Volume 5  Number 4                            Dennis R. Dinger                                1 February 2007

Updates

"... for Ceramists" Series Books

The paperback version of Characterization Techniques for Ceramists is available on the Books and Downloads page at the web site!    Retail price is $29.95 plus shipping and handling. The book has 256 pages and it covers 34 different characterization techniques that are commonly used by ceramists.  Order your copy NOW!

The book sets on the web site have also been revised to include this new book.  A 3-book set of paperbacks, including one each of Particle Calculations for Ceramists, Rheology for Ceramists, and Characterization Techniques for Ceramists, is now available for $64.85 plus shipping and handling.  This is a $10 saving off the total retail price of the 3 paperback books.  A 3-book set of downloads is also available for $52.85.  This, too, represents a $10 saving off the total retail price of the 3 downloadable books.  

The E-Book version of Characterization Techniques for Ceramists is available for downloading at the Books and Downloads page of the website for $24.95.  The download is a 2.889 Mb self-extracting Zip® file for the Windows® environment which unzips to the 2.998 Mb book in PDF file format.  Those of you who order the downloadable book will want to know that the PDF book is formatted to print on 5.5" X 8.5" paper (i.e., 8.5" X 11" sheets cut in half.)

The other two books, Rheology for Ceramists and Particle Calculations for Ceramists, continue to be available for purchase as downloadable E-books and as paperback books at the Books and Downloads page of the web site.

The E-zine

If this is the first issue of the Ceramic Processing E-zine that you've seen, you can add your name to the mailing list by clicking HERE.  All back issues can be accessed from the Publications page at the web site.  For those of you whose e-mail programs don't properly show the figures in these E-zines, go to the Publications page of the web site using your web browser to open any and all issues.  All figures should open properly when issues are accessed from the web site.  Questions, suggestions, and/or requests for topics to be covered in future issues of this e-zine can be sent to QuestionsandComments@DingerCeramics.com   .

If you have friends, business associates, etc., who are ceramists, materials engineers, or any other type of engineer or technician, and they are interested in receiving this e-zine, please forward this issue to them and encourage them to sign up.  Or simply point them to the Dinger Ceramics web site.  Also -- whether you are a new or continuing reader -- please send suggestions for topics you'd like to see addressed in future issues of this E-zine.

A Well-Controlled Particle Size Distribution --- WHY?

Introduction

For the new ceramic or materials engineer or engineering student, it is worthwhile to go back and examine the fundamentals every now and then.  In this article, we will consider the paraphrased question, "Who cares about particle size distribution anyway???!!!"

There are many process and ware properties that depend on the close control of particle size distribution (PSD).  Any time you are using powders in your process, PSD is not just important -- it is critical!  PSD in ceramic bodies fluctuates all the time -- from raw material bag to raw material bag -- from scoop to scoop -- etc.  PSD is a parameter that MUST be closely controlled from batch to batch, or downstream process and ware properties will vary.  

Let me emphasize, I am not saying this "may be the case occasionally."  PSD always varies.  You simply can't ignore it.  And you can't push its control off onto your suppliers either.  They don't have the capabilities to fine tune each powder precisely for each of their customers.  They have their standards and consistency goals, but each of their customers has precise demands beyond the suppliers' capabilities and out of the suppliers' control.  My recommendation is this:  Don't even think about trying to push your control problems onto your suppliers.  You are running your process with its particular demands.  Only you know what those are, and only you can make the proper adjustments to control your process.  Notice all of the "you"s and "your"s in the previous sentence?  It is your process, and you are in control of it (or not).  The suppliers are in business to give you consistent raw materials from shipment to shipment.  It is your responsibility to make that into consistent body from batch to batch for your process.  You can't pass the buck.  

Lots of process variables and ware properties at all stages throughout production are affected by PSD variations.  The biggest of these variables is viscosity of suspensions and plastic forming bodies.  As PSDs vary, so do viscosities.  As PSDs vary, the capability of the powders to pack varies.  As PSDs vary, the distances between particles vary.  As particles move farther apart, viscosities decrease.  As particles crowd one another, viscosities increase. 

