Advanced Powder Technology 25 (2014) 16721678

Contents lists available at ScienceDirect

Advanced Powder Technology

journal homepage: www.elsevier.com/locate/apt

Original Research Paper

Investigation on pre-weakening and crushing of phosphate ore using

high voltage electric pulses
Seyed Mohammad Razavian, Bahram Rezai , Mehdi Irannajad
Department of Mining and Metallurgical Engineering, Amirkabir University of Technology, Tehran, Iran

a r t i c l e

i n f o

Article history:
Received 19 October 2013
Received in revised form 12 April 2014
Accepted 10 June 2014
Available online 30 June 2014
Keywords:
High voltage electric pulse crusher
Comminution
Phosphate ore
Energy efciency

a b s t r a c t
Mechanical size reduction processes in the mineral processing industry are energy intensive. In this
study, a novel High Voltage Electric Pulses Crusher (HVEPC) was designed and developed to pre-weaken
and crush a phosphate ore. To compare the effectiveness of the new HVEPC in crushing with the conventional crushers, several rock samples of a phosphate ore were used and the effect of the capacitor, the
voltage level and voltage rise time on the phosphate ore breakage was investigated. Comparing the
particle size distribution of the crushed samples using both the HVEPC and the conventional crushers
revealed the difference between their breakage mechanisms. The nal results showed that applying
the high voltage pulses at specic energy of 35 kWh/t could signicantly increase and extend the cracks
and microcrakcs inside the rocks and consequently lead to decline in the Bond crushability and Abrasion
indices of the crushed samples by 10.6% and 28.1%, respectively.
2014 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder
Technology Japan. All rights reserved.

1. Introduction
Comminution, particularly in ner sizes to achieve the required
degree of liberation or specic surface, is a critical unit operation in
terms of energy consumption and process optimization. Several
research have shown that the specic energy consumption in
comminution varies from a few kWh/t for crushing to 1060 kWh/t
for AG/SAG and ball milling, and over 100 kWh/t for ultra-ne
grinding [1]. Sustainable development and use of new raw materials require new, better and more efcient processes. Accordingly,
in the eld of comminution, it is important to consider new
methods based on lower energy consumption and higher degree
of liberation. Studies have shown that conventional mechanical
comminution methods are largely based on application of
compressive force, resulting in high energy dissipation and slime
production. It is known that developing the tensile force based
methods could signicantly decrease the slime production while
reducing energy consumption. Therefore, several novel comminution methods such as High Pressure Grinding Roller (HPGR) [2],
microwaves treatment [3,4], Electro Hydraulic Disintegration
(EHD) [5] and Electrodynamic Disintegration (ED) [6] have recently
been studied and presented.

Corresponding author. Tel.: +98 2166460707; fax: +98 2166419728.

E-mail addresses: rezai@aut.ac.ir, b_rezai@hotmail.com (B. Rezai).

In this study, one of the novel comminution methods based on

particle disintegration using high voltage electric pulses was investigated. The history of research in this eld is mainly related to the
works done by Andres [710]. According to Andres [7], the idea of
using electrical disintegration methods was rstly developed by
Yutkin [11] in 1955 when he found that electrical current can crush
the suspended particles during electrolyzing water. Andres et al.
[9] investigated various minerals including oxide ores such as
hematite, PGM, Cu-containing suldes complexes and Pentlandite.
The experimental results indicated that higher percentage of the
liberated minerals and lower ne particles in electrical comminution exist in comparison with the case of mechanical comminution.
Cabri et al. [1222] at the Institute of CNT-MN in Canada have also
investigated electrical and electrohydraulic disaggregation and
especially the development of instruments in practical eld of mineral processing. Lastra et al. [13] compared the liberation study of
minerals in Merensky reef samples crushed by electrical and conventional crusher, and showed that the liberation of chromite,
Pentlandite, pyrrhotite and PGM using electrical crusher was
higher than that obtained by conventional jaw crusher. Ito et al.
[23] found that electrical disintegration on samples containing coal
represents a higher degree of liberation than mechanical grinding.
However, there is limited information about energy consumption
in electrical disintegration method.
The most comprehensive activity carried out on electrical
comminution in the past four years with the aim of energy

http://dx.doi.org/10.1016/j.apt.2014.06.010
0921-8831/ 2014 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.

