The use of a stack of metal sieves followed by a ro-tapping procedure is probably the most traditional method for measuring particle sizes of dry particles.  Several considerations that should be applied when using sieves will be discussed in this article.

Standard Sieve Sizes

Wire screens used in standard laboratory sieves have precisely defined square openings through which particles can pass.  In a sieve analysis, the size of a particle is defined by the size of the smallest opening through which it can pass and the size of the next smaller opening through which it cannot pass.  Sieves used in laboratories commonly cover the range from about 1/4" down to 38mm.  Large industrial screens have openings as large as 5" square, and standard metal sieves can have openings as small as 20mm.  Specialty sieves can go even smaller.  Depending on the powders being analyzed, the largest and smallest sieves in the analysis stack can be chosen to cover appropriate size ranges.

Sieves are designated by their mesh numbers and/or their opening sizes.  A 200 Mesh screen, for example, has 200 wires per inch forming the screen, and it defines an opening of 75mm.   A 100 Mesh screen has 100 wires per inch and it defines an opening of 150mm.  A 10 Mesh screen has 10 wires per inch and it defines a 2mm opening.  The smaller the mesh number is, the larger the opening size will be.  Both mesh numbers and opening sizes are routinely used to define sieve sizes -- it depends to whom you're speaking.  Be prepared to use both. 

All laboratory sieves in the commonly used range below 0.25" form a fourth root of two size series.  That is, the ratio between any two adjacent sieves is approximately equal to the fourth root of two.  Using every other sieve in the list forms a square root of two series (two such series are possible), and using every fourth sieve in the list forms a factor of two series (four such series are possible.)  The ratios between adjacent sieve sizes in the last two examples are (approximately) the square root of two and two, respectively.

For example, a fourth root of two series includes sieves with openings that are 500mm, 425mm, 355mm, 300mm, 250mm, and 212mm.  The ratio between any two adjacent sizes is approximately the fourth root of two.  Square root of two series from these same sizes can include 500mm, 355mm, and 250mm, and 425mm, 300mm, and 212mm.  Factor of two series from these same sizes can include 500mm and 250mm, and 425mm and 212mm.  Whichever series you choose (fourth root, square root, or factor of two), all sieves in the series covering the size range of interest should be used -- with none missing and none extra.

Charting Sieve Analysis Results

The results of sieve analyses should be plotted as histogram percentages (percentage in each size class) versus size on log-log axes.  Each size class should be defined as a bar (a size range within which the particles reside.)  When a complete fourth root, square root, or factor of two series is used, all sizes classes (i.e., all bars) shown on a log size axis will have equal widths.  When sieves are missing or extra sieves are present in the analysis stack, some size classes on the chart (i.e., some bars) will be different widths and then interpretation of results will be difficult.

Analysis results can also be plotted as Cumulative Percent Finer Than (CPFT) versus size (again, on log-log axes.)  This is an excellent way to show the percentages of particles smaller than any specific size (which is useful for production controls.)  But when individual powders are analyzed and then their results are combined to obtain the distribution of the whole body, histograms should be used.  Interpretation of results is easier when histograms (rather than CPFT curves) are used.  For example, it is simply easier to predict the shape of the combined histogram when two histograms (and two corresponding powders) are added together than it is to predict the shape of the CPFT distribution when two CPFT curves must be added together.

Sieve Analysis Procedure

A stack consisting of six or more sieves, covering the size range of interest, should be selected.  When larger numbers of sieves are desired, analysis procedures are conducted in several steps using the coarsest screens first and finer screens later.

For example, when six sieves are used in a stack, the largest must be on top, and all sieves must be placed in size order from large at the top to small at the bottom.  A pan goes on the bottom of the stack.  The powder sample is then placed on the top sieve and the whole stack is covered with a lid.  The whole stack (lid, sieves, and pan) is then placed on a ro-tap device which shakes (rotates) and taps (jars) the sieve stack so particles can drop through the screens to report in their appropriate size class.  At the completion of ro-tapping, sieves can be gently pulled apart and the amount of powder on each sieve can be measured and recorded.

When more than six sieves are used, the largest six are ro-tapped first.  Then the powder in the pan is transferred to the top sieve in the next smaller stack, which can then be ro-tapped.  Powder in the pan can then be transferred to the top of the next smaller stack and the process can be repeated, etc.

Wet versus Dry Sieving???

Ultra-fine powders tend to cling to coarser particles -- which produces errors because the masses of these fines report in the wrong size classes.  Wet sieving does a better job of separating such fines from the carrier coarse particles because water has sufficient mass to dislodge and flush the fines off the surfaces of the coarse.  When very clean, precise, separations are required, wet sieving may need to be considered.

This is typically a minor problem which is ignored in most cases.  But when precise details are required, wet sieving can be performed.  Wet sieving does not require ro-tapping because the water flowing through the stack performs the separations.  Special lids which connect to the water supply are available for use to perform wet sieve analyses.

'Blinding' of Sieves

One of the biggest problems when performing sieve analyses is blinding of the sieves.  Blinding occurs when there are so many particles on a sieve that the holes are blocked by large particles and the fines cannot pass through to report in their appropriate, characteristic size classes.