Packing capability of a system of particles is directly related to PSD.  If the natural capability of a distribution to pack is poor, a compact of this batch of powder will exhibit poor packing and high porosity.  If the natural capability of a distribution to pack is good, a compact of this batch of powder may still exhibit poor packing (if it is poorly mixed), but it should be expected to produce a very dense compact and low porosity.  

Low porosities are accompanied by small pores and low permeabilities.  High porosities may also be accompanied by small pores and low permeabilities, but they should be expected to produce large pores and high permeabiliites.  

Permeability relates directly to dewatering.  Systems with low permeabilities will dewater poorly (slowly).  Systems with high permeabilities will dewater quickly.  Note that drying is a dewatering process.

Shrinkages during drying and firing relate directly to compact porosities.  High compact porosities can be accompanied by high shrinkages.  Low porosities in a compact can be accompanied by low shrinkages.  Notice the use of the word 'can' in the preceding sentences.  A high porosity compact may not exhibit high shrinkage depending on the firing program.  High porosity dry bodies can produce high porosity fired bodies with low shrinkages when firing temperatures are low and firing times are brief.  With high firing temperatures and/or long firing times, high porosity dry bodies can be fired to produce low fired porosities with relatively high shrinkages.

In suspensions, additive chemistries and rheologies can vary considerably due to PSD and packing capabilities.  At fixed solids content, when particles can pack very well, suspensions will have relatively low viscosities because the high packing capability puts lots of distance between suspended particles.  In this case, to reach a particular target viscosity, suspensions will need lots of flocculating additives.  Again at fixed solids content, but in contrast to this, when particles pack poorly, they will produce relatively high viscosity suspensions because the fluids will fill the large quantities of pores without separating particles very much.  Viscosities will then be relatively high.  To achieve particular target viscosities, suspensions will need lots of deflocculating additives.  This applies to processes in which solids contents and final viscosities are controlled but PSDs are not.  In such bodies, the following occurs:  On days when particles pack well, rheologies will tend to be shear-thinning because the bodies are flocculated;  On days when particles pack poorly, rheologies will tend towards dilatant (i.e., shear-thickening) because the bodies are deflocculated.

Other process and ware properties are similarly related to PSD variations.  PSD is a very important, critical property in the control of any and all bodies made from powders.

                    A "Good" Distribution

Everyone talks about "good" distributions, although "good" means different things to different people.  Make sure when you hear someone talk about a "good" distribution, that you determine that person's definition of "good."  Each plant will have a "good" distribution which is the PSD that works well in their particular process.  Make sure you determine the definition of "good" in each case because one company's "good" distribution will not necessarily work in another plant or even in another process in the same plant.

                    A "Broad" Distribution

This is a term used frequently to mean a distribution that covers a very broad range of particle sizes.  A broad PSD is one that covers particles from (for example) sub-micron sizes all the way up to thousands of micrometers.  "Broad" usually implies that the distribution packs well -- but this is not necessarily the case for all broad distributions.  (Good packing is likely, but not an absolute.)  "Broad" usually implies that the distribution contains a few particles in each size category over the very wide range of particle sizes.  

Broad distributions usually have good rheologies, which means they tend to be shear-thinning.  This, too, is not an absolute.  Good rheologies result from that fact that there is a wide range of particle sizes represented and the vastly different particle sizes usually flow well and pack well with one another.  

                    A "Narrow" Distribution

This term is used to mean a distribution in which all particles are essentially the same size.  A barrel of BBs contains a narrow PSD because all of the BBs are identical in size.  Narrow distributions usually exhibit poor packing, high porosities, and when suspended, high viscosities and poor rheologies.  When narrow distributions are diluted sufficiently to exhibit low viscosities, they tend to be unstable (i.e., their particles settle.)

The three terms, "good", "broad", and "narrow", mean different things to different people.  So when using these terms, you MUST make sure that you understand what the other person means and that the other person understands what you mean.  The characteristics given above are typical of these different types of distributions, but once again they are not absolute.  Some narrow distributions are stable.  Broad distributions can be dilatant and narrow distributions can be shear-thinning.  (The "how"s and "why"s of these characteristics and phenomena are beyond the scope of this article.)