S.M. Razavian et al. / Advanced Powder Technology 25 (2014) 16721678

consumption optimization was in JKMRC (Julius Kruttschnitt Mineral Research Center) [1,2426]. The activities of JKMRC have been
conducted in two major areas using high voltage pulses: particle
pre-weakening and mineral liberation which both focused on
improving energy efciency. These studies showed that the ores
treated by high voltage electric pulses with low specic energy
input (a few kWh/t) developed cracks and microcracks in the rock,
and became weaker than untreated particles. Comparison of Bond
work indices showed that the electrical treatment can reduce the
energy consumption in downstream grinding processes by up to
24%. Moreover, research conducted in JKMRC showed that higher
mineral liberation can be achieved in the electrical comminution
[1]. In other words, lower consumption of energy in the electrical
method can result in proper liberated minerals. Wang et al. [24]
studied the factors affecting the electrical grinding for six ores. In
this study, parameters such as feed particle size, undersize
aperture, failure potential and energy input were considered. In
addition, numerical simulations were used to show the distribution and intensity of the electric eld around and within the particles. Accordingly, the electrical comminution is strongly dependent
on the electrical characteristics of the ore. Dal Martello et al. [27]
investigated the effect of the electrical comminution in reducing
impurities in quartz ore to produce silicon for solar cells application. Comparing the magnetic separation efciency of the samples
produced by the mechanical and electrical methods showed a
signicant decrease in iron impurities content of the samples
crushed using the electrical methods.
Dodbiba et al. [28] used both mechanical and electrical comminution for recovery of indium from LCD screens disposed of by
leaching, and reported the results. They concluded that sample
preparation using electrical comminution was more efcient and
environmentally friendly than mechanical comminution.
In this paper, the effect of high voltage electrical pulses on phosphate ore comminution is presented. Phosphate ore was chosen
because its comminution inherently produces a high level of slime
which disrupts the downstream processes and reduces the quality
of the nal product. In this study, a high voltage electrical pulses
unit was designed, developed and tested to investigate the effect
of the instrumental parameters on failure and comminution of
phosphate ore and compare the results with the mechanical
comminution (crushing and grinding).
2. Materials and methods
2.1. Materials tested
The sample used in this study was obtained from the Esfordi
phosphate mining complex located in Bafgh, Iran. About 400 tons
of a representative sample from different zones of the mine were
collected and crushed in the crushing plant. The samples were then
further homogenized in the blending room and approximately 100
tons were sent to IMPRC for the laboratory and pilot scale tests.
The particle size distribution of this sample showed that 80% of
particles were smaller than 9.6 mm, 25% were ner than 150 lm
which is the acceptable size for the downstream process and
11.5% were ner than 38 lm.
Mineralogical studies were done on the thin-smooth polished
sections prepared from the main sample and analyzed using a
polarizing optical Zeiss Axioplan 2 microscope with light reection
and transmission. These studies showed that the major minerals
present in the sample are apatite, hematite and quartz. Hematite
is mostly inter-locked by contact and in some cases as 0.1
0.01 mm inclusions in other minerals. Apatite is interlocked with
hematite, calcite and quartz mainly of double type and less frequently triple type which decreased with particle size. The result
of the XRF analysis done on the sample using Fillips Magic_pro
instrument is given in Table 1.

1673

2.2. Crushability work index measurement

Particles of broken phosphate rock passing a 75 mm square gird
and retained on a 50 mm square were mounted between two
opposing equal 13.6 kg weight which swing on bicycle wheels.
When the wheels were released the weights impact simultaneously on opposite sides of the measured smallest dimension of
the rock. The height of fall was progressively increased until the
rock broke. The work index was obtained from the average of at
least ten breaks and is calculated as [29]:

