3.

Particle Characterisation
The need to characterise particles is encountered in all engineering disciplines. Chemical
engineers encounter particulate solids in crushing, drying, crystallisation, catalytic reactions
and dust collection, to name a few. Civil engineers deal with concrete aggregates, soils and
river silting. Mining and metallurgical engineers encounter particles whenever an ore is handled.
In many industrial operations, a single number is required to characterise the particle or particle
assembly. This is not an easy task, and sometimes two (or more) parameters are required to
describe the particle(s) adequately.
This chapter will discuss the concept of equivalent diameters, both for single particles and a
population of particles. This knowledge will be used in subsequent chapters when unit
operations like settling, fluidisation, filtration, etc. will be designed and analysed.

3.1 The single particle.

Consider a rectangular box, and you are asked to give the size. To only measure the height,
you are only partially correct. A complete answer will be the height, width and length. The
situation becomes even more complicated when one tries to describe other complex shapes.
A quality control manager ideally would like to have a single answer to determine whether the
particle size has increased or decreased. Similarly, in modelling processes, a single particle
dimension is required.

3.1.1 Equivalent diameters

A sphere is the only shape that can be described by a single number, its diameter. An equivalent
diameter is defined as the diameter of a sphere that has the same properties of the particle.
3.1.1.1 Equivalent volume diameter (dV)
This is the diameter of a sphere that has the same volume as the particle
The volume of a sphere is /6d3. Thus /6dV3 = Vp

6V P
or dv  3 …1

3.1.1.2 Equivalent Surface area diameter
This is the diameter of a sphere that has the same surface area as the particle
The surface area of a sphere is d2.

AP
Thus dS2 = Ap or ds  …2

3.1.1.3 Equivalent Surface area: Volume diameter (Sauter diameter)
This is the diameter of a sphere which has the same volume to surface area ratio as the particle

 d sv3 Vp 6V p
6  or d sv  …3
d sv2 Ap Ap
3.1.2 Shape factors
The diameters defined above give an indication of the size of a particle, but the overall shape
is still unknown. Two particles can have the same equivalent diameter, but completely different
shapes. Table 2 lists some of the more general shape descriptions
Table 2 Qualitative terms for particle shape
Name Description
Acicular needle-shaped
Angular sharp-edged or having a rough polyhedral shape
Crystalline freely developed in a fluid medium, of geometric shape
Dentritic having a branched crystalline shape
Fibrous regular or irregular thread-like
Flaky plate-like
Granular approximate equidimensional irregular shape
Irregular lacking any symmetry
Modular having rounded, irregular shape
Spherical global shape

From a modelling point of view, one would like to have a single value that describes the overall
shape of the particle. The following two parameters are most frequently used.
3.1.2.1 Sphericity ()
The most common shape factor used in industry is sphericity and is defined as:

Surface area of a sphere having the same volume as the particle

 …4
surface area of the particle

From equations 1 and 2 the sphericity can be calculated as

2
d 
   v  …5
 ds 
3.1.2.2. Aspect ratio (AR)

Another shape factor used in industry is the aspect ratio. The aspect ratio is defined as the
largest dimension divided by the smallest orthogonal dimension.

3.2 Particle size distribution (PSD)

In industry, one usually encounters particles that range from small to large in the same sample.
There are several methods for particle size measurements where the raw data collected is
either as a number distribution (microscopy) or a mass or volume distribution (sieving). The
general principle of microscopy is that the number of particles of fixed size is counted and
expressed as a percentage of the total number of particles. For sieve analysis, the total mass
of a specific size range is determined and then expressed as a percentage of the total mass of
particles.
Particle size distributions (PSD) can be expressed as frequency distributions or as cumulative
mass curves. Figures 5 and 6 are examples of the corresponding data in Table 4.
If the frequency distribution is defined as f(di), then the cumulative frequency distribution is
di

