Minerals Engineering: Chris Aldrich

Minerals Engineering 49 (2013) 77–91
Contents lists available at SciVerse ScienceDirect
Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng
Consumption of steel grinding media in mills – A review

Chris Aldrich ⇑
Department of Metallurgical and Minerals Engineering, Western Australian School of Mines, Curtin University, Perth, WA, Australia
Department of Process Engineering, University of Stellenbosch, Private Bag X1, Matieland 7602, Stellenbosch, South Africa
a r t i c l e i n f o a b s t r a c t
Article history: In this study, the current understanding of the factors affecting the consumption of steel media in com-
Received 29 March 2013 minution systems in mineral processing are reviewed, together with models predicting wear losses in
Accepted 18 April 2013 grinding media. Media wear arises as a consequence of complex interaction between a range of variables
Available online 5 June 2013
related to processing conditions, the characteristics of the media, as well as the ores or slurries, and is not
well understood as yet, despite extensive study over the last 50 years and more. The three basic wear
Keywords: mechanisms, impact, abrasion and corrosion, can simultaneously influence mass loss in grinding media.
Comminution
Present studies are difficult to compare directly, owing to imprecise information with regard to the com-
Grinding media
Wear
position of the media or grinding conditions. As a result, most current models do not account for varying
Models conditions inside the mill and their use is restricted to conditions similar to those associated with their
calibration. This may not always be possible and alternative modelling methodologies are discussed and
demonstrated by means of a case study on simulated data.
Ó 2013 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
2. Properties of grinding media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.1. Microstructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.2. Effect of carbides in the metal matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.3. Media shape. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3. Grinding wear mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.1. Abrasive wear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.2. Impact wear. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.3. Corrosive wear. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4. Characterizing the grinding environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.1. Pulp potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.2. Dissolved oxygen concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.3. Oxyhydroxide species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.4. Slurry viscosity and surface tension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.5. Mill feed rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.6. Particle size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.7. Rotational speed of the mill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.8. Solids and crop load of the mill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.9. pH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5. Measurement methodology related to grinding media consumption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.1. Dropped ball test (DBT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.2. Marked ball wear test (MBWT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.3. Other laboratory tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.4. Tests in industrial mills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
⇑ Address: Department of Metallurgical and Minerals Engineering, Western

Australian School of Mines, Curtin University, Perth, WA, Australia. Tel.: +61
892664349; fax: +61 893584912.
E-mail address: [email protected]
0892-6875/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.mineng.2013.04.023
78 C. Aldrich / Minerals Engineering 49 (2013) 77–91
Nomenclature
A surface area of grinding media in general, m2 mimp,j grinding media mass loss owing to impact mechanism j,
Ab surface area of ball or sphere, m2 kg
Ai abrasion index, – Dm change in the mass of the grinding media, kg
Alball surface area of grinding ball in laboratory mill, m2 N rotational speed of mill, s1
Arball surface area of grinding ball in industrial mill, m2 Nc critical rotational speed of mill, s1
C cost of grinding media per unit mass, $/kg n wear rate exponent, –
CL crop load, % R volumetric wear rate of grinding media, m3 s1
Df final diameter of grinding ball, m T mass of ore milled, kg
Di initial diameter of grinding ball, m t time, s
D0 initial diameter of mill, m v velocity, m s1
Eabr,i energy dissipated owing to abrasion phenomenon i, J W mass loss of grinding media per unit surface area,
Eimp,j energy dissipated owing to impact phenomenon j, J kg m2
k0 wear speed or wear constant, m3n s1 q density of grinding media in general, kg m3
E
kd energy specific wear rate constant, m J1 kg1 qb density of steel ball, kg m3
km mass wear rate constant, kg s1 XE grinding media consumption based on energy usage,
kd linear wear rate constant, m s1 kg J1
m mass of grinding media, kg XM grinding media consumption based on amount of ore
mabr,i grinding media mass loss owing to abrasion mechanism ground, kg kg1
i, kg Xt grinding media consumption based on operating time,
mcorr grinding media mass loss owing to corrosion, kg kg s1
5.4.1. Wear rate criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

5.4.2. Scale-up from laboratory data to industrial environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6. Grinding media wear models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.1. Linear wear theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.2. Nonlinear (general) wear theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.3. Mechanistic models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.4. Empirical models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
6.5. Numerical simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
7. Case study with simulated data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
8. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
1. Introduction The cost associated with grinding media is chiefly determined

by two factors, viz. the price and wear performance of the grinding
Grinding circuit operators have long been aware of the signifi- media. Different operating conditions can be compared with the
cant impact of grinding media consumption on the cost of grinding. effective grinding cost or the cost-effectiveness of the application
Comminution accounts for an estimated 30–50% of typical mining (Sepúlveda, 2004). This is a challenging task, since different operat-
operating costs, and of these, liner wear and media consumption ing conditions in comminution circuits arising from changes in ore
account for roughly 50% of the cost. According to Moema et al. types, operational procedures and the properties and size distribu-
(2009), in some instances, media wear can constitute up to 40– tions of the grinding media themselves all need to be accounted for
45% of the total cost of comminution. An estimated consumption when the cost of grinding media is calculated (Chenje et al., 2004;
of steel grinding media of around 600,000 tons p.a. in the 1980s al- Lameck et al., 2006; Jayasundara et al., 2011).
ready gives an indication of the scale of the problem (Malghan, The consumption of grinding media has been studied exten-
1982). Likewise, in the cement industry, as mills are supplied for sively in the mineral process industries, where steel balls and rods
ever increasing capacities, the ball size distribution and wear exact are mostly used to reduce rock fragments and ore particles to the
a tremendous effect on the profitability of producing finished ce- fine sizes required for mineral liberation and further downstream
ment. Improper size distribution or filling level of the ball charge processing. Apart from a better understanding of the phenomena
can reduce the efficiency of grinding by 5–20%, amounting to losses involved in the wear of grinding media, many of these studies were
of millions per annum for a mill with a capacity of 150 tons/h also aimed at the development of models capable of predicting
(Longhurst, 2010). media consumption based on an understanding of the mechanisms
Apart from these cost factors, one of the major unsolved prob- involved in the process. In this paper, these studies are reviewed,
lems in the optimal design of ball mills concerns the equilibrium starting with an overview of the properties of grinding media in
media size distribution in the mill, which is determined by the rate Section 2, followed in Section 3 by consideration of the wear mech-
at which make-up media is added to the mill, as well as the rate at anisms on media consumption. This is followed by characterization
which these grinding media are consumed (Austin and Klimpel, of the grinding environment in Section 4, and measurement of
1985). Reliable prediction of grinding media consumption can grinding media consumption in Section 5. In Section 6, grinding
therefore play an important role in the management and control media wear models are reviewed and in Section 7, a simulated case
of these costs, and the overall cost of mining operations. study is considered to illustrate the potential of alternative ap-
C. Aldrich / Minerals Engineering 49 (2013) 77–91 79
Fig. 1. Materials used in wear resistant applications in comminution.
proaches to modelling wear losses in grinding media. The conclu- 2009). As can be expected, the cost of the media, which is ulti-
sions of the study are presented in Section 8. mately consumed in the comminution process, also plays a vital
role, and in this respect selection again depends on the comminu-
2. Properties of grinding media tion environment, as superior mechanical properties and corrosion
resistance are usually associated with higher cost.
A wide range of materials is used to resist wear in comminution A lesser factor that does not seem to be widely considered in the
processes (Durman, 1988; Moema et al., 2009). The abrasiveness of selection of grinding media at present, is the effect of iron accumu-
the material being processed is of prime importance in determin- lating in the mineral slurries or ores, as a result of the consumption
ing the absolute wear rate of the grinding media, but conflicting of grinding media. This should be as low as possible, since iron can
characteristics of high hardness for maximum wear resistance have an adverse effect on downstream processing (flotation), as
and adequate ductility, to avoid catastrophic brittle failure in appli- suggested by a number of studies on non-sulphide (Kinal et al.,
cation, always has to be balanced cost effectively. A range of mate- 2009) awasakind sulphide ores (Martin et al., 1991; Thornton,
rials has been developed for this purpose, which include abrasion 1973; Pavlica and Iwasaki, 1982; Yuan et al., 1996).
resistant steels, non-metallics and alloyed white cast irons, as indi- For example, Chenje et al. (2003a,b, 2004) have conducted com-
cated in Fig. 1. parative studies with different types of balls, consisting among
Of these materials, the manganese steels containing additional other of eutectoid steel, low alloy steel, medium chromium cast
alloying elements, such as Cr, Ni and Mo, are considered to be of iron, cast semi-steel and white cast iron. They have used the cost
the highest quality. The balls are typically hardened to 60–65 effectiveness (E) as a criterion for grinding media selection, i.e.
Rockwell C. The low alloy, low carbon steels are the least expensive
dm
and recommended for rough grinding only, where metallic con- E ¼ C ð1Þ
dT
tamination is not a problem. The austenitic stainless steels are typ-
ically only used in acid media requiring non-magnetic balls, owing where m is the mass of balls in the mill, T is the mass of ore milled
to their high cost. and C is the cost of the grinding media per unit mass. Accounting for
The Ni–Cr cast irons or nickel-hards are white cast irons alloyed the adverse effect that iron could have on downstream processing
with Ni and Cr. Two groups are used in grinding media, namely of the ores, the criterion proposed by Chenje et al. (2003a, 2004)
medium alloyed nickel hards (Ni-hard 1 and 2) and high alloyed could be extended to Eq. (2) to constrain the amount of iron re-
nickel hards (Ni-hard 4). The last group, the high chromium cast ir- leased into the ore system over a given period to some upper limit
ons represent a wide range of characteristics, owing to their range (UL).
of chemical compositions and heat treatability. Chromium carbides
dm dm
are harder than iron carbides and therefore more wear resistant, E ¼ C ; subject to < UL ð2Þ
dT dT
and also play a major role in wear resistance in corrosive environ-
ments. Some of these characteristics in grinding media are dis- Chen et al. (2006) have investigated the consumption of high Cr
cussed in more detail below. (26–30%) alloy balls in a phosphate mill. Mass loss of the balls in-
creased linearly with grinding time or 0.00036 g/h or 247 MPY,1
2.1. Microstructure with all other variables kept constant.
Gundewar et al. (1990) have found that high chromium cast
Given the mechanisms of material loss in grinding media (abra- iron had a significantly higher wear resistance than forged EN 31
sive, impact and corrosive wear), it is clear that wear resistant steel, which in turn exhibited a higher wear resistance than cast
media should generally be corrosion resistant and have superior hypersteel during the wet grinding of Kudremukh iron ore in India.
mechanical properties. The suitability of specific properties, such The high resistance of the chrome balls could be attributed to the
as hardness or toughness, depend on the milling environment. resistance of the balls to corrosion (passivation), especially in the
For example, steels with predominantly pearlitic structures pos- presence of oxygen, as discussed before.
sess excellent impact toughness, but inferior hardness. This may
make them more suitable to milling conditions where high impact 1
MPY calculated from MPY = 534Dm/(qbAbt), with density qb (g/cm3), ball surface
is required, such as when milling hard gold ores (Moema et al., area Ab (inch2), grinding time t (h) and mass loss Dm (mg).
Moroz (1984) has used marked ball wear tests in wet ores to vere conditions, as imposed by quartz in the ball mill test, lead to
examine the effect of the microstructure of 0.90% C forged steel increasing wear rates with an increase in the amount of carbides in
balls on their wear resistance. These balls were subjected to differ- the matrix. This can be explained by rapid removal of the metallic
ent heat treatment procedures (quenching, quenching and temper- matrix followed by microcracking of the exposed carbides. In less
ing, as well as normalization), together with quenched and severe conditions, such as when grinding hematite or phosphate
tempered AISI 4140 steel balls. Wear resistance generally in- rock, the carbides protect the metallic matrix from microcuts or -
creased with the surface hardness of the balls, but carbon content abrasions, and the wear rates decrease as the amount of carbides
was also found to be a key factor. increases, up to the eutectic composition.
By grinding quartzite, Jang et al. (1988) and Chandrasekaran Other investigators have observed similar behaviour. For exam-
et al. (1991) have concluded that the hardness of the worn surfaces ple, Gates et al. (2008) have used ball mill abrasion tests (BMAT) to
of grinding media cannot be used to predict their wear behaviour. predict the relative service lives of wear-resistant alloys for grind-
Microstructures like pearlite, spheroidite, bainite and martensite ing media in mineral grinding environments. The results showed
appeared to wear more when their hardness decreased. In contrast, that very hard (above 630 HV) martensitic steels and white cast ir-
microstructures containing martensite, retained austenite and ons only offer large performance benefits when grinding relatively
undissolved carbides showed a minimum wear at a certain level soft or weak abrasives (Mohs hardness less than about 6) and that
of austenite. In total, the differences in microstructures could lead this may alter the cost-benefit balance in favour of simple low-cost
to a change of up to 28% in mass losses in the grinding media. steels when grinding hard strong minerals. However, even modest
proportions of softer minerals in real ores could favour the use of
more sophisticated hard alloys.
2.2. Effect of carbides in the metal matrix
2.3. Media shape