How to Control PSD?

There are two ways to control PSD.  One is to control milling processes to produce desired distributions.  The other is to mix several PSDs to achieve the final, desired distribution.

                    Control the Milling

This is a sensitive control process because it is easy to over- and under-mill PSDs.  Most companies use large, batch, tumbling ball mills.  If the solids contents, powder contents, or chemical contents in the mill vary slightly (due to normal variations), milling properties and product PSDs from fixed milling times will vary.  Actual distributions can be analyzed at the end of each milling cycle.  One can pull samples for analysis from each mill to guarantee that the PSD is correct before the mill is dumped.  If the PSD is still too coarse, milling can continue.  Then, the analysis can be repeated.  When the PSD is too fine, however, no corrections can be made.  Dump the mill and move on.

Checking at the end of each mill is difficult and if rushed, it can be very inaccurate.  Can one guarantee that the sample pulled out of the mill is representative of the whole mill?  Then, while waiting for results, how long must the mill sit there idle before the decision is made to mill longer or to dump the mill? 

Instead of batch milling, it is also possible to use continuous milling, but most ceramic companies do not own or use continuous tumbling ball mills.  Continuous stirred ball mills (for fine milling), however, are common.  When continuous mills are used, one can pull samples of product regularly for analysis and make adjustments to the milling conditions as the mills are running.  

                    Mix Several Product Distributions

The best way to control milling is to mill some mills coarser than the target and other mills finer than the target.  Then, the products can be analyzed and PSDs can be controlled by careful blending and mixing based on analysis results.  For example, if a 40 micron mean size is desired, mill some powders to 50 micron median sizes, and other powders to 30 micron median sizes.  Then, after each batch product is analyzed, one can carefully mix some of each to achieve the desired 40 micron mean size.  Computer programs can easily calculate the exact percentages of each product to be mixed to achieve the target PSD.

This is the best method to control PSDs.  It applies to incoming materials from suppliers as well as to locally milled products.  There are, however, several requirements.  Lots of holding tanks are required and lots of PSD analyses are required.  Each mill product should be dumped in a holding tank until the PSD analysis results are available.  This requires many holding tanks and lots of extra time.  But the results can be tightly controlled PSDs.

Controlled mixing of several distributions is not the typical manufacturing method employed in the ceramics and materials industries, however.  Most processes go like this:  Fill the mill with the required amounts of all body ingredients, water, and chemical additives; mill for a fixed time (e.g., 150 minutes); dump the mill products into process tanks; and repeat for the next batch.  By contrast, the mixing/analysis method requires:  many more storage tanks for the batch products; milling of individual or carefully selected combinations of materials; many PSD analyses; and mixing the resulting products in a blunger to achieve body compositions which contain target PSDs.  

The important question is this:  How precisely must you control your PSD?  If your process requires a high level of precision and control, then the mixing method is the only method that can achieve the required level of control.  If more substantial variations are OK, then the "Control the Milling" method may be sufficient. 

Each company and process engineer must determine how close is "close enough".  If variations in product PSDs from the "Control the Milling" method are acceptable, then by all means, use that method.  If variations in product PSDs are too great, the "Mix Several Product Distributions" method will be required.  

Sometimes it is easy to implement the mixing method -- or variations on the mixing method.  For example, one company wanted to produce batches with 4 micron median particle sizes.  It just happened that their incoming raw materials were already characterized for median particle sizes, so they just needed to properly select incoming powders to be used in each batch.  When they use some 5 micron median size powders, they just need to also use some 3 micron median sizes to compensate.  

Sometimes it is not easy to implement the mixing method.  Then it is up to the process engineer(s) to determine how best to proceed.  Remember that with today's particle size analyzers, it is easy to pull PSD results into MS Excel® or other similar programs to calculate percentages of several different PSDs to be mixed to achieve the target PSD.

The Problem with Out-of-Control PSDs

There are two main problems when PSDs are out of control are:  first, strange problems come and go, and second, suspension rheologies vary with viscosity.

