~0.1 micrometers up to nearly 1/4".  (Our feed particles were generally 1/4".)

Today (in 2005) laser scattering instruments cover the range from ~0.02 micrometers up to ~2000 micrometers.  It sounds even more impressive when the range is specified as from ~200 Angstroms up to about 2 millimeters.  A single instrument can do this!  The whole analysis is done automatically using a single analysis technique in a single, commercially available instrument.  No more requirement for multiple analysis techniques!  No more requirement to combine multiple analyses!  Etc!  Similar advances have occurred in all other type of analysis instruments as well.

Duration of Analysis

Unlike some other analysis techniques which are labor intensive (sieve analysis, for example), or are limited by some other principle (sedimentation analyses, limited by Stoke's Law, can take from ~15 minutes to an hour to perform an analysis), laser scattering analyses can be performed quickly.  It depends how fast a result is needed and how long a technician, engineer, or manager is willing to wait for the data to be compiled.  Even 25 years ago, we could perform analyses on our laser instruments in a few seconds.  This still holds true today.  Such short duration sample results lack statistical significance, but the capability for fast results is there.

The trick is to set analysis durations long enough to produce statistically significant results without setting them too long so time is wasted.  Durations should probably be set to several minutes at least (taking into account instrument manufacturers' guidelines).  Even a 10 minute run is not long, considering the quantity and quality of data that can be produced in that time.

A recommendation:  Don't go nuts with the idea that quick analyses are possible!  Choose durations that produce good, sound data.  Sure, you may be able to maximize the output from an instrument by standardizing on 2 second analysis durations, but the results will not be significant and maybe not be accurate either.  Pick reasonable durations and let the instruments collect good data.  Sampling and sample prep are by far the most time consuming steps in particle size analysis.  Don't negate good sampling and sample prep procedures by running extremely short duration particle size analyses. 

Mixtures

Sieve analysis is not material- or density-specific.  The whole body, including all varieties of compositions in its components, can be analyzed by a single sieve analysis.  Sedimentation, however, requires all particles in a single analysis to be a single density.  All powders, regardless of density, will settle according to Stoke's Law.  Sedimentation analyzers can monitor the settling of a complete body sample, but unless all sedimentation samples contain particles of a single density, it is not possible to distinguish between the various components in analysis results.  For this reason, sedimentation samples are limited to particles of a single density.

Similar to sieve analysis, laser scattering has no density limitations.  The whole body can be sampled and analyzed in a single analysis in a laser scattering instrument.  Laser scattering instruments are sensitive to particle sizes and shapes -- not to particle densities.  So not only can particles of an extremely broad size range be analyzed in laser scattering instruments, but samples of the whole body (regardless of composition) can be analyzed as well.

Both sedimentation and laser scattering instruments have the same material limitation:  the powders cannot be soluble in the carrier fluid.  This limitation can be circumvented by using a carrier fluid in which the particles are not soluble.  Water, which is typically used in both types of analyses, can be replaced by alcohols or a variety of other organic fluids to successfully perform analyses on water-soluble powders.  Even when water is not used as the carrier fluid, sedimentation instruments are still sensitive to powder densities and laser scattering units still respond to particle size and shape (regardless of particle densities.) 

The Most Important Consideration:  Out of Sight, Out of Mind

This is the most important limitation of the laser scattering technique (in the author's opinion.)  Laser scattering instruments can only see particles in the size ranges covered by their sensors.  If there's no sensor covering a particular size range, no particle will ever be present in that size range.  If a particle cannot be seen, it is NOT PRESENT!  Out of sight, out of mind!  This is not a totally simplistic statement, either.  The greatest surface areas in powder samples are contributed by the finest particles which cannot be seen or analyzed by most particle size analyzers.  How fine is fine?  Sub-sieve?  Sub-micron?  Sub-0.1 micron?  Sub-0.01 micron?  Yes!  Can each particle size analyzer measure all such particles?  No.