Albertin and Sinatora (2001) have considered the effect of car-
bide volume fractions from 13% to 41% and matrix microstructure
Relatively little work has been done to determine the effect of
on the wear of 50 mm diameter cast iron balls tested in a labora-
grinding media shape on comminution. Shi (2004) has considered
tory ball mill during wet grinding of hematite, phosphate rock
the effect of tapered cylindrical media (cylpebs) on grinding and
and quartz sand. Martensitic, pearlitic and austenitic matrices
observed that they produced marginally less oversize than steel
were evaluated. Quartz sand caused the highest wear rates, rang-
balls with the same mass and size distribution.
ing from 6.5 to 8.6 lm/h for the martensitic balls, while the wear
Sinnott et al. (2011) have considered the importance of media
rates observed for the phosphate rock ranged from 1.4 to 2.9 lm/
shape on grinding performance in stirred mills. In these mills,
h. An increase in carbide volume fractions resulted in lower wear
the grinding media and charge in a chamber are mobilized by a
rates for the softer abrasives. The eutectic alloy performed best
rotating internal agitator. It is generally accepted that highly
against the hematite and phosphates, owing to the virtually com-
non-spherical debris from balls which break or spall due to manu-
plete protection of the matrix by carbides in the finely divided eu-
facturing defects reduce ball mill grinding performance (Sinnott
tectic microstructures.
et al., 2011).
In contrast, the quartz abrasive rapidly wore out the matrix,
More recently, Qian et al. (2013) have investigated the effect of
continuously exposing and breaking carbide branches in the pro-
grinding media shapes on the grinding kinetics of cement clinkers
cess. The martensitic steels performed best against the quartz
in balls mills. The specific breakage rates of the clinkers were high-
abrasive. The wear rate of 30% carbide cast irons in contact with
er when the cylinders were used, compared to balls.
phosphate particulates increased from 1.46 to 2.84 and to
6.39 lm/h as the matrix changed, respectively, from martensite
to austenite to pearlite. Wear profiles of non-martensitic balls 3. Grinding wear mechanisms
showed deep subsurface carbide cracking, owing to matrix
deformation. Mass losses in grinding media can be attributed to three basic
The effect of the amount of carbides on the wear resistance of mechanisms, viz. abrasion, impact and corrosion. These mecha-
high chromium cast irons depends on the severity of the tests. Se- nisms can be simultaneously active in a given grinding environ-
Fig. 2. Impact, abrasion and corrosion wear mechanisms of grinding media (partly after Radziszewski (2002)).
Fig. 3. Corrosion models for grinding balls: (a) differential abrasion cell and (b) ball-mineral galvanic cell (after Iwasaki et al. (1988)).
Table 1
Studies of mineral-grinding media in aqueous sulphide ore slurries.
Grinding media Mineral References

Mild steel, Cr-steel (15–30%) Arsenopyrite Huang and Grano (2006) and Huang et al. (2006)
Mild steel Pyrite Huang and Grano (2005) and Peng and Grano (2010)
C-steel, Ni-hard cast iron, 22% and 29% Cr white cast iron Sphalerite Isaacson (1989), Yelloji Rao and Natarajan (1989b, 1990) and Vathsala and Natarajan
(1989)
HCLAa, C-steel, Ni-hard cast iron, 22% and 29% Cr white Chalcopyrite Isaacson (1989), Ahn and Gebhardt (1991), Yelloji Rao and Natarajan (1988, 1989a) and
cast iron Natarajan (1996)
Mild steel, high Cr, C-steel, Ni-hard cast iron, 22% and 29% Galena Isaacson (1989), Yelloji Rao and Natarajan (1990) and Peng et al. (2002)
Cr white cast iron
Mild steel, HCLAa steel, austenitic stainless steels Pyrrhotite Pavlica and Iwasaki (1982), Natarajan et al. (1984), Natarajan and Iwasaki (1984) and
Gangophadhyay and Moore (1985)
a
High carbon (1.7%), low alloy.
ment, leading to complex interactions, some which are discussed propagate through the balls radially from the points of contact.
in more detail below. When the compression waves reach areas of acoustic impedance
mismatch in the interior of the impacted body, tension waves re-
flect back and create spalling at points where these waves exceed
3.1. Abrasive wear
the tensile strength of the material. Mass loss from grinding media
can also arise from other mechanical effects, as observed by several
In mills operating at low cascading speeds, abrasive wear is
authors (Moore et al., 1988; Rao et al., 1991), for example, as indi-
considered to be the dominant wear mechanism (Hukki, 1954).
cated in Fig. 2.
In highly abrasive ores, approximately 1–2 kg of grinding media
Sepúlveda (2004) has done calculations on ball breakage based
can be consumed per ton of ore milled (Moema et al., 2009). In
on impact, showing that the speed (v) in metres per second at
wet milling, this assumes proper pulp coverage of the ball surfaces,
which a ball could be moving, can be estimated by
i.e. not too thick to cushion the impact between balls and not too
dilute to result in insufficient coverage of ball surfaces and exces-
sive ball wear. Iwasaki et al. (1988) have concluded that abrasive v ¼ 0:3894pNc D0:5
mill ð3Þ
wear is strongly dependent on slurry rheology, which is in turn
governed by solids loading and viscosity modifiers, if present. In where Nc (rad/s) is the critical mill speed, and Dmill the mill diameter
addition to the effects associated with coating of the grinding med- (m). In practice, the velocities of grinding media in tumbling mills
ia with slurry, the slurry viscosity also affects the movement of the are in the order of 10 m/s (Gates et al., 2007) and generally impact
grinding media in the mill and hence abrasion of the grinding med- breakage of media is more pronounced in larger mills.
ia (Klimpel, 1982, 1983).
Fig. 2 gives a summary of the mechanisms that can affect mass
3.3. Corrosive wear
loss in grinding media. 2-body abrasive wear assumes grit or hard
particles to remove material from opposite surfaces, while 3-body
Corrosive wear of steel grinding media is strongly associated
wear occurs when the particles are not constrained, and are free to
with wet milling environments and has been studied by a number
roll and slide down a surface. Likewise, the contact environment
of investigators by use of electrochemical measurements. Two
determines whether the wear is classified as open or closed. An
models have been postulated, viz. one based on differential abrad-
open contact environment occurs when the surfaces are suffi-
ing cells and one based on galvanic cells, as indicated in Fig. 3. In
ciently displaced to be independent of one another.
the differential abrading cell, the unabraded surfaces of the ores
act as cathodes, where oxygen is reduced, while the freshly
3.2. Impact wear abraded surfaces act as anodes, where iron is oxidized.
In the galvanic cell system, the mineral particles are cathodic
Abrasive wear is generally less in harder grinding media, where and the steel grinding media anodic, again leading to accelerated
spalling owing to impact loading can be more pronounced instead. wear through oxidation of the iron in the media. A complicating
As a phenomenon, spalling has been studied intensively in other factor in accounting for the effects in these galvanic cells is that
disciplines, but does not seem to have been covered in much detail the effects of galvanic interactions between sulphide minerals
in the context of grinding media wear. Essentially, when two balls and grinding media need to be considered in conjunction with
collide, the sudden release of energy causes compression waves to the galvanic interactions that can occur between sulphide miner-
Table 2
Studies of mineral-grinding media in aqueous non-sulphide ore slurries.
Grinding media Mineral References