W i 53:49

ICS
SG

where ICS is the impact crushing strength in Joules per mm of rock

thickness, SG is the specic gravity of the rock, and Wi is the work
index in kWh/t.
2.3. Abrasion index measurement
The Bond Abrasion test determines the Abrasion index, which is
used to determine steel media and liner wear in crushers, rod mills
and ball mills [30]. The Abrasion index is determined from the
weight loss of the standard alloy steel paddle hardened to 500 Brinell under standard operating conditions. 1.6 kg of ore sample in
size class of (19 + 12.5 mm) was used to perform the test. From
plant data, empirical equations correcting with Abrasion index
were developed to estimate rod, ball, mill liner and crusher liner
wear rates. These equations are shown in Table 2 [31]. Abrasion
index tests were carried out on both high voltage pulses and
conventional crusher products to assess the pre-weakening effect
on media and liners wear consumption rate.
2.4. High Voltage Electric Pulse Crusher (HVEPC)
The pulse generator as a standard electrical device which was
capable of generating high voltage pulses and comparable with
that of natural lightning came into the existence at the end of
19th century [32]. Nicolas Tesla used a coils resonant transformer
circuit for the wireless transfer of electrical energy and created the
electrical super weapon. His coils circuit included a transformer,
several high voltage capacitors, and a spark gap for generating high
voltage pulses. Nevertheless, the contemporary pulse generators
were developed in 1927 by Edwin Marx in the Braunschweig University in the form of a network of high voltage capacitors connected in parallel and interspersed with resistors to produce an
additive capacitance of the network. The capacitors were connected together in series and were charged in parallel through
the resistors but a gap was left in this circuitry [32].
Several research studies have used the SELFRAG lab machine to
investigate the effect of high voltage electrical pulses on comminution circuits [1,2427]. However, the effect and role of the
instrumental parameters on the nal product quality has not
comprehensively been studied. Therefore, it was aimed in this
study to design and fabricate a new High Voltage Electric Pulse
Crusher (HVEPC) consisted of three main parts to further
consider the effect of design and process parameters on the
comminution circuits. The rst part of the HVEPC contains the
electrical pulse generation circuit with required specications as
pulse generator.
A schematic of the HVEPC developed in this study is shown in
Fig. 1. Three main parameters including voltage, Capacitance and
voltage rise time affect the operation of the instrument. Using an
HV Transformer, urban 220 volt AC circuit was converted to a maximum of 100 kV AC. Then, the voltage was increased to a maximum
value of 140 kV DC V. The same voltage could also be used as the
pulse. Another parameter was the capacitance of the capacitor

1674

S.M. Razavian et al. / Advanced Powder Technology 25 (2014) 16721678

Table 1
Chemical analysis of phosphate ore sample.
Composition

Fe2O3

P2O5

SiO2

CaO

MgO

Al2O3

K2O

SO3

Cl

Y2 O3

L.O.I

Mass (%)

42.50

15.74

14.27

19.60

3.97

1.01

0.17

0.15

0.11

0.08

2.40

Table 2
Empirical equations correcting with Ai to estimate rod, ball and mill liner wear.
Rod mill
Wet

Rod wear rate (kg/kW-h)

0.159(Ai-0.020)0.2

Liner wear rate (kg/kW-h)

0.0159(Ai-0.015)0.3

Ball mill
Wet
Dry

Ball wear rate (kg/kW-h)

0.159(Ai-0.015)0.34
0.023(Ai)0.5

Liner wear rate (kg/kW-h)

0.0118(Ai-0.015)0.3
0.0023(Ai)0.5

before the spark gap which had a direct relation with the discharge
pulse. The capacitor was slowly charged from a DC source until the
spark gap broke down. This spark gap acted as a voltage-limiting
and voltage-sensitive switch of which its ignition time (time to
voltage breakdown) was very short.
The second part of the HVEPC, indicated as EPC in Fig. 1, was
the crushing chamber consisting of a rod-shaped copper electrode. In addition, the electrical disintegration system had also a
current transformer and a coil type amperemeter that was connected to the oscilloscope and a computer. The voltage between
the generator and ground electrode made of a steel plate was
measured by a probe connected to the oscilloscope. The ground
electrode consisted of circular apertures to pass the crushed
particles and prevent over-crushing. The sample was immersed
in the water in the chamber between two electrodes and then a
high-voltage pulse was directly discharged to disintegrate the
sample.
The third part was related to the safety accessories. Considering
the safety issues in all cases of using high voltage power supplies
was a crucial part of this study to design and install the necessary
safety equipment.
The energy of discharged electrical pulse was calculated by
Eq. (2):