                    Strange Problems Come and Go

As PSDs vary, strange problems come (without warning) and go (without correction.)  When problems show up in more-or-less random fashion, and they last a day or two, or a batch or two; and then they disappear for no apparent reason, these problems could be due to PSD variations.  

We have taught quite a few seminars over the years, and everyone has such strange problems in all of their processes.  There is no need to qualify these statements with words like "almost everyone" and "almost every process" because such variations are present in all processes.  I know of no examples where batches and processes are under perfect control.. 

When such variations happen, one must look for the underlying problem which is not very easy to do.  Out-of-control PSDs can cause variations that last a few batches and then they disappear.  Just about the time one has started to study the problem, it may disappear on its own, without correction.  This is the nature of out-of-control PSD problems.   

          Rheologies Vary With Viscosity

This problem is more subtle.  Many companies control their processes to a fixed viscosity at a fixed solids content.  When PSDs vary without control, body rheologies vary widely.  The reason for this is that at a fixed solids content, PSD variations will produce viscosity variations.  When viscosities are high, deflocculants are added to achieve target viscosities.  This can produce severely deflocculated suspensions at the target viscosity as measure by process viscometers.  At the other extreme, when viscosities are low, flocculants are added to achieve target viscosities.

This produces bodies or suspensions that can be extremely shear-thinning or extremely dilatant at the target solids content and the target viscosity.  Processing properties will vary widely when using suspensions or bodies that have major rheological variations at target process conditions.

It is also more difficult to relate such rheological variations back to PSD variations, but PSD variations are frequently the principle cause of such problems.  PSD variations can have the biggest effects on body properties and can easily cause problems like this.  Just because the problem is with forming properties, or rheologies, or viscosities, does not rule out PSD as the culprit.  PSD variations are the underlying causes of many process problems.  Consider this:  If you hand a technical assistant a suspension and request the amount and type of additive to achieve the target viscosity -- he/she can usually give a very precise answer that successfully achieves the goal.  But if the PSD of the sample is out-of-control, their precise solution to achieve the target viscosity may cause other process problems.

Once the PSD of a body is set, it is difficult to go back and make adjustments to it.  One can adjust and re-adjust additive chemistries, and adjust and re-adjust solids contents, but it is impossible to adjust and re-adjust PSD after the body is mixed and sent off to the process.

Summary

The particle size distribution of a body is one of the first parameters that is set in any process.  Once mixing is complete and the body is sent off for further adjustment and processing, the PSD of the body cannot be changed.  Chemistry can be adjusted.  Solids content can be adjusted.  Most such changes are reversible.  Added too much deflocculant?  Add a little flocculant.  Solids content too high?  Add a little water.  Solids content too low?  Remove some water or add more dry powders.  Etc.  But once the PSD is fixed -- that is, once the batch has been mixed and sent to holding tanks, PSD is fixed and no longer adjustable.

PSD variations in bodies come from PSD variations in incoming raw materials and from PSD variations in mill outputs.  Since PSD measurements usually take an hour at best, it is difficult and/or time-consuming to analyze PSD at each step along the process.  Notice:  PSD measurements are not impossible to perform -- they are just inconvenient and/or time-consuming to perform.  So when PSDs are not measured in favor of moving on with the process and taking the chance that the PSD is OK, one has moved out of the realm of control of PSD to the realm of non-control of PSD.  If one is not actively controlling PSDs, one is accepting out-of-control PSDs.  If you are not willing to take the necessary steps to control PSD, then you are saying that you are willing to accept out-of-control PSDs.

In this author's opinion, particle size distribution is the most important variable in any powder suspension or forming body.  To treat this step lightly, when it is one of the first to be controlled in any process and it is the most important variable to ceramic processing, is to accept a whole variety of random problems that come and go later in the process.  In other words, to treat this first variable lightly is to invite a whole bunch of later problems that come and go and that cannot be controlled at that point in the process.  If PSD is the underlying problem, and it is out-of-control, then the best we can do is try to control symptoms rather than attack and eliminate the underlying problem.

How important is PSD control?  It is that important!!!