a
Mild steel, HCLA steel, austenitic stainless steels Magnetite Natarajan and Iwasaki (1984)
High Cr cast Fe, EN31, cast hypersteel Hematite Gundewar et al. (1990)
HCLAa cast and high Cr cast steel, 1018 C steel Phosphates Deshpande and Natarajan (1999), Tao and Parekh (2004)
and Tao et al. (2005)
High Cr, forged steel PGMs (low S) Miettunen et al. (in press)
Mild steel; HCLA steel; Cast iron balls (0%, 8.5%, 11%, 16%, 21%, 26% and 29% Cr) Quartz Pitt et al. (1988) and Rajagopal and Iwasaki (1992b)
Low alloy, eutectoid, med Cr cast iron, unalloyed cast iron, cast semi-steel Granite Chenje et al. (2004)
AISI 1020, HCLA, SS 304 Taconite Iwasaki et al. (1985)
a
High carbon (1.7%), low alloy.
als, as discussed by Rao and Finch (1988), Cheng and Iwasaki data from the literature, corrosion causes less than 10% of the wear
(1992), Cheng et al. (1993, 1999), and Li and Iwasaki (1992). of grinding media in commercial mills.
Corrosive wear can be studied by comparison of grinding media This is in contrast with the observations of Lui and Hoey (1973),
wear losses in wet grinding in corrosive environments with grind- Hoey et al. (1977) and Tao and Parekh (2004), for example, who
ing in similar environments, where corrosion is suppressed. Sup- estimate that approximately 50% or more of grinding media con-
pression of corrosion can be accomplished based on dry grinding, sumption can be attributed to corrosion. They have found the solu-
grinding in organic liquids and wet grinding in nitrogen atmo- tion pH to have the most significant effect on grinding media
spheres. Of these, the latter is considered to give the best estimate consumption. For 1018 carbon steels, minimum wear was associ-
of abrasive wear. ated with a grinding environment characterized by a solution pH
On this basis, the performance of grinding media with different of 7.36, rotation speed at 70.31 rpm, solids percentage at 75.50,
compositions has been studied widely in different corrosive envi- and crop load (%total volume of grinding media, ore, and water rel-
ronments. Corrosive wear becomes particularly significant in the ative to volume of the mill) at 71.94%. For high chromium media,
presence of ore with a high sulphide content in an oxygen-rich minimum wear was observed at a solution pH of 8.69, rotation
environment, owing to galvanic coupling between the grinding speed at 61.13 rpm, solid percentage at 64.86, and crop load at
media and the minerals. Broadly speaking, corrosion in mineral 57.63%.
slurries can be differentiated based on the presence or absence of If anything, this underscores the complexity of corrosion phe-
sulphide ores, as indicated in Tables 1 and 2, which refer to sul- nomena in milling environments, as further illustrated by a US Bu-
phide and non-sulphide slurries respectively. reau of Mines study. In this study, Isaacson (1989) has determined
In the sulphide slurries, grinding media with a high chromium the effect of common sulphide minerals on various types of ferrous
content tended to suffer significantly less wear than their counter- alloy grinding media. The minerals included chalcopyrite, galena
parts containing less chromium. This can be ascribed to the well- and sphalerite. In the presence of oxygen, chalcopyrite was found
known effect of passivation, where initial rapid corrosion of the to increase the corrosion rate through galvanic coupling, galena
steel media leads to the formation of a protective chromium oxide was found to decrease the corrosion rate through an oxygen scav-
film tightly adhering to the steel surfaces. enger mechanism, and the effect of sphalerite was dependent upon
For example, Vathsala (sic) and Natarajan (1989) have studied the type of grinding media.
the galvanic interaction between steel balls and sphalerite that re- Many of the studies outlined above are difficult to compare sys-
sulted in the anodic corrosion of the cast steel balls that became tematically, as the materials they refer to, although roughly simi-
more severe in the presence of oxygen. Similar studies were con- lar, are not the same and in many cases not described in
ducted for pyrite, pyrrhotite and galena by Pavlica and Iwasaki sufficient detail for comparative purposes. The same goes for the
(1982), Adam et al. (1984), and Learmont and Iwasaki (1984). Even conditions in which these experiments were conducted. This is
though oxygen-rich sulphide environments can be seen as aggres- not helpful as far as continuous process improvement and a better
sive, general statements with regard to the effects of corrosion in understanding of wear mechanisms are concerned, as was also ob-
different systems should be treated with care, as indicated by the served by Albertin and Sinatora (2001).
contradictory points of view with regard to the importance of cor-
rosion in the consumption of grinding media in the literature
4. Characterizing the grinding environment
(Rajagopal and Iwasaki, 1992b). Some researchers suggest that cor-
rosion plays a major role in the consumption of grinding media
Changing conditions inside the mill that can have an effect on
during wet grinding, with the contribution of corrosion to the total
grinding media consumption include pulp potential, dissolved oxy-
consumption of grinding media estimated to range from 10% to
gen concentrations, pH, particle size and other, as discussed in
90%.
more detail below.
In contrast, other researchers suggest that corrosion is respon-
sible for only a small part of the total wear. For example, Tolley
et al. (1984) have investigated corrosion rates of alloys commonly 4.1. Pulp potential
used in media in synthetic mill water using an ultrasonic grinding
device especially designed for this work. The grinding media alloys During grinding with iron-based media or in iron mills, the pulp
consisted of forged and cast steel (carbon content from 0.61% to potential can change, the extent of which is naturally related to the
0.79%, as well as so-called nickel hard and high chromium steels type of media used. For example, Leppinen et al. (1998) found, that
(discussed in more detail in Section 3). Electrochemical measure- depending on ore type, the potential difference (after milling) for
ments showed variations in corrosion rates for these alloys be- complex sulphide ores between grinding in normal steel and stain-
tween 9.8 MPY2 for medium-carbon steels and 0.14 MPY for white less steel mills was about 100–250 mV, while Kelebek et al. (1995)
irons. The authors concluded that based on Bureau data, along with have reported differences of 500–600 mV for pyrrhotite-rich cop-
per-nickel sulphide ores in the initial stages of grinding. In princi-
ple, these changes could affect the corrosion rate of the grinding
2
1 MPY = 0.0254 mm/y (SI units). media, but little has been done to show this.
4.2. Dissolved oxygen concentrations ing of coarse granite. In addition, they have observed that during
wet grinding of raw granite, the mineral constituents present in
Interaction between sulphide minerals and mild steel grinding the ground material may have a significant influence on the wear
media leads to a reduction in the amount of dissolved oxygen pres- behaviour of the media. For granite grinding, feldspar can act as
ent in the pulp during the course of grinding (Martin et al., 1991). a bonding agent, gluing fine quartz particles to the coarse granite
Reduction in oxygen level can also occur by other means and may and to the surfaces of the balls. This can alter the wear behaviour
be strongly influenced by ore type. Dissolved oxygen levels of of the media, resulting in a weaker correlation between relative
pulps following laboratory grinding with mild steel mill/media size of the particles and media consumption rates.
have been reported to be as low as 1 ppm (Kelebek, 1993), owing
to corrosion of the grinding media and in conventional full-scale 4.7. Rotational speed of the mill
iron mill discharge pulps, dissolved oxygen levels of less than
0.1 ppm have been measured (Grano et al., 1994). This changes Effect of speed variation on ball wear is confounded by several
the environment experienced by the grinding media and could contradictory trends and factors. As an example, by use of high
have a significant effect on mass loss attributable to corrosion. speed photography, Kumar et al. (1989) have observed that at
70 rpm in a 2D mill setup used to wet grind quartz, all the balls
4.3. Oxyhydroxide species in the mill interact with each other and with the ore through mu-
tual sliding, rolling and colliding. At 86 rpm approximately 50% of
At regions of higher alkaline pulp pH, the iron can dissolve from the balls went out of circulation, reducing the probable number of
the grinding media during milling as ferrous ions and subsequently interactions in one revolution of the mill, decreasing efficiency, as
oxidize to the ferric form (Bruckard et al., 2011). Ultimately it can well as ball wear. In addition, as the speed increased from 50 to
precipitate at the cathodic sulphide mineral sites as oxy-hydrox- 70 rpm, the ball distribution pattern in the mill and the number
ides species, such as Fe(OH)2, FeOOH, and Fe(OH)3. These iron of revolutions did not change substantially, but the coating thick-
hydroxides, which may be hydrophilic, can coat completely or par- ness of the balls decreased. This increased the prospect of metal-
tially the sulphide mineral surfaces, which can in turn affect the to-metal contact and increased wear of the balls. Increasing the
interaction between the grinding media and the mineral species speed from 70 to 86 rpm reduced the number of interactions per
in the pulp. revolution and led to an increase in the coating thickness of the
balls. The net effect of this is was reduced wear.
4.4. Slurry viscosity and surface tension Rajagopal and Iwasaki (1992b) have observed that an increase
in the rotational speed of a mill resulted in an increase in the high
Although the effect of slurry rheology has already been dis- chromium steel balls (austenitic stainless steel, 26% and 29% Cr
cussed in general terms in Section 3.1, the mechanisms are still cast iron) materials exhibited low corrosion. However, increasing
not well understood. Klimpel (1982, 1983) has drawn attention rotation speed, had a comparatively small effect on the corrosion