E 1=2CV max 2

where C, is the capacity in Farad, Vmax is the maximum DC voltage

in volt and E is the energy discharged by electrode in joules (w.s).
The value of E in the Eq. (2) is the amount of energy discharged
from one electric pulse. Therefore, if more pulses were used, the
total number of pulses and energy were calculated. It should be
noted that the control table of the system was equipped with a
periodic shutdown system. Thus, by adjusting the frequency of
pulse discharge, the required period of time could be easily determined and then the number of pulses could be obtained. For better
comparison with the amount of energy consumption in mechanical
systems of mineral processing, taking into account the weight of
the sample to be tested and conversion factors, energy consumption unit was presented as Kilowatt hour per ton (kWh/t).
Table 3 shows the conditions of HV pulses tests. Each HV pulses
test was conducted with a batch sample of 200 g. By considering
the sample weight and conversion coefcients the energy of discharge pulse in kWh/t was calculated and by reverse calculation,
the number of discharge pulse was counted to achieve 50 kWh/t
of total cumulative energy consumption. It should be noted that
the control table of the system was equipped with a periodic shutdown system and so, by adjusting the frequency of pulse discharge,
the required period of time could easily be determined and then
the number of pulses could be obtained.
The tests were done by changing the particle size fraction, the
voltage and the capacitance each of which in at least three levels.
Based on the characteristics of the Esfordi phosphate ore, the three
levels of the particle size fractions were 19 + 12.5 mm,
12.5 + 6.35 mm and 6.35 + 3.35 mm. The clearance of the electrodes was adjusted proportional to the particle size, such that
the specic size of each fraction (geometrical average) was calculated, and was considered as clearance.

Fig. 1. High voltage pulse electric crushers schematic diagram and circuit.

1675

S.M. Razavian et al. / Advanced Powder Technology 25 (2014) 16721678

Table 3
HV tests Calculations to achieve 50 kWh/t of energy consumption.
No.

Sample weight (g)

No. of pulses

Capacitor (nF)

Voltage (kV)

Energy of pulse (w.s)

Energy of pulse (kWh/t)

Total Energy (kWh/t)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200

6000
2667
1500
960
667
4500
2000
1125
720
500
3600
1600
900
576
400
750
480
334

30
30
30
30
30
40
40
40
40
40
50
50
50
50
50
60
60
60

20
30
40
50
60
20
30
40
50
60
20
30
40
50
60
40
50
60

6
13.5
24
37.5
54
8
18
32
50
72
10
22.5
40
62.5
90
48
75
108

0.008
0.019
0.033
0.052
0.075
0.011
0.025
0.044
0.069
0.100
0.014
0.031
0.056
0.087
0.125
0.067
0.104
0.150

50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.000
50.100

3. Results and discussion

3.1. Analysis of comminution process
Fig. 2 shows the effect of the operating parameters in the electrical crushing of phosphate ore in three size fractions of
19 + 12.5 mm, 12.5 + 6.35 mm and 6.35 + 3.35 mm. The
concept of comminution has been considered as the cumulative

passing percent from control sieve in each fraction. To achieve

the same conditions for comparison in all cases, the energy consumption for all of experiments was 50 kWh/t. Two operating
parameters including voltage (kV) and capacitance (nF) were also
considered. The following results were deduced by observing
Fig. 2:
 In each particle size fraction, increase in voltage and capacitance resulted in improvement in crushing. This was due to discharge energy of each pulse, which is proportionate to Eq. (1).
Although, the total energy consumption was the same in these
tests, the energy of each pulse was the effective factor in better
(more) crushing.
 Comparing the crushing mechanism of the particle size fractions showed that the breakage was improved in smaller size
fractions. In other words, however the total energy consumption was identical in each test, the size-specic energy (dened
as energy divided by particle mass) for the smaller particles was
larger due to the smaller mass of each particle.
 It was found that the voltage had a more signicant effect on
the crushing than the capacitance which might be related to
the amount of energy consumption per pulse.
The effect of the specic energy of pulse in J/cm on breakage for
different size fractions is shown in Fig. 3. The energy of pulse
discharge directly affected the crushing of particles and the minimum energy of pulse for breakage was related to the particle size

Fig. 2. Effect of instrumental factors in electrical crushing of 3 fractions of 12.519,

6.3512.5 and 3.356.35 mm phosphate ore.

Fig. 3. Effect of specic energy of the pulse in J/cm on 3 fractions of 12.519, 6.35
12.5 and 3.356.35 mm phosphate ore crushing.

