16/09/15

Fundamentals of
mineral processing
and recycling
Prof. Rodrigo Serna
Dept. of Materials Science and Engineering
Fall 2015

Week 2

Particle technology
•  Solid particle properties
•  Liberation
•  Particle Size distribution
•  Density
•  Solids handling
•  Transport
•  Storage

1  
 16/09/15  

…so you have extracted some rocks

from the ground. What next?
•  Can you concentrate a
2 m-wide ore body?

•  The first step in a

mineral processing plant
is Comminution
•  Crushing and grinding

•  Two purposes:
•  Liberation
-  Gain access to the valuables entrapped in the mineral matrix
•  Obtain a target particle size
-  Ease of handling
-  Concentration operations have an optimum size range

Mineral processing operations:

Comminution circuits
•  A typical comminution step (i.e., crushing or milling) is
followed by a sizing step
•  Helps achieving the desired size distribution of particles
•  Can be used to prevent overgrinding and waste of energy
Example: Ball mills and hydrocyclones at a gold mine in Alaska
Feed

Sizing/
Undersize
Classificaiton

Oversize

Crushing/
Grinding

2  
 16/09/15  

Liberation
•  Liberation is the process by which valuable minerals are
exposed
•  The size reduction necessary for liberation is dictated by the
valuable grain size
-  Low concentration valuables such as Ag or PGMs require grinding down to
particles with a few µm

Liberation
•  Comminution produces particles with various degrees of liberation
•  “Degree of liberation” can be defined as the fraction of liberated particles in a
population of liberated and non-liberated (“locked”) particles
-  Strictly speaking, this means only particles with 100% valuable grade are considered
“liberated”
-  In practice, “liberated” particles can have a lower grade, depending on the grade at which
recovery is possible
•  There are certain “domains” where access to valuables will be more problematic
than others

Comminution

Example: Chalcopyrite concentrate from

Typical problem: is it economic to regrind these particles Papua New Guinea
(Source: http://cms.unige.ch/sciences/terre/research/Groups/
(called middlings) in order to liberate the valuables? mineral_resources/ore_dressing/ore_dressing.php)

3  
 16/09/15  

Liberation

Example: What is the degree of liberation of the mineral sample

described below?
Particle grades (%) 0,1-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-99,9 100 (free)

% Mineral in particle 6,5 3,5 1,5 1 1 1 2,5 6,5 11,5 23,5 41,5

% Recovery 34,1 92,2 98,7 97,7

Note: Since minerals are natural products, their properties are always a distribution

Liberation

•  And what about recycling of waste metals?

•  “Liberation” is also needed!

Cutting/
Shredding

4  
 16/09/15  

Particle size distribution

•  Through the use of comminution and classification/screening
stages, it is expected to produce particles with a target size
•  This process however, do not produce particles with perfect
geometrical shape

•  Challenge: how to define particle size for irregular shapes?

Particle size distribution

•  The simplest “dimension” to use is the diameter (or radius) of a

sphere

•  But since particles are not perfectly spherical, it has been proposed
to use
•  Equivalent diameters
•  Average diameters ≈
•  Weighted-average diameters

•  We have to keep in mind that these are approximations, and the less
spherical the particles will be, the more their behavior may deviate
from theory (since they are usually assumed to be spherical)
•  This intends to be corrected by using a shape factor

5  
 16/09/15  

Particle size distribution

•  Some equivalent diameters (d) are:
•  From geometrical properties:
-  Volume diameter (dV): The diameter of a sphere with the same volume as the particle
-  Surface area diameter (dS): The diameter of a sphere with the same surface area as the
particle
-  Surface-volume diameter (dSV): The diameter of a sphere with the same surface area-
to-volume ratio as the particle
-  Projection area diameter (dC): The diameter of a disk with the same projected area as
the particle
-  Martin diameter (dM): The length of a cord necessary to split the particle in two
equivalent areas

•  From physical properites:

-  Stokes diameter (dSt): The diameter of a sphere with the same settling velocity as the
particle
-  Drag diameter (dd): The diameter of a sphere with the same resistance to motion as the
particle