to the importance of rheological properties in ball mill grinding rates of low Cr balls (containing less than 21% Cr), which was
on the specific rates of breakages of minerals in the presence and markedly higher than those of the high Cr balls.
absence of chemical additives. Meloy and Crabtree (1967) have
studied the effect of viscosity and surface tension of various liquids 4.8. Solids and crop load of the mill
on grinding efficiency. Abrasion of grinding media is further influ-
enced by the coefficient of friction that depends on the type of me- With regard to the effect of solids loading, Iwasaki (1985) has
tal (media material), the hardness and concentration of minerals in used mild steel and high carbon low alloy (HCLA) steel balls to
slurries and by solution properties such as viscosity and pH. wet grind a taconite ore and observed that ball wear decreased
as the solids loading and hence pulp density increased. This could
4.5. Mill feed rate be attributed to the pulp viscosity controlling the thickness of the
pulp layer on the balls, thereby affecting grinding efficiency and
Mill feed rate is thought to influence consumption of grinding ball wear.
media in that it influences the wear mechanisms acting on the In another study comprising pH, rotational speed of the mill,
media. At low feed rates, impact forces are considered to dominate crop loading and solids concentration in the slurry, Chen et al.
over abrasion (Gates et al., 2007) and Howat and Vermeulen (1988) (2006) have proposed a multilinear regression model as a good
have observed that at low feed rates, when the residence time of fit for their experimental data. Of these variables, pH had the larg-
particles within the mill is long, the consumption of grinding med- est effect on wear rate, and the variables could be ranked as fol-
ia is high and the grind is fine. lows: pH > rotational speed > solids loading > crop load.
4.6. Particle size 4.9. pH
The effect of particle size on the wear rates of grinding media is In sulphide minerals in particular, pH is a critical variable that
not yet fully understood. Chandrasekaran and Kishore (1993) have influences the effect the galvanic interaction between grinding
shown that wear rates during dry sand-quartz grinding increase media and minerals (Peng, 2002), but this can also be the case in
relatively rapidly with mean particle size up to some critical size, other minerals. For example, Chen et al. (2006) have observed a de-
beyond which the wear rate increases at a lower rate. They attrib- crease in wear rate of high Cr balls with increasing solution pH, in a
uted the observed behaviour to accumulation of fine particles on phosphate mill, e.g. a wear rate of approximately 224 MPY at
the ball surface. Pintaude et al. (2001) have considered the effect pH = 3.1, which was approximately twice the wear rate of
of particle size on the abrasive wear of high-chromium white cast 115 MPY at pH = 10.0. Corrosion was observed to increase at low
iron mill balls. They have concluded that the ratio of steel ball to pH values as a result of the increased availability of H+ for reduc-
particle size is the critical factor. The highest ball wear rates oc- tion by the cathodic half-cell reaction (Davis, 2000).
curred during grinding of fine granite under dry (120 mg/cycle) These observations could be explained by the fact that no pas-
and wet (129 mg/cycle) conditions. The lowest wear rate (approx- sive film was formed in a strong acidic solution and the chromium
imately 50 mg/cycle) was observed during the wet and dry grind- alloy corroded by active dissolution and enhanced access of dis-
solved oxygen to the alloy, which further increased the corrosion point metal alloy. Upon recovery, the plug is melted and the ball
rate. The formation of a passive film in the presence of dissolved identified.
oxygen at a pH > 10 slowed down the rate of corrosion Another approach is to use different configurations of holes
significantly. (number of holes, hole diameters and orientations) in each. This
Likewise, Hebbar (2011) has studied the grinding wear behav- has to be done with care, so as not to change the microstructure
iour of austempered ductile iron as media material in the commi- and mechanical properties of the balls as a result of localized heat-
nution of Kudremukh haematite iron ore in a ball mill. The grinding ing during drilling.
experiments were conducted under different pH conditions, i.e. Regardless of the method used to identify the balls, they are all
pH = 7.0 and pH = 8.5, while keeping other parameters fixed, at a weighed individually before placing them into an operating mill.
mill speed of 74 rpm for 1 hr. The ore samples were crushed to With periodic recovery of the balls, for example when the mill is
the size of 10 and +30 mesh size using the laboratory jaw crusher. down for scheduled maintenance, they can be retrieved, identified
The mill charge consisted of a set of 200 balls including 25 marked and weighed to determine the rate of mass loss.
ones. After each grinding experiment the 25 marked balls were
washed, dried and weighed. Similar grinding experiments were 5.3. Other laboratory tests
carried using 200 forged EN 31 steel balls as media material. The
volumetric wear rate (R) of the grinding media per revolution of Although the MBWT and DBT have found wide acceptance as
the mill was calculated using following equation: standard tests of the performance of grinding media, other test
apparatus have also been proposed, of which one such is discussed
Dm here. Fiset et al. (1998) have proposed a three-body laboratory im-
R¼ ð4Þ
Nq pact-abrasion pin test. In this test, samples were subject to impac-
Hebbar (2011) concluded among other that the austempered tion and rotation in a cup containing the abrasive (ore), under
ductile iron balls offered better wear resistance than the forged conditions that closely simulated the forces inside a mill. The re-
EN 31 steel balls, and that the wear loss of the media was lower sults correlated well with those obtained by a marked ball wear
in the slurry with higher pH values. Raghavendra et al. (2010) have test, at a significantly lower cost. Other such tribological systems
reported similar findings with austempered ductile iron balls com- are reviewed by Spero et al. (1991), many of which were developed
pared to EN 31 steel balls. for the coal industry. Pons et al. (2004) have made use of multivar-
iate image analysis to compare the degree of abrasion of grinding
media, but it is not clear how these data should be interpreted.
5. Measurement methodology related to grinding media
consumption 5.4. Tests in industrial mills
In a recent review paper, Sepúlveda (2004) gave an overview of 5.4.1. Wear rate criteria
the test methodology for grinding media consumption, covering In laboratory experiments, media wear is typically determined
the dropped ball test, marked ball wear test and testing in indus- by measuring the mass of the media before and after sequential
trial mills, as discussed in more detail below. grinding experiments (Hebbar, 2011), but wear rates of breakage
constants according to Eqs. (7), (8) are not commonly used in
industry to analyze results. Instead, the following consumption
5.1. Dropped ball test (DBT)
indicators are used (Sepúlveda, 2004; Sepúlveda et al., 2006):
The dropped ball test (DBT) was originally developed by the US
Consumption based on energy consumed, XE (g/kW h).
Bureau of Mines (Blickenderfer and Tylczak, 1983, 1985) and later
Consumption based on operating time, Xt (g/h).
adapted by the international Moly–Cop Grinding Systems organi-
Consumption based on amount of ore ground, XM (g/ton).
zation to assess the resistance of any given sample or lot of balls
to repeated severe ball-to-ball impacts. The DBT facility consists
The last indicator (XM) is used most commonly, but is also the
of a J-shaped tube, 10 m high.
least reliable (Sepúlveda, 2004), since it does not account for the
The curved, bottom part of the tube is filled with a known num-
energy required to grind the ore.
ber of balls (e.g. 24, when testing 129 mm balls). When another
On industrial plants, advantage can be taken of parallel lines of
ball is dropped through the tube the top ball retained below in
mills, as reported by Banisi and Farzaneh (2004) for example,
the tube suffers the direct impact of the falling ball, which is rep-
where they have made use of four identical parallel lines at the Sar-
licated through the whole line of balls retained in the curve at
cheshmeh copper mine to test three different ball charges and one
the bottom of the J-tube.
combined charge simultaneously.
This also results in the removal of the first ball in the line
Plant trials can be based on sequential, concurrent or cross-ref-
through the lower tip of the tube, which is replaced by the last ball
erenced evaluation. In sequential evaluation, historical rates of
dropped. The balls removed from the tube are continuously lifted
consumption in the same mill are compared before and after purg-
with a bucket elevator back to the top of the tube to be dropped
ing.3 In concurrent evaluation, two or more mills are operated in
down once again. The DBT is run until a certain maximum number
parallel over exactly the same time interval, once purging has been
of balls are broken (say, five balls) or a reasonable number of total
completed. During cross-reference evaluation, the difference in con-
cycles have been completed (e.g. 20,000 drops).
sumption rate of mill 1 is subtracted from the difference in con-
sumption rate of mill 2, and this is normalized with respect to the
5.2. Marked ball wear test (MBWT) wear constant of the first mill (discussed in more detail in the next
section), e.g.
The marked ball wear test (MBWT) is a reliable approach to
compare the wear rates of different materials under identical con-
ditions. Different approaches can be used to identify grinding med- 3
Lapse of time required for complete consumption of media A before beginning of
ia. One of these is to drill a hole in each ball to me identified, put an test with media B, or time required for complete consumption of first new ball
identification tag in the hole and seal it up with some low melting charged at beginning of test with media B, whichever takes the longest.