1676

S.M. Razavian et al. / Advanced Powder Technology 25 (2014) 16721678

Fig. 4. Relationship between the minimum specic energy of pulse and particle size
of phosphate ore.

of each fraction. In other words, the instrument was capable of

crushing the mineral when minimum energy of pulse discharge
was provided. It is clear that this minimum is proportional to the
type of ore minerals in addition to particle size. The initial tests
for providing minimum energy of pulse discharge must therefore
have been done separately for each ore. Fig. 4 shows the relationship between the minimum specic energy of pulse and particle
size of the phosphate ore sample. As it is shown in Fig. 4, Eq. (3)
can be presented with acceptable approximation to calculate the
minimum pulse energy required for crushing the sample.

Es 17:29 lnd  15:16

where Es, is the specic energy of discharge pulse (J/cm) and d, is

the particle size (mm).
Two points should be considered in this regard; rst, although
determining the minimum pulse energy species the minimum
operating conditions, these conditions are not clear for maximum
energy of discharge. This point becomes more pronounced when
crushing is enhanced with increased energy as shown in Fig. 3.
Moreover, the results showed that increase in energy of pulse discharge could improve the material crushing. However, this could
not only lead to excessive crushing that may cause problems, but
could also cause damages to both the measuring equipment
around the instrument and the mineral structure. In other words,
instead of creating cracks in contact surfaces of minerals, electrical
discharge occurs within the minerals, which would not be appropriate in such mineral processing operations. According to these
explanations, selection of pulse discharge energy should have been
optimized and preliminary tests should have been conducted
regarding the type of minerals. For generating a given energy of
pulse discharges both voltage and capacitance can be varied

Fig. 6. Average particle size distribution of jaw crushing (JC) and HVEPC in 2 forms:
cumulative passing and retained on sieve (normal).

according to Eq. (2). Thus, the effect of each of these parameters

should be taken into account. Experimental tests showed that the
choice of capacitor was more important than voltage in crushing of
various particle sizes. This means that although the increase in
voltage was more effective on pulse discharge energy, the type of
capacitor affected the way the pulse was discharged. To better
understand this latter phenomenon, some tests were conducted
on discharge voltage of electrical pulses at two different
conditions: 10 nF capacitor with 40 kV voltage and 40 nF capacitor
with 20 kV voltage. The results showed that the 40 nF capacitor
with 20 kV could have higher breakage effect particularly for larger
particle size fractions when the same rate of pulse discharge
energy was applied. To investigate the reason for this, the wave
shapes were recorded in these two conditions using an oscilloscope, which is shown in Fig. 5. As observed in Fig. 5, the effective
factor was the wave rise time. The rise time was approximately
500 ns in the rst (10 nF and 40 kV) and 1400 ns in the second
(40 nF and 20 kV) test condition. It is also apparent that the
discharge energy which was sharper in the rst case had a lower
width or effective range whereas in the second condition, the
effective range of discharge energy was higher.
3.2. Particle size distribution of nal crushed products
To compare the size distribution of the nal crushed products of
the conventional crushers, regardless of energy consumption, part
of the main sample was crushed using a jaw crusher. Fig. 6 shows

Fig. 5. Waveshapes of discharge voltage in two conditions: C:40 nF, V:20 kV and C:10 nF, V:40 kV.

1677

S.M. Razavian et al. / Advanced Powder Technology 25 (2014) 16721678

Table 4
Bond crushability test results (in kWh/t) of phosphate ore in terms of: non treatment,
pre-weakening by 3 kWh/t and 5 kWh/t (by HVEPC) respectively.
Test No.

Non-treatment

3 kWh/t

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

3.07
5.32
5.67
5.86
6.23
6.35
6.56
6.66
6.75
7.16
7.68
7.72
8.05
8.23
8.60
10.90
14.73
15.23
18.75
20.78

2.67
5.25
5.58
5.66
5.72
5.78
6.10
6.29
6.82
6.98
7.08
7.36
7.51
8.26
9.17
10.71
12.52
15.07
16.23
16.46

2.18
4.52
5.27
5.28
5.64
5.77
5.84
6.20
6.29
6.40
6.91
7.05
7.29
7.74
8.88
11.08
14.12
14.98
15.21
16.65

9.02
0

8.32
7.8

8.06
10.6

Average
Relative difference (%)

explosive plasma streamers at the interface of mineral components

of the ore aggregates with different permittivities and electrical
conductivities [9].

