In mill, size reduction of mineral particle assemblies is accomplished stepwise in commercial devices. The complicating features in the breakage of large assemblies of particles are the different particle sizes and strengths and the sometimes random way in which stress is applied in the various steps. In mining operations, blasting with explosives is usually the first step in the comminution sequence. During blasting, high-velocity gas pressure pulses created by the explosives load blocks of ores to provide initial fragmentation. The next step in mineral processing operations is typically crushing, which is accomplished by slow compression of large particles (greater than 1 cm) against rigid surfaces. The mechanism of breakage in the crushing step is contrasted with subsequent size reduction by tumbling or stirred mills in which breakage of smaller particles (less than 1 cm) is accomplished by a combination of impact, chipping, and abrasion events caused by energy transferred from grinding media, such as balls, rods, or large particles. The characteristics of the product required in a given application determines what device or series of devices is required.

Crushing is performed in one or more stages with small reduction ratios, (d80 F /d80 P is between 3 and 6 per stage). Practically speaking, the reduction ratio represents the ratio of feed size opening (the gape) to the discharge size opening (the set). The first stage, primary crushing, usually produces a product that contains particles finer than 10 cm with an attendant energy consumption of less than 0.5 kWh/t. The energy efficiency is on the order of 80%. Secondary crushing can generally achieve size reduction to less than 1 cm with an energy consumption of less than 1.0 kWh/t. Here the efficiency is closer to 50%. The next stage of size reduction for most mineral processing operations is accomplished by wet grinding in rotating cylindrical vessels termed “tumbling mills.” In these mills, particle breakage occurs by compression, chipping, and abrasion caused by the tumbling action of the grinding media. Preliminary grinding can be done with a rod mill, in which case the grinding media consists of an assortment of rods, a ball mill using balls, an autogenous mill that uses no grinding media, or a semiautogenous mill that uses a light load of balls. The product from primary grinding can be as fine as 300 μm. The energy required can be between 5.0 and 25.0 kWh/t depending on the ore and the product size. The energy efficiency of these devices ranges from 15% down to 3%. Sometimes a special crusher called a high-compression grinding roll is used at this stage; it has an efficiency of up to 30%.

Final stages of grinding are normally accomplished in tumbling ball mills or stirred mills. Here it is possible to reach product sizes of a few microns but at costs of as much as 50 kWh/t. In this case, the energy efficiency may drop to as low as 1% when based on the energy for single-particle slowcompression loading..

In this section descriptions of different types of comminution devices are given together with specific design criteria and operating characteristics. For the past quarter century, the size of comminution equipment and associated drives required for commercial-scale comminution tasks has been selected on the basis of the specific energy (energy per unit mass of product) necessary to reduce a feed material to the desired product size. The choice of specific energy as a scale-up criterion is based

on two important premises: (1) equipment of different sizes delivering the same specific energy will yield identical products when fed the same feed material, and (2) existing equipment size/power draft relationships are accurate enough to allow an equipment size that will deliver the necessary energy at the design throughput to be selected.

Whenever possible, the specific energy requirement for a given feed to product transformation is determined in such a way as to minimize the design risk. In other words, the value is determined from an existing full-scale operation or from a pilot-plant circuit that is operated in a fashion similar to that anticipated for the commercial installation. When commercial- or pilot-scale data are not available, design engineers often use the Bond energy size reduction equation, or its equivalent, to estimate specific energy requirements. Recently other techniques involving population balance model parameters have been shown to reduce design risk.