There is a fine balance required here.  Sufficient powder must be used to produce good statistical results, but you don't want too much powder which will cause blinding.  Technicians must pay attention as they are performing each analysis to insure blinding has not occurred.  To obtain good statistics without blinding problems, they can adjust the starting mass of powder up or down, accordingly.

Non-Spherical Particle Shapes -- What Is A Characteristic Particle Size?

Many particle size analyzers use equivalent spherical diameters (ESD) to define particle sizes.  ESD can also be used for sieve analysis results, but one must remember that sieve analyses define size by determining which size hole is the finest through which each particle can pass.  A 6" long, 0.25" diameter pencil can fit through a 0.25" square hole (theoretically in a perfect world, it can.)  But does 0.25" properly characterize the size of a pencil?  0.25" is NOT characteristic of most pencils -- because we know pencils are usually long -- much longer than 0.25".  Many pencils are discarded when their eraser has been used up -- which frequently occurs long before the pencil becomes very short.  4" or 5" may be a more characteristic measure of a pencil than 0.25".  Proper characterization of a pencil, however, requires both length and cross-sectional size to be specified.  Sieve analyses cannot make such measurements.

Similar to pencils, fibers, plates, and other non-spherical particles will typically not report in characteristic size classes.  Which size class is actually characteristic of a non-spherical particle?  A good question!  Implied in this question is the use of a single length to characterize each particle.  If the diameters of a system of fibers are important, sieve analysis should not be used.  Similarly, a 1.414" diameter thin plate or disk can fit through a square hole with edge length 1".  In both fiber-shaped and plate examples, the magnitude and type of jarring of the particles determines how well the particles will bounce around on the metal screens and that determines the probability that each particle will line up properly so it can pass through each screen.  It's highly probable that most fibers will not pass reach the sieve that properly characterizes their diameter.  It's also highly probable that many round discs will pass through the sieve that properly characterizes their diameter.  For these reasons, it is important to pay close attention to particle shapes when performing sieve analyses, and to consider particle shape when interpreting results.

Combining Sieve Analyses with Other PSD Analyzer Results

Until recently, particle size analyzers that could effectively measure the whole range of important sizes were not available.  The results from fundamentally different types of PSD analyses have been, and will continue to be, frequently combined.  One must be aware that different types of analyzers measure different characteristic dimensions.  Sedimentation measures ESDs of settling particles.  Laser techniques measure ESDs by laser scattering of tumbling particles.  Sieve analyses measure the sizes of square holes through which the particles can or cannot pass.  Electrical resistance techniques measure ESDs by electrical resistance changes of the particles as they pass through a pair of electrical contacts in an orifice.  None of these size definitions are exactly equivalent. 

What is the correct analysis technique to use? 

This question cannot be answered.  There is NO single, correct technique to use in all cases.  Each technique has its own advantages, disadvantages, and best applications.  Within a single plant, whichever technique is used should be applied consistently to all particles.  Consistency is the key word!

When size results from different analyzers are (must be??) combined, discontinuities usually occur at the sizes where results are joined.  This is common, and it is a fact of life with which we must live.  There is no way around this problem because a 1mm particle measured by any one technique will invariably not be exactly 1mm measured by any other technique.  Even in two different labs with identical PSD analysis techniques, a 1mm particle measured by one technician, one set of sieves, or one instrument will frequently not be a 1mm particle when measured by another technician, another similar set of sieves, or even another similar instrument.

What Is The Actual Size of a Particle? 

This is another trick question because there is no way to know the actual size of any particle.  Each technique uses a different measurement process and each produces a different characteristic size.  The complete details of the size and shape of each particle must be measured and defined to properly characterize any particle.  This usually requires several characteristic dimensions -- and the more accurately you want to define the exact shape of a particle, the more dimensions need to be measured and recorded.  The true definition of a particle's size and shape is simply not possible when a single size characterizes the particle (as it does in most available size analysis techniques.)

Another part of this question deals with sampling, and yet another has to do with analyses performed by other labs.  No two samples are ever truly identical, and no two labs will ever produce identical results for the same sample.  (If results appear to be identical, go celebrate -- because you've experienced a truly rare occasion!!)   So what is the actual size of any particle or system of particles?  We do not know!  We develop and use the best techniques available to us and we apply those techniques consistently from sample to sample.  That's the best we can do.

Remember -- in the broad sense, lab procedures should be characteristic of production procedures.  Sometimes, lab procedures must differ from production procedures.  But PSD analysis procedures for all samples should be consistently applied.  Use common sense.  Think about how you're doing what you're doing and why you're doing it.  Follow the fine details of standard operating procedures very carefully.  Specific details are usually there for a reason (even when no one seems to know what that reason is.)  Finally, and once again, Be consistent! 

Conclusion

Sieve analysis is a commonly used PSD measurement technique.  Sieve analysis will remain a commonly used procedure, even with the advent of modern, fully-automatic analyzers.  Use the technique with confidence. But be aware of its strengths, weaknesses, and limitations.

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.