•  From measuring techniques:

-  Sieve size (dp): The minimum opening size in a sieve through which the particle passes

Particle size distribution

•  The “sphericity” of a particle can be defined with some shape
factors, such as
•  Circularity (ratio of perimeter and area)
4π A
fcirc =
P2

•  Waldell sphericity (based on dS and dV) 2

2
"d %
Ψ S,V =$ V '
# dS &
•  Aspect ratio (ratio of minimum and maximum orthogonal diameters)
dmax
dmin AR =
dmin
dmax
•  The closer these factors are to 1, the more spherical they should be

6  
 16/09/15  

Particle size distribution

Example: What are the dS, dV, dSV , dC and sphericity of the
following particles?

4
5

3
10 8

Particle size distribution

•  Now that we have agreed on using an “equivalent diameter”
as measurement of size, we can characterize the particles in a
sample
•  Remember that these are natural products, and they will not
have a single size, but rather a distribution
•  A distribution can be based on number, length, area, volume
or mass of the particles with a particular size
Frequency

Particle size

7  
 16/09/15  

Particle size distribution

•  Particle size distribution (PSD) is presented typically as either
frequency or cumulative distribution

Frequency Cumulative

Example of zircon sand PSD measured by laser diffraction

•  Expressed in mathematical terms:

•  Frequency = f(dp): dF(d p )
f (d p ) =
d(d p )
•  Cumulative frequency = F (dp): max
F(d p ) = ∫ f (d p )d(d p )
min

Particle size distribution

•  Question: For practical purposes, we need to use a single

value as measurement of particle size. What value would you
use?

•  Particle size can be expressed as:

•  Average
-  Arithmetic, quadratic, cubic or harmonic mean
-  Mode
-  Median
•  Passing size
-  The size at which a particular fraction of the sample is equal or smaller
-  A common measure in mining industry is d80, or the size at which 80% of sample
is smaller

8  
 16/09/15  

Measuring methods: Sieving

Sieves have different opening sizes, based on

industry standards (e.g. ISO R-565, DIN 4188, ASTM E-11)

Shake +
time

Measuring methods: Sieving

•  Sieving is simple and straight forward

method
•  It is however, defined by statistics
•  i.e., how probable it is that a particle will pass
through the mesh?
•  It is particularly suited for coarse particles
•  It has limitations at sizes smaller than 100 µm
•  Sieving of slurry (wet sieving) is possible and
it may aid in the measurement of fine particles
100%
•  Sieves require calibration and regular
Passing (Weight %)

inspection for wear 80%

60%
•  The concept of “passing” particle size is
40%
related to sieving
•  For example, a 60% passing particle size (d60) 20%
refers to the sieve opening dimension that
0%
would retain 40% of the sample 0 200 400 600 800 1000 1200 1400
Sieve opening in microns

9  
 16/09/15  

Measuring methods: Laser diffraction

•  Fast and easy measurement based on the
scattering of light by particles
•  Diffraction angle is inversely proportional
to particle size
•  Results also depend on refractive indices of
solid particles and liquid medium
•  Typically performed using dilute
suspensions
•  It is necessary to maintain a homogeneous
suspension for reliable measurements
•  Not for high density or soluble materials
•  Depending on the instrument and theory
applied, measuring of particles down to 0,1
µm is possible
•  A downside is that it does not produce
results that are directly comparable to
other methods
•  Useful for comparison of samples measured
with this technique
•  Otherwise, a correction factor may be used

Measuring methods: Microscope and

image analysis
•  Direct measurement of the
actual particles forming the
sample
•  It is by principle only two-
dimensional
•  Time consuming
•  Manual analysis is tedious
•  Automatic analysis require
sufficient computational power

Mineral composition and particle size of a chromite ore

measured with QEMSCAN

10  
 16/09/15  

Particle size distribution

•  Since measuring methods produce discrete data points, one

can use size distributions to produce a continuous PSD
profile
•  Developed before the computational-era, to present size
distributions with ease

