Fly Ash

The desired Planetary Mill PULVERISETTE 5 premium line is a high energy planetary mill. Typically, such a type of mill is comminuting brittle samples extremely fast with much more of impact force (> 50x g-force), compared to classic planter mills or pearl mills (22x g-force for the comparable classic line model). Because of the additional impact force, usually samples are not ground for hours as requested.

With a start size with ~ 75 µm for the max. fly ash (determined by static light scattering), we recommend using 1 mm Ø grinding balls. With a factor > 10x, we will be on the safe side and also the largest particles will be ground. We recommend using a proper solvent which is not affecting the sample, nor grinding media (zirconium oxide in our case) or the gasket. Also the solvent should have a higher boiling point (> 65 °C to avoid over pressure) and a lower viscosity. We added about 104 grams of isopropyl alcohol (IPA) to maintain a motor oil like slurry, for optimum grinding results.

The mill was programmed to comminute in cyclones with maximum speed (800 rpm) for one minute, followed by 10 minutes of programmed pausing cycles for chilling. The outside temperature should be checked on a regular basis and grinding times or pausing times should be adapted when the outside of the bowl reaches temperatures close to the boiling point of the picked solvent.

We checked the viscosity and particle size distribution in intervals and observed an immense improvement in fineness after the first 10 minutes of grinding time. The d50 sunk from < 14 µm to < 1 µm within this time frame. Also the viscosity was still ok and no further isopropyl alcohol was required to add. With a d90 < 2 µm, it would be possible to exchange the grinding balls to the smallest 0,1 mm dia. zirconium oxide grinding balls already. We left the 1 mm grinding balls inside to check if particle size distribution still improves and picked a small sample from demonstration before we proceeded grinding. After a total grinding time of 30 minutes, the d50 sunk to < 720 nm with a d90 < 1,9 µm. The static light scattering wasn’t showing a significant improvement of larger particles anymore. A small sample was picked for demonstration. With the actual level of filling, we would estimate that up to 200-250 grams of sample can be ground this way, per bowl. We exchanged the grinding balls to 0,1 mm Ø zirconium oxide (see result 2).

After 30 minutes of grinding in result one, now the 1 mm grinding balls were exchanged against 0,1 mm Ø zirconium oxide grinding balls. The used 1 mm grinding balls were flushed with fresh isopropyl alcohol to regain a maximum of sample. About 26 grams of isopropyl alcohol were added to flush the 1 mm grinding balls and maintain a motor oil like viscosity again.

The sample was ground for a total grinding time of 45 minutes (+ 15 minutes with 0,1 mm Ø grinding balls) and the particle size distribution was checked again. A d50 < 510 nm with d90 < 900 nm was determined.

We proceeded, grinding for a total grinding time of 60 minutes and checked the particle size distribution again. The d50 value sunk to 450 nm with d90 < 800 nm.

After 90 minutes of total grinding time, the particle size distribution could not be determined anymore, the smallest particles agglomerated again and were detected as larger fragments. This effect is typical for nanometer fine ground samples and happens often. It would require a proper additive to hinder that such find particles reattach. Those additives can be surfactants (cationic, anionic, amphoteric), in watery slurries, also the pH can be modified by salts, acids, bases or buffers to stabilize a slurry without using of surfactants. We aborted our trial at this point and packed the sample best possible.

The used equipment was cleaned by short time comminution of sand and water (2 minutes at maximum speed). Afterwards, the grinding bowl and balls were flushed with fresh water and dried afterwards.