Gypsum rock sample

The large quality of gypsum can be ground directly with our Disk Mill PULVERISETTE 13. To achieve a fine result, small gap settings are required. The disadvantage of a 0,05 mm gap (with a majority of output < 250 µm) would be much more smeared residues inside the grinding chamber. We estimate that a setup of 0,2 mm gap would achieve > 80 % < 250 µm results with only a few traces of smeared particles.

The complete sample was fed into the mill; the grinding sound started to vanish after ~ 3 minutes. The mill was stopped after 3 ¾ minutes. At this point, a bit of sample remained loose inside the feeding funnel, also on top of the fixed disk. Electrostatic charged particles stay attached to the housing and could be brushed off easily. Also after this trial, we found a few traces, compressed and smeared where the ribs of the disk will be crush samples before fine grinding.

Most likely this will happen per se when a sample like gypsum contains crystalline water.

The original rock sample was too large to be used for a direct comminution. It was required to pre crush. We adjusted a gap setting of 1,5 mm which would open up to ~ 3,5 mm during operation.

After feeding of the rock, the grinding sound was smoother; the grinding sound vanished after a total grinding time of 20 seconds. Minimum residues of gypsum were visible on the surface of the crushing jaws. The sample output was showing a maximum length about 10 mm.

The pre-crushed rock sample of result two was used for this trial. The thickness of sample allows using 20 mm Ø grinding balls which will speed up the processing (compared to 6x 30 mm Ø grinding balls when larger fragments are fed). We used a new bowl made of zirconium oxide with smoothest zirconium oxide grinding balls.

The sample was ground for only one minute with the maximum speed setting (400 rpm = 22x g-force). After one minute only, the sample was ground very fine. The sample already started to stick to the grinding balls and bowl. A longer dry grinding of the sample is not recommended. To improve the fineness, we recommend a grinding in suspension. By this way, fine ground particles will not become compressed by the flying grinding balls. A motor oil like viscosity would work best for a short grinding time and least abrasion. By standard, water, benzine, isopropyl alcohol etc. are often used. To avoid reagglomeration of fine ground samples, we recommend a rapid using a rotary speed evaporator.

If dry grinding is mandatory, it is possible to delay sticking somehow, when dry grinding agents (few droplets of ethylene glycol or methanol) are added. It is also possible to grind with lower speed setting and separate the fine fraction in short intervals by sieving.

Cleaning after this trial was performed by comminution of a proper portion of sand and water. After two minutes of grinding, the equipment was flushed with fresh water and dried afterwards.

For our last trial, we used the pre-crushed sample of result two again. The rest of the sample (~ 180 grams) was placed into a 250 ml steel grinding set. About ½ of available space was filled with the sample material.

With this type of mill, the sample would be ground with more impact forces and less of friction. After only one minute of grinding, the sample was ground to a fine powder as well. Few traces are mainly located between the puck and ring. A stronger brush was required to remove most of those particles.

We packed the sample best possible and sieve out a portion of 20 grams to determine the level of particles < 250 µm. With this trial, the best result was achieved. We determined 96,3 % within the desired range.

Also here, we cleaned the grinding set by short time comminution of sand and water.