Executive Summary
This project aims in developing a system engineering approach for understanding,
optimizing and scaling industrial dry grinding processes, with a special focus on the
manipulation of the material properties and, thus, the grinding and classification efficiency
by grinding aids. Grinding aids are defined here as liquid or dry substances that are added
to the process in order to increase the product throughput, decrease the specific energy
consumption and/or to reach a certain product fineness. During milling operations, grinding
aids impact powder material mainly in:
Product fineness after grinding;
Tendency of fine particle agglomeration;
Powder flowability;
Total mass of product inside the mill and residence time
Amount of material coated on equipment surfaces.
In this project, dry grinding of the materials alpha alumina and calcium carbonate is
studied. Three substance classes were adopted as grinding aids: An Alcohol, a Carboxylic
acid and a Glycol. For the experiments, a 4 liter batch ball mill and a 47 liter, batch-wise
or continuously operated ball mill as well as an reflector-wheel air classifier are selected.
After defining materials and establishing experimental methods, first batch grinding tests
and powder flowability measurements of the product were conducted. In the case of
alumina, the grinding aids show different efficiency in reducing energy consumption:
Alcohol and glycol were quite effective, while the use of carboxylic acid resulted in results
similar to grinding without additive. For the calcium carbonate the additives presented
similar effects compared to each other, but improving efficiency in comparison to no
additive.
In terms of flowability measurements, the carboxylic acid promoted notable increase in
flowability for both materials. The glycol presented negligible effect on powder flowability,
but with good results for milling efficiency for both materials. It is clear that improved flow
behavior, although beneficial for continuous processing, does not guarantee an effective
grinding. The results also confirm the necessity of describing the phenomena of powder
stressing and transport independently.
Moreover, the continuous grinding process modelling approach is being structured as a
dynamical population balance model predicting product size distribution and energy
consumption in relation to:
Product breakage characteristics;
Product transport behavior;
Process parameters;
Mill geometry and design;
Type and dosage of grinding aid.