E
kd
E
kd kd
E E
ðkd Þmill2;after
E
kd This means that the kinetics is linear and that kd is not depen-
mill1;before mill1;after mill2;before mill1;before
dent on the ball diameter at any given time, i.e. a ball will lose
ð5Þ 1 mm of its diameter over a given period, regardless of whether
it is large or small. The mill is therefore continuously or at least
5.4.2. Scale-up from laboratory data to industrial environments at regular intervals recharged with monosized media of diameter
Data obtained from laboratory studies are not directly applica- D0.
ble to industrial systems. One obvious reason is that impact ener-
gies in laboratory mills are negligible compared to those in 6.2. Nonlinear (general) wear theory
industrial mills. For example, Dodd et al. (1985) have estimated
that an increase in ball kinetic energy from 1.4 J for a 0.025 m ball When the assumptions of linear wear theory do not hold, the
in a 0.20 m ball mill to 350 J for a 0.127 m ball in an 8.5 m ball mill. kinetics become more complex, but can still be accounted for with
A study by Albright and Dunn (1983) have conducted a study that appropriate models (Sepúlveda, 2004). For spherical media, the
yielded wear rate data for 22 different alloys (including pearlitic, mass wear rate can be generalized as follows:
martensitic and bainitic steels and white cast irons) in contact with
dm
molybdenite ore with gangue comprising quartz, granite and a small ¼ kðqpD2 ÞDD ð9Þ
dt
proportion of MoS2) in two 2.9 m diameter ball mills. The same 22 al-
loys were subjected to a pin-on-plate version of the pin abrasion test where k is again a linear wear rate constant. If D = 0, the wear law
(PAT). Some significant differences were seen between the mill trial reduces to the linear model, i.e. mass loss is proportional to the ex-
results and the PAT results, which indicate some fundamental differ- posed surface area of the sphere and the rate of decrease of the ball
ences between the laboratory test and the industrial service environ- radius (dD/dt) is constant. This is known as Bond’s wear law (Bond,
ment. Similar relatively disappointing service performance of white 1943). If D = 1, the rate of mass loss is proportional to the volume of
cast irons has been reported in many anecdotal accounts of minerals the sphere, which is known as Davis’s wear law (Davis, 1919). In
industry experience (Gates et al., 2007). In fact, this disappointing most tests, Bond’s wear law is observed (linear wear theory).
performance of white irons in the plant compared to predictions of An alternative (simpler) form of the above wear law for steel
laboratory tests became the defining characteristic underlying balls was given by other authors (Lorenzetti et al., 1977; Meulen-
hypotheses of impact-abrasion wear mechanisms. dyke and Purdue, 1989; Moroz and Goller, 1986). In these papers,
Spero et al. (1991) highlighted the need for improved correla- they present the basic wear law as
tion between laboratory test results and production mill wear rates
dm n
based on a more fundamental understanding of the wear processes ¼ kD ð10Þ
involved was identified. They have cited several examples giving dt
reasonable indications that wear rates in full-scale mills can be ob- In Eq. (10), m is the mass of the ball, t is time, D is the diameter
tained empirically from the many results of laboratory test meth- of the grinding ball, k is a proportionality constant and n is an
ods (based on some moderate correlation rates of exponential constant, that is the wear rate of the ball is propor-
R2 = approximately 60%). They have concluded that further work tional to the ball diameter. Different exponential constants ranging
is required to obtain more reliable simulation of the wear condi- from 2 to 3 have been proposed by different authors (Davis, 1919;
tions in production mills, and to set the limits of predictability of Norquest and Miller, 1950; Bond, 1943; Austin and Klimpel, 1985;
three-body laboratory wear tests. In addition, they have identified Vermeulen and Howat, 1986; Azzaroni, 1987), depending on the
scope for further work based on the application of statistical distri- relative importance of surface dependent abrasion and corrosion
butions such as the log-normaI, Gaussian, and Weibull functions to (n = 2) or volume (mass) dependent impact forces (n = 3).
characterize wear rates. The solution of Eq. (10) gives (Meulendyke and Purdue, 1989)
0 D3n
i D3n
f
6. Grinding media wear models k ¼ ð11Þ
tð3 nÞ
6.1. Linear wear theory where k0 (m3n s1) is the wear constant or wear speed, t (s) is the
grinding time, Di and Df (m) are the initial and final grinding ball
The linear wear theory originally developed by Prentice (1943) diameters and n is the wear rate exponent. Vermeulen and Howat
and Norquest and Miller (1950) to characterize the slow sustained (1986) postulated the theory of combined wear, where the wear
consumption of grinding bosies in rotating mills is the most widely rate constant may vary during the course of grinding, depending
accepted approach in current use (Sepúlveda, 2004). According to on the prevalence of abrasion or impact phenomena and changes
this theory, the rate of mass loss of a body being ground in a mill in the grinding environment. For example, as the mill diameter
is directly proportional to its exposed surface area, i.e. and charge diameter increases and the mill volume decreases, the
dm intensity of impact forces will be larger than that of abrasive forces
¼ km Ab ð6Þ
dt (Azzaroni, 1987). Corrosion may also play a part in the apparent
In Eq. (6), m (kg) is the mass of the body at time t (s), while km time variance of the exponent.
(kg s1 m2) is the mass wear rate constant and Ab (m2) is the sur- Yildirim and Austin (1998) have extended Eq. (9) to the con-
face area of the body exposed to wear. sumption of cylinders, the geometry of which is specified by two
If the geometry of the body (assumed to be a sphere) is taken parameters (length and radius). This yielded a radial and axial wear
into account, then model, the details of which is beyond the scope of this document.
dD
¼ 2km =qb ¼ kd ð7Þ
dt 6.3. Mechanistic models
3
In Eq. (7), D (m) is the diameter of the ball, qb (kg m ) is the
density of the ball and kd (m/s) is the linear wear constant. Assum- More recent attempts have been made to construct models
ing kd to be time invariant, integration of Eq. (7) gives based on an understanding of the mechanistic principles of wear
of the grinding media. Radziszewski (2002) have proposed a
D ¼ D0 kd t ð8Þ
semi-empirical total media wear model on the assumption that
the effect of each wear mechanism can be determined indepen- could explain 96% of the variance in the mass loss of the alloy as a
dently to give an additive model of the form function of solution pH, solids loading, crop load and rotational
speed of the mill, whereas a linear model could explain only 80%.
dm X 3
dmi Although the modelling methodology is sound, in this particular
¼ ð12Þ
dt i¼1
dt case it is not clear whether the model would be able to generalize
well, as it consisted of 25 IF-THEN rules fitted to 29 measurements
Eq. (12) has three terms to account for abrasion, impact and cor- obtained from a study by Chen et al. (2006).
rosion, as presented in an expanded form in Eq. (13).
n n
dm X abr
dmabr;i dmcorr Ximp
dmimp;j 6.5. Numerical simulation
¼ ðEabr;i Þ þ Arball þ ðEimp;j Þ ð13Þ
dt i¼1
dt A lball dt j¼1
dt
Ashrafizadeh and Ashrafizadeh (2012) have used numerical
The abrasion and impact terms make provision for different (discrete element) simulation to predict the wear caused by solid
phenomena and are expressed as functions of the abrasion and im- particle impact. More specifically, the discrete element method
pact energies (Eabr;i ; Eimp;j ) associated with the specific phenomena. (DEM) was used to simulate the behaviour of a jet of particles that
However, the parameters of the model have to be determined exit a nozzle and hit a flat plate for various impact angles, particle
experimentally, which may not be feasible in practice. Moreover, velocities and particle concentrations. The main idea was to evalu-
the assumption of additivity of the different mechanisms may ate DEM as a reliable method for predicting wear rate and erosion
not hold, as abrasion and corrosion may work together to enhance mechanism in problems associated with solid particle interactions.
grinding media consumption. Radziszewski (1997) has proposed a Comparing the simulation results with reported experimental data,
mathematical model describing ball mill wear as a function of mill it was confirmed that there is a correlation between shear impact
operating variables. The wear model incorporated the energy dissi- energy and the wear rate. In future this may become a useful ap-
pated in crushing, tumbling and grinding zones of the charge pro- proach to predicting the performance of grinding media.
file with adhesive and abrasive wear descriptions.
Radziszewski (2000) has given and overview of abrasive and
7. Case study with simulated data
corrosive wear mechanisms and has proposed an experimental
procedure to facilitate prediction of the effect of changes in charge
Although the basic phenomena underlying the wear mecha-
media composition. In addition, he has presented experimental re-
nisms of these media, viz. impact, abrasion and corrosion, have
sults for four cases and used these to predict changes in media
been studied extensively, the interaction between grinding media
wear.
and ore particles or mineral slurries is too complex to generally al-
He has used a simple rule-of-thumb procedure based on the
low modelling from first principles that can account explicitly for
observations that (i) in laboratory tests, corrosion can represent
these phenomena.
anywhere from 25% to 75% of metal loss depending on the ore–me-
As a consequence, in practice, predictive wear models for grind-
tal-environmental factors involved (Rajagopal and Iwasaki,
ing media (mostly steel balls) are empirical in nature and depend
1992a,b), while (ii) corrosion represents less than 10% of total me-
on grinding media mass loss data to parameterize. These models
tal loss in typical large diameter balls mills.
generally predict wear from a single variable, typically the diame-
Radziszewski (2000) further assumed that regarding the batch
ter or surface area of the grinding media (usually assumed to be
mill corrosion test procedures: (i) the stainless steel mill has a neg- n
spherical), i.e. dm ¼ kD . In effect the model parameters (k and n)
ligible effect on real corrosion wear rates, (ii) no constant temper- dt
are assumed to be constant, but this is not generally the case, as
ature, (iii) no air circulation in the test mill, (iv) lack of control of
they depend on other variables that may change during the grind-
the grain size of the slurry has a negligible effect on real corrosion
ing process.
wear rates, and (v) similar charge volume/geometry exists be-
Since these models are primarily used in the selection of grind-
tween test mills and real mills. Based on these observations and
ing media, it also implies that models have to be used under the
assumptions, he has used a few simple rules to predict grinding
same conditions for which they had been calibrated. It would
media wear.
clearly be futile to attempt prediction of wear losses in the media
in a wet mill, based on a model calibrated under dry mill condi-
6.4. Empirical models
tions. However, even if the same nominal systems are observed,
the variation in the ore and internal conditions in the mill may still
One of the earliest models was that of Bond (1943), who pro-
be affect the reliability of the models.
posed a specific consumption rate for balls XE (kg/kW h) in wet
In principle, this can be accounted for by including additional
mills of
variables in the model, as was done by Chen et al. (2006) for a lab-
XE ¼ 0:16ðAi 0:015Þ1:3 ð14Þ oratory system. In practice it may be more difficult to measure
variables representative of the environment of grinding media
and for dry balls and very little work has been done in this context. For example,
conditions in the mill should be embedded in other state variables
XE ¼ 0:023ðAi Þ0:5 ð15Þ of the mill, such as the mill load or power consumption. According
where Ai is an abrasion index that depends on the ball material, to the theorem of Takens, these variables could be unfolded in a
which could range from Ai = 0.016 for dolomite to Ai = 0.891 for state or phase space to account for the behaviour of unseen state
alumina. variables, which could potentially be used as predictor variables
More recently, Kor et al. (2010) have developed a fuzzy logic for grinding media consumption.
model4 to predict the wear rate of a high chromium alloy. The model Grinding wear models may also benefit from better instrumen-
tation, including online analyzers of iron in the ore slurry that
4
could potentially give an indication of the corrosive wear of the
Fuzzy systems essentially consist of compact sets of fuzzy IF-THEN rules and a
mechanism to interpret these rules (e.g. IF the Cr content of the steel ball is high, and
media in ore systems. This could also inferential models based
milling is dry, THEN consumption is low, with some means to interpret or quantify the on wear of other components, such as special coupons with mate-
labels high, dry and low). rial properties similar to the media placed in the mill, the state of
Fig. 4. Grinding media wear models.
Fig. 5. Prediction of wear with neural network model.
Fig. 6. Variable importance analysis with neural network model.
which could be easier to monitor online. These ideas are diagram- industry, can be generalized to include additional variables x1,
matically portrayed in Fig. 4. x2, . . . , xm. The functional relationship can be determined explicitly,
In this figure, modelling can be quantitative or qualitative. The e.g. by multiple linear regression, or otherwise, e.g. by means of
quantitative models are expressed as simple deterministic models fuzzy models, neural networks, etc.
wear the mass loss rate of grinding media depend on the dimen- Qualitative modelling is also possible, especially where corro-
sions (ball diameter) of the media only. This assumes constant sion could be an important factor in media losses. For example,
comminution conditions that are represented by the two parame- in this instance, knowledge-based systems (expert systems, case-
ters, k and n, of the model. This linear model that is widely used in based reasoning, etc.) could be used to capture some of the knowl-
edge with regards to the effect of different ores on corrosive wear variables could be determined as well. As indicated in Fig. 6, time
of the grinding media. played the most important role in the consumption of the media,
To illustrate these points, the data of Chen et al. (2006) were while pH was less important, but also significant. The other vari-
used to simulate the specific mass loss of grinding media (W, g/ ables had a negligible impact on the response variable.
m2) as a function of time t, (h), pH, rotational speed of the mill In addition, partial dependence analysis (Auret and Aldrich,
RS, (rpm), mill crop load CL (%) and solids concentration SC (%). 2011, 2012) could be used to assess the influence of each variable
150 data points of the form [t, pH, RS, CL, SC, W] were simulated on the response. This influence could depend on the values of the
and a multilayer perceptron with a single sigmoidal hidden layer other variables in the model, and could potentially be positive or
was trained with the Levenberg–Marquardt algorithm to fit the negative. The effect of time is indicated in Fig. 7. Note that all vari-
data. Fivefold cross-validation was used to ensure that the model ables were scaled to the range [1; 1]. As expected, time had a
could generalize on data not used during training. The model could monotonously increasing effect of mass loss, i.e. more mass was
explain 86.4% of the variance of the data, as indicated in Fig. 5. lost the longer the media were used. Owing to the effect of the
By permuting the variables one at a time in the model, the other variables in this case, the relationship was not linear over
importance of each variable on the response (specific mass loss the entire duration of mill operation.
of the media) could be determined, as explained in more detail The problem could also be treated as a classification problem,
by Auret and Aldrich (2011, 2012). The results of this variable by discretising the response variable into ‘Low’ (W 6 3.9 g/m2),
importance analysis are shown graphically in Fig. 6. By comparing ‘Med’ (3.9 < W 6 8.7 g/m2) and ‘High’ (W > 8.7 g/m2) values. In this
the results with those of a random variable, the significance of the case, for illustrative purposes, a classification tree was fitted to the
Fig. 7. Relationship between wear and time.
Fig. 8. Structure of a classification tree fitted to the wear loss data.