Not everyone will own one of the very newest size analyzers.  Older instruments are certainly capable of performing excellent particle size analyses, but all such instruments are not capable of measuring to sizes as small as 0.01 micrometers, or to 0.1 micrometers, or even to 1 micrometer.  In all such cases, the finest particles which produce the greatest surface areas cannot be measured.

Another question one might ask is:  How can one guarantee that an instrument has seen and analyzed all particles in a distribution?  It cannot be done.  No one can guarantee this!  How does one prove that they have not missed any particles?  They can't!  How does one prove a negative?  It can't be done.

In the case of laser scattering instruments, unless sensors are positioned to analyze all sizes, one will never know how many particles were not included in analysis results.  The assumption, of course, is that all particles have been analyzed, but this simply doesn't happen.  Such assumptions are generally incorrect.  There are, of course, exceptions, but they are rare.  Each laser scattering instrument has a lower size limit below which it cannot see particles.  Different analyzers have different lower limits.  Some limits are 1 micrometer, some are 0.1 micrometers, some are 0.01 micrometers, etc.  Below the lower size limit in each case, particles will not (cannot) be analyzed.

Don't Use CPFT Plots -- Use Histograms

Why is it important that the finest particle sizes are not included in the results?  Why is it important to use histograms rather than CPFT plots?  Many people like to use CPFT (cumulative percent finer than) plots.  But CPFT plots are not accurate when it's not know how many particles are finer than the lower size limit of an instrument.  By definition, CPFT plots start at 0 CPFT finer than some finest size, and they grow to 100CPFT finer than some coarsest size.  When it's unknown how many particles are finer than the lower size limit of an instrument, CPFT plots cannot be calculated.  (The starting point is unknown.)  If CPFT plots are calculated anyway, they are inaccurate and relatively meaningless.

Note that CPFT plots can be calculated and printed by most particle size analyzers.  How?  It's simple!  Make the assumption that ALL particles have actually been analyzed and plot the CPFT curve.  Instead of CPFT plots, histograms, showing quantities of particles in each size class, should be used.  If there are no particles in a size class, or if the instrument cannot see particles in a size class, histogram results will be the same:  zero particles in that size class.  This is an accurate representation of the results.  When no particles are present, or none are visible to the instrument, histograms will show zero particles in the size class.  The histogram will accurately state the results and interpreters of the data can wonder whether or not particles are actually present in the channels that contain zero particles. 

Take for example an instrument that can only see particles larger than 1 micrometer.  If a sample of Bentonite, containing all particles smaller than 1 micrometer, is analyzed, the instrument will show no particles at all in the sample.  Everyone involved in the process will be able to see the cloudy sample solution and recognize that particles are actually present.  But if the size analyzer cannot see those particles, it will show the same result as if a solution of clear water had been analyzed.

Remember:  Zero percent in a histogram channel means one of two things:  (1) no particles were present in that size range;  or (2) particles were present in that size range but the instrument could not see them.

Summary

Today's laser scattering instruments are incredible!  Analysis ranges are from ~20nm to ~2mm.  Instruments with such broad analysis ranges and with full, automatic computer control were not available two to three decades ago.  Take full advantage of the great capabilities available in today's instruments.

Make it a practice to routinely analyze samples of the whole body.  Laser scattering is not sensitive to material densities, so samples of the whole body can (and should) be analyzed.

Use histograms to display laser scattering results.  (Don't use CPFT plots, which are generally invalid for laser scattering analyses.)

Use analysis durations of several minutes to insure statistically significant results.

Maintain a healthy skepticism concerning particle size analysis results.  Test materials with alternate analysis techniques if necessary to make sure that results are reasonable and accurate.  Then, with confidence, take best advantage of the wonderful capabilities of these instruments.

Miscellany

Suggested topics .... Please continue to send your ideas or questions for future topics.  Thanks.  Until next time ...

Thanks for signing up to receive Ceramic Processing E-zine
To unsubscribe, click the "Remove Me" link below.

Copyright © 2005  Dennis R Dinger

103 Augusta Rd, Clemson, SC 29631   (864) 654-5731

consulting@DingerCeramics.com

www.DingerCeramics.com

All Rights Reserved.

All Rights Reserved.