5 kWh/t

the average particle size distribution of the nal products of the

conventional jaw crusher and the developed HVEPC in the forms
of the cumulative passing percentage and retained percentage on
the sieve. Comparing the diagrams of cumulative passing percentage in Fig. 6, the distribution curve for the electrical method was
almost linear. In addition, comparing the distribution curves of
the retained percentages on the sieve indicated that particle size
distribution in the electrical comminution was closer to normal
pattern. In other words, the HVEPC produced less large particles
as well as too small particles. So in comparison of jaw crusher,
HVEPC produces proper particle size for grinding stage due to less
larger particles. These observations in the particle size distribution
clearly reected the differences between the electrical and
mechanical comminution mechanisms. In the mechanical method
using crushers, the dominant comminution force is pressure and
cleavage while the specic liberation effect at the disintegration
of minerals by electrical pulses is due to the locality of the

3.3. Phosphate ore crushability

Due to higher energy consumption in the electrical comminution, such method could be used for pre-weakening of the phosphate ore [1]. Applying low energy in the electrical method is
known to increase cracks and microcracks within the rock and
facilitate the mechanical comminution later in the process [1,7].
To compare the effect of the pre-weakening on crushability of
phosphate ore, the standard Bond test was used. Twenty pieces
of rock in a size range of 57.5 cm from the crushing plants average feed were prepared and two energy levels of 3 kWh/t and
5 kWh/t were considered for pre-weakening. Regarding the size
of the particles, 50 nF capacitor, 40 kV voltage and a 2 Hz frequency
were considered as conditions of the electrical comminution test.
The standard test of crushability index was performed on each
sample series the results of which are presented in Table 4.
As can be seen on Table 4, the relative reduced levels of energy
at 3 kWh/t and 5 kWh/t of energy levels were 7.8% and 10.6%,
respectively. This proved that pre-weakening stage of the rocks
before crushing stage could reduce comminution energy consumption by about 10%. In addition to reducing energy consumption, the
way the cracks are developed in the ore is also important. As the
mechanism of cracks production is based on electrical characteristics of the minerals, the cracks occur mostly in contact surface of
the minerals. This causes the beginning of the disintegration of
particles from contact surface of the minerals in later comminution
which is a favorable event in downstream concentration processes.
This latter effect can be observed in microscopic images shown in
Fig. 7.
As depicted in Fig. 7, the cracks created within the ore were in
contact surface of the minerals which led to separation of minerals
from their contact surfaces in downstream comminution.
3.4. Phosphate ore abrasion index
To compare the abrasion index of the phosphate ore sample in
both the mechanical and electrical treatments, primary particles of
the size 12.519 mm were obtained. In the electrical method, the

Fig. 7. Microscopic images (ppl) of phosphate ore pre-weakened by HVEPC.

Table 5
Wear rate (kg/kW-h) of rod, ball and mill liner by phosphate ore prepared by conventional mechanical comminution and electrical comminution.
RodMill
Wear rate (kg/kW-h)

Rod

Mech. comminution
HVEPC comminution

0.1135
0.1054

Reduction ratio (%)