1.5
12
1
0.5 10
Scaled pH
0 8
-0.5 6
-1 4
-1.5 2
-1.5 -1 -0.5 0 0.5 1 1.5 f7
Scaled Time
Fig. 9. Response surface generated by the neural network regression model (left) and the classification tree model (right).
data, recast as 300 samples of the form [t, pH, RS, CL, SC|CLAS- effect of the environment on the consumption of grinding
S = {‘Low’, ‘Medium’, ‘High’}]. The Gini information criterion was media, e.g. pH or electrochemical noise in corrosive slurry sys-
used to recursively split the variable space based on an exhaustive tems, mill load or power consumption, etc.
univariate search, class probabilities estimated from the data and Direct or inferential measurement of the variables representing
equal misclassification costs. This yielded a tree model with 7 ter- the interior of mills also remains to be further investigated.
minal nodes as shown in Fig. 8. When steel media are used, it could possibly include online
The tree could predict the correct class 88.7% of the time. The measurement of the iron in the mill product, or measurement
seven terminal nodes are equivalent to seven IF-THEN rules, viz. of particle size distributions in the feed and mill product by
use of machine vision systems.
IF t 6 1.26 THEN Class = Low Scale-up from laboratory test data to industrial systems
IF t > 1.26 AND t 6 1.73 AND pH 6 6.9 THEN Class = Med remains an issue, as some of the wear mechanisms change fun-
IF t > 1.73 AND t 6 2.88 AND pH 6 6.9 THEN Class = High damentally during scale-up, although studies in large labora-
IF t > 1.26 AND t 6 2.88 AND pH > 6.9 THEN Class = Med tory ball mills are probably a good compromise.
IF t > 2.88 AND pH 6 8 THEN Class = High A number of studies relating the properties of media to their
IF t > 2.88 AND pH > 8 AND RS 6 65 THEN Class = Med wear resistance in different systems have been done, but these
IF t > 2.88 AND pH > 8 AND RS > 65 THEN Class = High have been mostly system specific and more work can be done to
develop models of a more generic nature, even if these are qual-
The response surface (variables t and pH only) of the classifica- itative only. Generally, studies on the relationship between the
tion tree is a coarser approximation of that of the regression model grinding media composition and microstructures are difficult to
derived with the neural network, as shown in Fig. 9. Even coarser compare.
approximations are possible when the predictor variables are also Linear models are well-established and reliable, but have lim-
discretized. Rules such as these could be used in conjunction with ited application, as they require grinding conditions to be sim-
an inference engine as part of an expert system. It would also be ilar to those they had been calibrated on. As was indicated in
possible to fuzzify the rules to form the basis of a fuzzy rule-based this paper, in principle at least, it would be possible to construct
system or alternatively, fuzzy rules could be derived directly from more general models with wider applicability by using a few
the data, as was done by Kor et al. (2010) in this case. Such models additional measurements currently available on many grinding
could be used as expert systems to guide the selection of grinding circuits.
media for different grinding systems, or at least in principle, to
guide the addition of grinding media to the mill.
Acknowledgement
8. Conclusions
Donhad’s support of this study is gratefully acknowledged.
The consumption of grinding media plays an important role in
the economics of grinding and as a consequence also in the overall
processing of a large variety of ores. Although the effects of individ- References
ual variables on the mass loss of grinding media may be well
understood in principle, the large number of variables and their Adam, K., Natarajan, K.A., Iwasaki, I., 1984. Grinding media wear and its effect on
the flotation of sulphide minerals. International Journal of Mineral Processing
interaction, coupled with the difficulty in directly observing these 12, 39–54.
variables, makes quantitative estimates of the wear of grinding Ahn, J.H., Gebhardt, J.E., 1991. Effect of grinding media-chalcopyrite interaction on
media a formidable challenge. Despite steady progress over several the self-induced flotation of chalcopyrite. International Journal of Mineral
Processing 33 (1–4), 243–262.
decades of research, a number of areas could still benefit consider-
Albertin, E., Sinatora, A., 2001. Effect of carbide volume fraction and matrix
ably from further investigation. These include the following: microstructure on the wear of cast iron balls tested in a laboratory ball mill.
Wear 250, 492–501.
Characterization of the grinding environment is difficult, and Albright, D.L., Dunn, D.J., 1983. Wear behaviour of iron and steel castings for the
mining industry. Journal of Metallurgy 35, 23–29.
little progress has been made to date to systematically identify- Ashrafizadeh, H., Ashrafizadeh, F., 2012. A numerical 3D simulation for prediction of
ing potential state variables that could be used to represent the wear caused by solid particle impact. Wear 276–277, 75–84.
Auret, L., Aldrich, C., 2011. Empirical comparison of tree ensemble variance Howat, D.D., Vermeulen, L.A., 1988. Fineness of grind and the consumption and
importance measures. Chemometrics and Intelligent Laboratory Systems 105, wear rates of metallic grinding media in tumbling mills. Powder Technology 55
157–170. (4), 231–240.
Auret, L., Aldrich, C., 2012. Interpretation of nonlinear relationships between Huang, G., Grano, S., 2005. Galvanic interaction of grinding media with pyrite and
process variables by use of random forests. Minerals Engineering 35, 27–42. its effect on flotation. Minerals Engineering 18 (12), 1152–1163.
Austin, L.G., Klimpel, R.R., 1985. Ball wear and ball size distributions in tumbling Huang, G., Grano, S., 2006. Galvanic interaction between grinding media and
ball mills. Powder Technology 41, 279–286. arsenopyrite and its effect on flotation: Part I. Quantifying galvanic
Azzaroni, E., 1987. Considerations on the performance of grinding balls in operating interaction during grinding. International Journal of Mineral Processing 78
conditions and eventual effects in determining the wear law factors. In: 5th (3), 182–197.
Symposium on Grinding, Armco, SAMI, Vina del Mar, Chile. Huang, G., Grano, S., Skinner, W., 2006. Galvanic interaction between grinding
Banisi, S., Farzaneh, M., 2004. Effect of ball size change on the performance of media and arsenopyrite and its effect on flotation: Part II. Effect of grinding on
grinding and flotation circuits. European Journal of Mineral Processing & flotation. International Journal of Mineral Processing 78 (3), 198–213.
Environmental Protection 4 (3), 194. Hukki, R.T., 1954. Correlation between principal parameters affecting mechanical
Blickenderfer, R., Tylczak, J.H., 1983. A large-scale impact spalling test. Wear 84, ball wear. Transactions of AIME 190, 642–644.
361–373. Isaacson, A.E., 1989. Effect of Sulfide Minerals on Ferrous Alloy Grinding Media
Blickenderfer, R., Tylczak, J.H., 1985. Laboratory Tests of Spalling, Breaking, and Corrosion. US Bureau Report of Investigations, RI 9244. 23pp.
Abrasion of Wear-resistant Alloys used in Mining and Mineral Processing. Iwasaki, I., 1985. Effects of percent solids and mill loading on ball wear in laboratory
USBM RI 8968 [Report], 1–17. taconite grinding. Minerals and Metallurgical Processing 2, 185–192.
Bond, F.C., 1943. Wear and size distribution of grinding balls. Transactions AIME Iwasaki, I., Pozzo, R.L., Natarajan, K.A., Adam, K., Orlich, J.N., 1988. Nature of
153, 373–384. corrosive and abrasive wear in ball mill grinding. International Journal of
Bruckard, W.J., Sparrow, G.J., Woodcock, J.T., 2011. A review of the effects of the Mineral Processing 22, 345–360.
grinding environment on the flotation of copper sulphides. International Journal Iwasaki, I., Riemer, S.C., Orlich, J.N., Natarajan, K.A., 1985. Corrosive and abrasive
of Mineral Processing 100, 1–13. wear in ore grinding. Wear 103, 253–267.
Chandrasekaran, T., Kishore, 1993. Effect of microstructure and hardness on the Jang, J.W., Iwasaki, I., Moore, J.J., 1988. Effect of martensite and austenite on