F (d )   f ( d )dd …6
0

dF (d )
and  f (d ) …7
dd

Note that  f (d )dd  1
0
…8

3.2.1 Methods to measure the particles size distribution

3.2.1.1 Sieve analysis
Sieve analysis is performed using a stack of sieves, as shown in Figure 1 with the biggest
aperture size at the top. Consecutive lower sieves have smaller apertures ending with a solid
pan at the bottom.
The aperture size or the mesh number identifies the sieve size. The mesh size is defined as the
number of wires per linear inch. The US Tyler and British fine mesh sizes are shown in Table
3.
Sieve analyses are typically used for particles larger than 45 m, although wet sieving analysis
can be performed to as low as 25 m. Sieve analysis is a mass distribution equal to the sieve
size. If it is assumed that the solid density of all the size fractions is the same, it also represents
the volume distribution.
Table 3 Mesh sizes
Aperture British fine Aperture
US Tyler (m) Mesh (m)
16 991 14 1204
20 833 16 1003
24 701 18 853
28 589 22 699
32 495 25 600
35 417 30 500
42 351 36 422
48 295 44 353
60 248 52 295
65 208 60 251
80 175 72 211
100 147 85 178
115 124 100 152
150 104 120 124
Figure 1. Stack of sieves on a
170 88 150 104
shaker.
200 74 170 89
230 61 200 76
270 53 240 66
 325 43 300 53
Example 3.1
Mass Cum
British Size Avg size retained Fraction
mesh (m) (m) (g) Fraction > dP
36 422
52 295 358.5 0 0.0 0.0
72 211 253 20 10.0 10.0
100 152 181.5 50 25.0 35.0
150 104 128 60 30.0 65.0
200 76 90 50 25.0 90.0
300 53 64.5 20 10.0 100.0
Pan 0 0.0 100.0

3.2.1.2 Microscopy
A sample of the particles is spread in a single layer such that individual particles can be
observed, as shown in Figure 2.

Figure 2. Typical microscope image with individual particles

Each particle is measured with a corresponding characteristic diameter. Martin’s or Ferret’s

characteristic diameters can be used. The choice will influence the result.
To obtain a statistically valid distribution, at least 2,000 particles should be measured. Manual
microscopic analysis can be a tedious task. With the use of image analysis computers, the task
is performed much faster and more accurate
Microscopy gives a number distribution.

3.2.1.3 Laser diffraction

Laser diffraction analysis, also known as laser diffraction spectroscopy, is a scientific technique
that utilises diffraction patterns of a laser beam passed through any object ranging from
nanometres to millimetres in size to measure the geometrical dimensions of a particle. Laser
diffraction analysis is based on the Fraunhofer diffraction theory, which states that the intensity
of light scattered by a particle is directly proportional to the size of the particle.
The laser is passed through the substance being analysed. The diffracted light is focused onto
a detector, which measures the angular distribution of the intensity of the scattered light. A
helpful diagram is shown in Figure 3. Laser diffraction reports a volume distribution.
 Figure 3 Laser diffraction principle of working
3.2.1.4 Electrozone Sensing (Coulter Counter)
A Coulter counter is an apparatus for counting and sizing particles suspended in an electrolyte.
It is a suitable process for particles, cells, bacteria and viruses. A schematic drawing of a Coulter
Counter is shown in Figure 4.
A typical Coulter counter has one or more microchannels that separate two chambers
containing the electrolyte solution. The suspended particle-fluid is drawn through a
microchannel in one of the chambers. Each particle causes a brief change to the electrical
resistance of the liquid. The counter detects these changes in electrical resistance. By
monitoring these pulses in electric current, the number of particles for a given volume of fluid,
can be counted. The amplitude of the electric current is related to the particle size; thus, a
particle size distribution is also measured. The result can be correlated to mobility, surface
charge, and concentration of the particles.

Figure 4 Schematic of a Coulter Counter

3.2.1.5 Sedimentation techniques
This method depends on the terminal settling velocity of different-sized particles in a fluid (see
Chapter 4). Using the sedimentation technique, it is assumed that laminar flow conditions
prevail, and a Stokes diameter is determined. Both incremental changes in time, as well as
cumulative depositions, can be determined.
Sedimentation techniques give a weight distribution.
3.3 Conversion between distributions
For a specific sample, the frequency distribution will look different depending on whether the
volume or mass, surface area or number distribution is measured as shown in Figure 8.

45

f N(d) number
40

35 f s (d) Surface area

30
Frequency (%)

25

20 f v(d) Volume

15

10

0
0 50 100 150 200 250 300 350 400
Particle diam eter

Figure 8 Number, surface area and volume distribution for data in Table 4.

If the number distribution is known, a volume distribution can be calculated provided, the shape
factor for the particles is constant, regardless of size. This is best illustrated using an example

Example 3.2
Size Number of particles Number distribution
1 1000 1000/1554 = 64.4
2 500 500/1554 = 32.2
3 50 50/1554 = 3.2
4 4 4/1554 = 0.3
 =1554

A particle with a characteristic length has a volume of kvd3, where kv is a shape factor.