7.1574

Ballmill (wet)
Liner

Ball

Ballmill (dry)
Liner

Ball

Liner

0.0097
0.0087

0.0905
0.0801

0.0072
0.0064

0.0104
0.0089

0.0010
0.0009

10.2265

11.5086

10.2265

15.1483

15.1483

1678

S.M. Razavian et al. / Advanced Powder Technology 25 (2014) 16721678

particles larger than 19.5 mm were prepared and crushed by

HVEPC using a control sieve of 19 mm, and the desired sample
was prepared. The electrical tests were conducted using 60 nF
capacitor, 80 kV voltage and 1 Hz frequency conditions. The abrasion index was determined to be 0.2057 g and 0.1481 g in the electric and mechanical methods respectively. The relative 28.1%
decrease of this index indicated the abrasion of grinding media
(ball and rod) and mills liners. According to the equations in
Table 2, the decline was as shown in Table 5.
4. Conclusion
A new High Voltage Electric Pulses Crusher (HVEPC) was constructed and developed in this study. The HVEPC consisted of three
main parts including the H.V. generator, the crushing chamber and
the safety accessories. The H.V. generator creates impulses based
on the Marx type generators using some capacitors and resistors.
To optimize the process variables, three parameters including particle size, voltage and capacitance were investigated at three different levels. The results of the crushing tests conducted on the
phosphate ore sample using both the HVEPC and the conventional
crushing methods revealed that:
 The discharge pulse energy was the main process variable
related to the capacitor and voltage in the electrical crushing.
To change the electro-hydraulic disintegration to the electric
disintegration, the minimum level of discharge pulses energy
was required. It was found that there was a logarithmic relation
between the discharge pulse energy and the particle size in the
phosphate ore crushing.
 The choice of capacitor was more important than voltage in creating of the proper discharge pulse energy. It was observed that
the larger was the capacitor the higher was the effective range
of discharge energy.
 Comparing the diagrams of cumulative passing percentage
exhibited that the distribution diagram of the electrical method
was very close to a linear curve. In addition, comparing the distribution curves of the retained percentage on sieve indicated
that particle size distribution in the electrical comminution
was closer to a normal distribution.
 The standard Bond crushability tests showed that the pre-weakening of the phosphate ore sample using the HVEPC with energy
consumption of 3 and 5 kWh/t reduced the crushability indices
by 7.8% and 10.6%, respectively. It was clear in the microscopic
images that the cracks were created at the minerals surfaces.
However the total energy consumption of crushing circuit
should be investigated in the continuous pilot or industrial
circuit.
 The relative reduction of standard Bond abrasion tests in the
mechanical crushing and the high voltage electrical crushing
was 28.1% for the phosphate ore sample.
Acknowledgments
The authors are grateful to the Iranian Mines and Mining Industries Development and Renovation Organization (IMIDRO) for
funding this PhD research and Iran Mineral Processing Research
Center (IMPRC) and Esfordi phosphate complex for their cooperation and substantial input into the research essential for this paper.
References
[1] E. Wang, F. Shi, E. Manlapig, Pre-weakening of mineral ores by high voltage
pulses, Miner. Eng. 24 (2011) 455462.
[2] R. Klymowsky, N. Patzelt, J. Knecht, E. Burchardt, Selection and sizing of high
pressure grinding rolls, in: A.L. Mular, D.N. Halbe, D.J. Barratt, (Eds.), Mineral
processing plant design, practice and control, SME, vol. 1, 2002, pp. 636668.