grinding abrasive wear resistance of a ball bearing steel. Wear 161 (1–2), 105– grinding media wear. Wear 122 (3), 285–299.
109. Jayasundara, C.T., Yang, R.Y., Yu, A.B., 2011. Effect of the size of media on grinding
Chandrasekaran, T., Natarajan, K.A., Kishore, 1991. Influence of microstructure on performance in stirred mills. Minerals Engineering. http://dx.doi.org/10.1016/
the wear of grinding media. Wear 147 (2), 267–274. j.mineng.2011.10.012.
Chen, G.L., Tao, D., Parekh, B.K., 2006. A laboratory study of high chromium alloy Kelebek, S., 1993. The effect of oxidation on the flotation behaviour of nickel–copper
wear in phosphate grinding mill. International Journal of Mineral Processing 80 ores. In: Proceedings of the XVIII International Mineral Processing Congress,
(1), 35–42. Sydney, vol. 4. The Australasian Institute of Mining and Metallurgy, Melbourne,
Cheng, X., Iwasaki, I., 1992. Effect of chalcopyrite and pyrrhotite interaction on pp. 999–1005.
flotation separation. Minerals & Metallurgical Processing 9 (2), 73–79. Kelebek, S., Fekete, S.O., Wells, P.F., 1995. Selective depression of pyrrhotite using
Cheng, X., Iwasaki, I., Smith, K.A., 1993. Electrochemistry of chalcopyrite– sulphur doxide-diethylenetriamine reagent combination. In: Proceedings of the
pyrrhotite–mild steel interactions and its relevance to the flotation of XIX IMPC, San Francisco, Flotation, Operating Practices and Fundamentals, vol.
complex sulfide ores. In: Reddy, R.G., Weizenbach, R.N. (Eds.), Extractive 3, pp. 181–187 (Chapter 30).
Metallurgy of Copper, Nickel and Cobalt, Fundamental Aspects, vol. 1. The Kinal, J., Greet, C., Goode, I., 2009. Effect of grinding media on zinc depression in a
Minerals, Metals & Materials Society, Warrendale, pp. 971–991. lead cleaner circuit. Minerals Engineering 22, 759–765.
Cheng, X., Iwasaki, I., Smith, K.A., 1999. Electrochemical study of multielectrode Klimpel, R.R., 1982. Slurry rheology influence on the performance of mineral/coal
systems and their relevance to the differential flotation of complex sulfide ores. grinding circuits. Part 1. Minerals Engineering 34 (12), 1665–1668.
Minerals & Metallurgical Processing 16 (1), 69–71. Klimpel, R.R., 1983. Slurry rheology influence on the performance of mineral/coal
Chenje, T.W., Simbi, D.J., Navara, E., 2003a. Wear performance and cost effectiveness grinding circuits. Minerals Engineering 35 (1), 21–26.
– a criterion for the selection of grinding media for wet milling in mineral Kor, M., Abkhoshk, E., Tao, D., Chen, G.L., Modarres, H., 2010. Modeling and
processing operations. Minerals Engineering 16, 1387–1390. optimization of high chromium alloy wear in phosphate laboratory grinding
Chenje, T.W., Simbi, D.J., Navara, E., 2003b. The role of corrosive wear during mill with fuzzy logic and particle swarm optimization technique. Minerals
laboratory milling. Minerals Engineering 16, 619–624. Engineering 23 (9), 713–719.
Chenje, T.W., Simbi, D.J., Navara, E., 2004. Relationship between microstructure, Kumar, K.U., Gundappa, P., Bai, B.N.P., Natarajan, K.A., Biswas, S.K., 1989.
hardness, impact toughness and wear performance of selected grinding media Laboratory studies on the wear of grinding media. Tribology International
for mineral ore milling operations. Materials and Design 25, 11–18. 22 (3), 219–225.
Davis, E.W., 1919. Fine crushing in ball mills. Transactions AIME 61, 250. Lameck, N.S., Kiangi, K.K., Moys, M.H., 2006. Effects of grinding media shapes on
Davis, J.R., 2000. Corrosion: Understanding the Basics. ASM International, Materials load behaviour and mill power in a dry ball mill. Minerals Engineering 19,
Park, Ohio, p. 218. 1357–1361.
Deshpande, R.J., Natarajan, K.A., 1999. Studies on grinding media wear and its effect Learmont, M.E., Iwasaki, I., 1984. Effect of grinding media on galena flotation.
on flotation of ferrugenous phosphate ore. Minerals Engineering 12 (9), 1119– Minerals and Metallurgical Processing 1, 136–143.
1125. Leppinen, J.O., Hintikka, V.V., Kalapudas, R.P., 1998. Effect of electrochemical control
Dodd, J., Dunn, D.J., Huiatt, J.L., Norman, T.E., 1985. Relative importance of abrasion on selective flotation of copper and zinc from complex ores. Minerals
and corrosion in metai loss in baII milling. Mineral and Metallurigical Processes, Engineering 11 (1), 39–51.
212–216. Li, X., Iwasaki, I., 1992. Effect of cathodic polarization on the floatability of
Durman, R.W., 1988. Progress in abrasion-resistant materials for use in chalcopyrite in the absence of oxygen. Minerals & Metallurgical Processing 9
comminution processes. International Journal of Mineral Processing 22 (1–4), (1), 1–6.
381–399. Longhurst, D.A., 2010. Economics and methodology of ball mill media maintenance.
Fiset, M., Huard, G., Grenier, M., Jacob, C., Comeau, G., 1998. Three-body impact- In: IAS/PCA 52nd Cement Industry Technical Conference, Colorado Springs, CO,
abrasion laboratory testing for grinding ball materials. Wear 217 (2), 271–275. USA, 28 March–1 April 2010, Code80751.
Gangophadhyay, A.K., Moore, J.J., 1985. The role of abrasion and corrosion in Lorenzetti, J.J., Zecca, A.R., Azzaroni, E., 1977. The Armco approach to predicting
grinding media wear. Wear 104, 49–64. grinding ball size and weight distribution in a ball mill. In: Symposium on
Gates, J.D., Dargusch, M.S., Walsh, J.J., Field, S.L., Hermand, M.J.-P., Delaup, B.G., Saad, Management of Copper Resources. Lubin, Poland, 27–28 April.
J.R., 2008. Effect of abrasive mineral on alloy performance in the ball mill Lui, A.W., Hoey, G.R., 1973. Corrosive and erosive wear of metals in mineral slurries.
abrasion test. Wear 265 (5–6), 865–870. Canadian Metallurgical Quarterly 12 (2), 185–190.
Gates, J.D., Gore, G.J., Hermand, M.J.P., Guerineau, M.J.P., Martin, P.B., Saad, J., 2007. Malghan, S.G., 1982. Methods to reduce steel wear in grinding mills. Minerals
The meaning of high stress abrasion and its application in white cast irons. In: Engineering 32, 684–690.
Hutchings, I.M. (Eds.), Wear: 16th International Conference on Wear of Martin, C.J., McIvor, R.E., Finch, J.A., Raoi, S.R., 1991. Review of the effect of grinding
Materials. Montreal, Canada, 6-35, 15–17 April 2007. media on flotation of sulphide minerals. Minerals Engineering 4 (2), 121–132.
Grano, S.R., Wong, P.L.M., Skinner, W., Johnson, N.W., Ralston, J., 1994. The effect of Meloy, T.P., Crabtree, D.D., 1967. Surface tension and viscosity in wet grinding.
autogenous and ball mill grinding on the chemical environment and flotation of Dechema Monographien 57 (1), 405–526.
the copper ore of Mount Isa Mines Limited, Australia. In: Castro, S., Alvarez, J. Meulendyke, M.J., Purdue, J.D., 1989. Wear of grinding media in the mineral
(Eds.), Proceedings of the IV Meeting of the Southern Hemisphere on Mineral processing industry: an overview. Minerals and Metallurgical Processing, 167–
Technology, Concepcion, Flotation – A Volume in Memory of Alexander Sutulov, 172.
vol. II. Andros Ltd., Concepcion, pp. 351–388. Miettunen, H., Kaukonen, R., Corin, K., Ojala, S., Keiski, R.L., 2012. Effect of reducing
Gundewar, C.S., Natarajan, K.A., Nayak, U.B., Satyanarayana, K., 1990. Studies on ball grinding conditions on the flotation behaviour of low-S content PGE ores.
wear in the grinding of Kudremukh hematite–magnetite ore. Minerals Minerals Engineering. doi: http://dx.doi.org/10.1016/j.mineng.2012.03.029.
Engineering 3 (1/2), 207–220. Moema, J.S., Papo, M.J., Slabbert, G.A., Zimba, J., 2009. Grinding media quality
Hebbar, R., 2011. Investigation on grinding wear behaviour of austempered ductile assurance for the comminution of gold ores. In: World Gold Conference, The
iron as media material during comminution of iron ore in ball mills. Southern African Institute of Mining and Metallurgy.
Transactions of the Indian Institute of Metals 64 (3), 265–269. Moore, J.J., Perez, R., Gangopadhyay, A., Eggert, J.F., 1988. Factors affecting wear in
Hoey, G.R., Dingley, W., Freemen, C., 1977. Corrosion behaviour of various steels in tumbling mills: Influence of composition and microstructure. International