•  Gates-Gaudin-Schumman ! d $
n

F(d p ) = ## p &&
" d p,max %

•  Roslin-Rammler-Sperling " " %

n %
d
F(d p ) = 1− exp $ − $$ p '' '
$ # d p,ref & '
# &

Week 2
“Landscape” by Grace Grothous

Solid particle properties

11  
 16/09/15  

Density of solid particles

•  Density is evidently a very important property to define the

behavior of any material
•  In the case of solid particles, and particularly solid powders,
density may not be defined as straight forward as in the case
of fluids
•  Question: What is density? What can be a good measure of
density for a mineral particle (i.e., mixture of solid
components)?

Density of solid particles

•  Solids particles will have: ρbulk

•  Solid density
-  Sometimes also referred as “wall density”, this
is the actual density of the solid material
-  In minerals processing, this will likely be used
only in suspension (slurries)
•  Particle density
-  If the material is porous, the absence of matter
in the porous voids have to be considered
-  A function of porosity (or pore volume)
•  Bulk density
-  The apparent density of solid particles filling
certain volume
-  Depends heavily on particular conditions of the
material, from particle shape, to stacking, to ρsolid
measuring technique, so it is usually tested on
site ρparticle
-  This is the density used in practice when
dealing with dry particles

12  
 16/09/15  

Recapitulating

•  Mineral particles have:

•  Irregular shapes
•  A distribution of several particle sizes
•  Various degrees of liberation (depending on grain size and particle
size)
•  Three types of density: solid, particle and bulk

•  These characteristics will influence the behavior of the

mineral
•  During concentration operations
•  In their transport and storage

Transport of dry particles

•  Dry minerals are transported by:

•  Gravity (e.g. unloading of excavators)
•  Conveyor belts and elevators (skips)

13  
 16/09/15  

Storage and discharge of dry particles

•  Silos
•  The proper design of silos is of interest since they need to
-  Withstand the pressure and stress of static particles
-  Allow the proper discharge of solids, as required
•  There are some typical discharge profiles for powders in silos

Mass flow Funnel flow Tunnel flow Rathole Arching

What we want Inefficient use No flow!

of silo (most common in
cohesive solids)

Storage and discharge of dry particles

•  Basics: stress forces in solid particles
•  At the point of contact, there are normal and
shear forces
•  If particles are non-cohesive, normal force can
only be compressive
•  If shear force is higher than the product of the
friction coefficient (fc) and the normal force,
the granules will slide
•  Shear stress force is measured experimentally
using methods such as Jenike’s ring shear test

14  
 16/09/15  

Storage and discharge of solid particles

Obtained from shear
stress testing

•  Design of silos
•  Geometry

(a) Conical; (b) Wedge; (c) Transition hopper 1;

(d) Chisel-shape; (e) Pyramidal; (f) Transition hopper 2

Storage and discharge of solid particles

•  The exertion of external forces on powders can produce different

phenomena
•  When sufficient force is provided to overcome a critical yield stress, the
powder will flow
•  A sufficient air flow can fluidize powders with the proper size and
density, making them behave as fluids
•  Powders can decrease their bulk density, this process is called
consolidation
•  Particles with similar properties (e.g. density) can tend to localize in
different areas of the powder bed, this process is called segregation
•  Other factors influencing flow of powders in a silo are
•  Moisture content
•  Temperature
•  Storage time

15  
 16/09/15  

Assignment
The following dry classification setup is used to separate the fine
fraction of a chalcopyrite slurry. Determine:
•  Particle size distribution, and mass pull of the overflow
•  d80 particle size for all streams
•  Compare the d50 particle size and the average particle size (arithmetic
mean) of the underflow product
•  If due to an operational failure, the underflow would decrease to only
80% of its original value, how would that affect the PSD and d80 of the
underflow product (assuming PSD of overflow does not change)?
Overflow (coarse fraction)
Feed Flow rate
Feed rate Bulk density
Bulk density Classification Particle size distribution
Particle size distribution
Underflow (fine fraction)

16