[3] S.W. Kingman, W. Vorster, N.A. Rawson, The inuence of mineralogy on

microwave assisted grinding, Miner. Eng. 13 (3) (2000) 313327.
[4] K. Barani, S.M.J. Koleini, B. Rezaei, Magnetic properties of an iron ore sample
after microwave heating, J. Separation Purication Technol. 76 (2011) 331
336.
[5] U. Andres, Parameters of disintegration of rock by electrical pulses, Powder
Technol. 58 (1989) 265269.
[6] U. Andres, Electrical disintegration of rock, Miner. Process. Extr. Metall. Rev. 14
(1995) 87110.
[7] U. Andres, Development and prospects of mineral liberation by electrical
pulses, Int. J. Miner. Process. 97 (2010) 3138.
[8] U. Andres, I. Timoshkin, M. Soloviev, Liberation of valuable inclusions in ores
and slags by electrical pulses, Powder Technol. 114 (2001) 4050.
[9] U. Andres, J. Jirestig, I. Timoshkin, Liberation of minerals by high voltage
electrical pulses, Powder Technol. 104 (1999) 3749.
[10] U. Andres, Dielectric separation of minerals, J. Electrostat. 37 (1996) 227248.
[11] L.A. Yutkin, Electrohydraulic Effect, Mashgis, Moscow, 1995.
[12] L.J. Cabri, New technology for process mineralogy: Electric Pulse
Disaggregation (EPD), CNT Mineral Consulting Inc. Ottawa, Canada, <http://
www.cnt-mc.com> 2008.
[13] R. Lastra, L. Cabri, P. Weiblen, Comparative Liberation study by image analysis
of Merensky Reef samples comminuted by electric-pulse disaggregation and
by conventional crusher, Proceeding of XXII International Mineral Processing
Congress, Cape Town, South Africa, vol. 1, 2003, pp. 251260.
[14] L. Cabri, N. Rudashevsky, V. Rudashevsky, T. Oberthur, Electric-Pulse
disaggregation (Epd), Hydroseparation (Hs) and their use in combination for
mineral processing and advanced characterization of ores, Proceeding of the
40th Annual Canadian Mineral Processors conference, Paper 14, 2008, pp. 211
235.
[15] L.J. Cabri, New developments in process mineralogy of platinum-bearing ores,
in: Proceedings of the Canadian Mineral Processors, 36th Annual Meeting,
Ottawa, 2004, pp. 189198.
[16] L.J. Cabri, A.M. McDonald, C.J. Stanley, N.S. Rudashevsky, G. Poirier, B.R.
Durham, J.E. Mungall, V.N. Rudashevsky, Naldrettite, Pd2Sb, a new palladium
antimonide from the Mesamax Northwest deposit, Ungava region, Qubec,
Canada, Mineralogical Magazine 69 (1) (2005) 8997.
[17] L.J. Cabri, M. Beattie, N.S. Rudashevsky, V.N. Rudashevsky, Process mineralogy
of Au, Pd, Pt ores from the Skaergaard Intrusion, Greenland, using new
technology, Miner. Eng. 18 (2005) 887897.
[18] L.J. Cabri, N.S. Rudashevsky, V.N. Rudashevsky, R. Lastra, Hydroseparation: a
new development in process mineralogy of platinum-bearing ores. CIM
Bulletin 99(1092) (2006) 1-7 <http://www.cim.org>.
[19] L.J. Cabri, D. Hoy, N.S. Rudashevsky, V.N. Rudashevsky, Mineralogical
evaluation of Au-Sb-As mineralization from the AD-MW zones, Clarence
Creek property, New Brunswick. Proceedings of the Canadian Mineral
Processors, 39th Annual Meeting, Ottawa, 2007, pp. 517533.
[20] L.J. Cabri, N.S. Rudashevsky, V.N. Rudashevsky, V.Ya. Gorkovetz, Study of native
gold from the Luopensulo deposit (Kostomuksha area, Karelia, Russia) using a
combination of Electric Pulse Disaggregation (EPD) and Hydroseparation (HS),
Miner. Eng. 21 (2008) 463470.
[21] L.J. Cabri, Y. Choi, C.H. Hamilton, P. Kondos, R. Lastra, Hydroseparation
concentrates and automated precious metal searches used to characterise
process products from selected mines, Ninth International Congress for
Applied Mineralogy ICAM 2008. The Australasian Institute of Mining and
Metallurgy, Publication Series No 8/2008, 261264.
[22] V.N. Rudashevsky, Advantages of EPD Spark-2 (CNT Mineral Consulting Inc.,
Canada) 2008 <http://www.cnt-mc.com/downloads/>.
[23] M. Ito, S. Owada, T. Nishimura, T. Ota, Experimental study of coal liberation:
electrical disintegration versus roll- crusher comminution, Int. J. Miner.
Process. 92 (2009) 714.
[24] E. Wang, F. Shi, E. Manlapig, Factors affecting electrical comminution
performance, Miner. Eng. 34 (2012) 4854.
[25] E. Wang, F. Shi, E. Manlapig, Mineral liberation by high voltage pulses and
conventional comminution with same specic energy levels, Miner. Eng. 27
28 (2012) 2836.
[26] E. Wang, F. Shi, E. Manlapig, Experimental and numerical studies of selective
fragmentation of mineral ores in electrical comminution, Int. J. Miner. Process.
112113 (2012) 3036.
[27] E. Dal Martello, S. Bernardis, R.B. Larsen, G. Tranell, M. Di Sabatino, L. Arnberg,
Electrical fragmentation as a novel rout for the renement of quartz raw
materials for trace mineral impurities, Powder Technol. 224 (2012) 209216.
[28] G. Dodbiba, H. Nagai, L.P. Wang, K. Okaya, T. Fujita, Leaching of indium from
obsolete liquid crystal displays: comparing grinding with electrical
disintegration in contex of LCA, Waste Management 32 (10) (2012) 19371944.
[29] T.J. Napier-Munn, S. Morrell, R.D. Morrison, T. Kojovic, Mineral Comminution
Circuits, Their Operation and Optimization, JK Mineral Research Center,
Australia, 1996.
[30] F.C. Bond, Metal Wear in Crushing and Grinding, vol. 07, Allis-Chalmers
Publication, 1963. p. 1701.
[31] A. Chester, Jr. Rowland, Selection of rod mills, ball mills and regrind mills, in:
A.L. Mular, D.N. Halbe, D.J. Barratt, (Eds.), Mineral processing plant design,
practice and control, SME, vol. 1. 2002, p. 754.
[32] E. Kuffel, W.S. Zaengl, J. Kuffel, High Voltage Engineering Fundamentals,
second ed., Butterworth-Heinemann, 2000.