ore grinding. CIM Bulletin, 105–109. Journal of Mineral Processing 22 (1–4), 313–343.
Moroz, P.J., 1984. The effects of carbon content, matrix hardness and microstructure Rajagopal, V., Iwasaki, I., 1992b. Corrosion properties of cast iron ball materials in
on the wear of steel grinding balls during wet copper ore grinding. In: Ninth wet grinding. Corrosion 48 (2), 124–131.
District Six Meeting, October 24–27, Kamloops, BC. Rao, S.R., Finch, J.A., 1988. Galvanic interaction studies on sulfide minerals.
Moroz, P.J., Goller, G.A., 1986. Improving the Wear Characteristics of Moly–Cop Canadian Metallurgical Quarterly 27 (4), 253–259.
Forged Steel Grinding Balls. Technical Report. Rao, Q., He, B., Zhou, Q., 1991. A study of the impact fatigue resistance of grinding
Natarajan, K.A., 1996. Laboratory studies on ball wear in the grinding of a balls – I. Matrices and retained austenite. Wear 151, 13–21.
chalcopyrite ore. International Journal of Mineral Processing 46 (3–4), 205–213. Sepúlveda, J.E., 2004. Methodologies for the evaluation of grinding media
Natarajan, K.A., Iwasaki, I., 1984. Electrochemical aspects of grinding media-mineral consumption rates at full plant scale. Minerals Engineering 17, 1269–1279.
interactions in magnetite ore grinding. International Journal of Mineral Sepúlveda, J.E., De Araugo, B., Wardell-Johnson, L., 2006. Comparative evaluation of
Processing 13, 53–71. grinding media consumption rates at full plant scale. In: Proceedings of the SAG
Natarajan, K.A., Riemer, S.C., Iwasaki, I., 1984. Influence of pyrrhotite on the 2006 Conference. Vancouver, BC, Canada, pp. 1–18.
corrosive wear of grinding balls in magnetite ore grinding. International Journal Shi, F., 2004. Comparison of grinding media – Cylpebs versus balls. Minerals
of Mineral Processing 13 (1), 73–81. Engineering 17, 1259–1268.
Norquest, D.E., Miller, J.E., 1950. Relative wear rates of varying diameter grinding Sinnott, M.D., Cleary, P.W., Morrison, R.D., 2011. Is media shape important for
balls in production mills. Transactions AIME 187, 712–714. grinding performance in stirred mills? Minerals Engineering 24, 138–151.
Pavlica, J.J., Iwasaki, I., 1982. Electrochemical and magnetic interactions in Spero, C., Hargreaves, D.J., Kirkcaldie, R.K., Flitt, H.J., 1991. Review of test methods
pyrrhotite flotation. Transactions of the SME/AIME 272, 1885–1907. for abrasive wear in ore grinding. Wear 146 (2), 389–408.
Peng, Y., 2002. Grinding Environment Studies in the Control of Oxidation and Tao, D., Parekh, B.K., 2004. Corrosion Protection of Grinding Mills in the Phosphate
Interactions between Sulphide Minerals and Grinding Media. PhD Thesis, Industry Using Impressed Current Technology. Technical Report 01-170-208.
University of South Australia, Adelaide, SA, Australia. Florida Institute of Phosphate Research, 1855 West Main Street, Bartow, FL
Peng, Y., Grano, S., 2010. Effect of grinding media on the activation of pyrite 33830, USA.
flotation. Minerals Engineering 23 (8), 600–605. Tao, D., Chen, G.L., Parekh, B.K., 2005. Corrosion protection of mild carbon steel
Peng, Y., Grano, S., Ralston, J., Fornasiero, D., 2002. Towards prediction of oxidation media in phosphate grinding mill using impressed current technology. Minerals
during grinding I. Galena flotation. Minerals Engineering 15 (7), 493–498. Engineering 18 (5), 481–488.
Pintaude, G., Tschiptschin, A.P., Tanaka, D.K., Sinatora, A., 2001. The particle size Thornton, E., 1973. The effect of grinding media on flotation selectivity. In:
effect on abrasive wear of high-chromium white cast iron mill balls. Wear 250 Proceedings 5th Annual Meeting Canadian Mineral Processors. Mines Branch,
(1–12), 66–70. Ottawa, pp. 224–239.
Pitt, C.H., Chang, Y.M., Wadsworth, M.E., Kotlyar, D., 1988. Laboratory abrasion and Tolley, W.K., Nichols, I.I., Huiatt, J.I., 1984. Corrosion Rates of Grinding Media in Mill
electrochemical test methods as a means of determining mechanism and rates Water. US Bureau of Mines Report of Investigations, RI 8882, 18pp.
of corrosion and wear in ball mills. International Journal of Mineral Processing Vathsala (sic), Natarajan, K.A., 1989. Some electrochemical aspects of grinding
22 (1–4), 361–380. media corrosion and sphalerite flotation. International Journal of Mineral
Pons, M.N., Belaroui, K., Simer, N., 2004. Characterisation of grinding media wear by Processing 26, 193–203.
visual texture. International Journal of Mineral Processing 74, S255–S265. Vermeulen, L.A., Howat, D.D., 1986. Abrasive and impactive wear of grinding balls in
Prentice, T.K., 1943. Ball wear in cylindrical mills. Journal of Chemical Metallurgy, rotary mills. Journal of the Southern African Institute of Mining and Metallurgy
Mining Society of South Africa, 99–131. 86 (4), 113–124.
Qian, H.Y., Kong, Q.G., Zhang, B.L., 2013. The effects of grinding media shapes on the Yelloji Rao, M.K., Natarajan, K.A., 1988. Electrochemical aspects of grinding media-
grinding kinetics of cement clinkers in ball mill. Powder Technology 235, 422– mineral interaction on sulphide flotation. Bulletin of Material Science 10 (5),
425. 411–422.
Radziszewski, P., 1997. Predictive model for ball mill wear. Canadian Metallurgical Yelloji Rao, M.K., Natarajan, K.A., 1989a. Effect of electrochemical interactions
Quarterly 36 (2), 87–93. among sulphide minerals and grinding media on chalcopyrite flotation.
Radziszewski, P., 2000. Developing an experimental procedure for charge media Minerals and Metallurgical Processing, 146–151.
wear prediction. Minerals Engineering 13 (8–9), 949–961. Yelloji Rao, M.K., Natarajan, K.A., 1989b. Effect of galvanic interaction between
Radziszewski, P., 2002. Exploring total media wear. Minerals Engineering 15 (12), grinding media and minerals on sphalerite flotation. International Journal of
1073–1087. Mineral Processing 27, 95–109.
Raghavendra, H., Bhat, K.L., Rajendra Udupa, K., Rajath Hegde, M.M., 2010. Grinding Yelloji Rao, M.K., Natarajan, K.A., 1990. Effect of electrochemical interactions among
wear behaviour of stepped austempered ductile iron as media material during sulphide minerals and grinding medium on the flotation of sphalerite and
comminution of iron ore in ball mills. In: International Conference on Advances galena. International Journal of Mineral Processing 29 (3–4), 175–194.
in Materials and Processing Technologies (AMPT2010), 24–27 October 2010, Yildirim, K., Austin, L.G., 1998. The abrasive wear of cylindrical grinding media.
Paris, France, AIP Conference Proceedings, 1315, pp. 1341–1346. Wear 218, 15–20.
Rajagopal, V., Iwasaki, I., 1992a. The properties and performance of cast iron Yuan, X.-M., Pålsson, B.I., Forssberg, K.S.E., 1996. Flotation of a complex sulphide ore
grinding media. Mineral Processing & Extractive Metallurgy Review 11, 75– II. Influence of grinding environments on CuFe sulphide selectivity and pulp
106. chemistry. International Journal of Mineral Processing 46 (3–4), 181–204.

Minerals Engineering: Chris Aldrich

Uploaded by

Copyright:

Available Formats

Minerals Engineering: Chris Aldrich

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Minerals Engineering: Chris Aldrich

Uploaded by

Copyright:

Available Formats

How do different microstructures affect the wear resistance of grinding media?

What factors influence the wear rate of cast iron balls during wet grinding?

Minerals Engineering 49 (2013) 77–91

Contents lists available at SciVerse ScienceDirect

Consumption of steel grinding media in mills – A review

⇑ Address: Department of Metallurgical and Minerals Engineering, Western

5.4.1. Wear rate criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

1. Introduction The cost associated with grinding media is chieﬂy determined

Fig. 1. Materials used in wear resistant applications in comminution.

2.3. Media shape

Grinding media Mineral References

Grinding media Mineral References

4.6. Particle size 4.9. pH

Fig. 4. Grinding media wear models.

Fig. 5. Prediction of wear with neural network model.

Fig. 6. Variable importance analysis with neural network model.

Fig. 7. Relationship between wear and time.

Fig. 8. Structure of a classiﬁcation tree ﬁtted to the wear loss data.

You might also like