Size Number of particles Volume of particles

1 1000 Kv(1)3x1000 = kv1000
2 500 Kv(2)3x 500 = kv4000
3 50 Kv(3)3x 50 = kv1350
4 4 Kv(4)3x 4 = kv256
 =1554 = kv6606
 Size Volume fraction
1 kv1000/kv6606 = 15.1
2 kv/4000/kv6606 = 60.6
3 kv1350/kv6606 = 20.4
4 kv256/kv6606 = 3.9

Assuming that the density of the particles across all size ranges is the same, then the volume
fraction is equal to the mass fraction.
Similar calculations can be done for conversion from a number to a surface area distribution,
or from a volume to a number or a surface area distribution.
Doing conversions one should be careful, as a number distribution will be magnified in the
volume distribution (e.g. a 2% error in number distribution will lead to a 6% error in volume
distribution). For this reason, conversions between distributions should be avoided wherever
possible.

3.4 Population of particles

Particle size distribution can also be represented with equivalent diameters. The purpose of a
mean is to present a group of individual particles in a concise order to obtain an understanding
of the group.
From the frequency curve, one can determine the mode, median and average value.
The mode is the most frequently occurring size in the sample.
The median is where 50% of the particles are bigger, and 50% is smaller than this value. Using
the cumulative frequency distribution, it is the size corresponding to 50% cumulative size.
The mean size can be defined in different ways. This will depend on the measurement
technique and the application of the value.
The mass mean particle size is calculated as dm = (fidi)/(fi). …9

3.4.1 Equivalent means

Consider a sample with n particles of different sizes. An equivalent diameter can be defined as
the diameter of a particle such that the properties required (i.e. volume, surface area, surface
area to volume ratio) of all the particles are the same as that of the original sample. The total
number of particles should also be the same as the number of particles in the original sample.
Hence this diameter retains both the property as well as the total number of particles, as shown
in figure 7.
 Original Sample

Equivalent Sample

Figure 7. A pictorial description of equivalent diameters

3.4.2 Equivalent volume diameter

Number distribution
The volume of a particle with size di is kdi3, where k is a volume shape factor. If there are ni
particles of this size, the volume of this size is kdi3ni. The total volume of all the particles is then

VT   kd i3 ni ….10

Let dv be the diameter of the equivalent particle. The volume of this particle is kdv3 – (note: we
assume the same shape factor for the equivalence). The total volume of all the equivalent
particles is then:

VT   kd v3 ni …11

The total volume as given by equations 10 and 11 should be the same for equivalence, thus

dv  3
n d i i
3

…12
n i
Mass (volume) distribution
In size fraction i, the total mass of the particles will be the volume of one particle with size di,
multiplied by the total number of particles in that size range, i.e.

mi = fimT= s ni k di3 ...13

𝑛 = …14

From equation 12 and 14

∑
𝑑 =
∑

and because the total mass (mT), the density of the particles (s) and the shape factor (k) are
constant, they can be removed from the summation. This gives

∑
𝑑 = . ..15
∑

Similar deductions can be made for the equivalent surface area diameter as well as the Sauter
mean diameter, as shown in Table 5.
Table 5. Equivalent diameters for population of particles.
Number Mass/Volume
∑𝑛 𝑑 1
= =
∑𝑛 𝑑 𝑓
dsv ∑ 𝑑
(Sauter)

∑𝑛 𝑑
= ∑𝑓 𝑑
dSA ∑𝑛 =
Surface Area ∑𝑓
𝑑

dV ∑𝑛 𝑑 1
= =
Volume ∑𝑛 ∑𝑓
𝑑
3.5 Bulk properties of particles
3.5.1 Porosity
The porosity of a bed of particles is defined as the open volume / total volume of the bed, i.e.
the volume that is not occupied by solids.

open volume (Vopen ) Vtot  V particles

  …16
total volume (Vtot ) Vtot

Vpart = masspart / solids density (s)

Vtot = masspart / bulk density (B)

B
   1 …17
s

3.5.2 Surface area

The specific surface area of the particles, S is given by:

S = AP/VP (m2/m3 particles) …18

and using equation 3, this is
6 6 (m2/m3 particles) …19
S 
d sv  fi 
1/ 
 d 

The particle area per unit bed volume

SB = S (1-) (m2/m3 bed) …20

6(1   )
or SB  (m2/m3 bed) …21
d sv

The particle area per unit mass is sometimes important. If the density of the particles is s, the
surface area per unit mass of particles is
6 6
SW   (m2/kg particles) …22
 s d sv  1 / fi 
s

 d i 