1.

PROPERTIES OF PARTICULATE SOLIDS

1.1. Introduction

In a great number of chemical industries, we are in presence of particulate

solids, whether as raw materials or products. For example, calcium
carbonate and clay in the cement industry, phosphate rock in the fertilizer
industry, soda ash, quartz and limestone in the glass industry etc…. That
is why in the present chapter, we will be concerned in understanding the
main properties that characterize solid particles:
These are the density, the shape and the size of these particles.

1.2. Density

The density of a solid material is a characteristic of this material. This is

termed “The true density” or “Particle density, p”. If now the material
is crushed, then, according to the size and shape of the crushed product,
pores or voids will appear between the particles. The total volume of the
crushed solids is termed “The Bulk volume, Vb”. This is the sum of the
true volume of solids (or Solid volume, Vp) and that of the pores Vpores.
The “porosity, ” is defined by:
Vpores Vb  Vp Vp
   1 (1.1)
Vb Vb Vb

Now, the bulk density b is defined as b = m/Vb where m is the mass of

the particles and the particle density is p = m/Vp.
Substituting in Eq. (1.1), we get:

  1 b (1.2)
p

Table (1.1) shows the particle and the bulk densities of some materials.

Table (1.1)
Material Particle density, kg/m³ Bulk density, kg/m³
Cement 3100 1300
Limestone 2400 1500
Dry sand 2600 1700
Gravel 2600 1600
Coal ( Anthracite ) 1600 700
Coke 1300 550

1
1.3. Particle Shape

In order to attribute a quantitative measure to the shape of a particle, we

first consider a spherical particle of diameter Dp.
3
Its surface area is A = D²p and its volume is V = D p/6.
The Specific Area per unit volume is s = A/V
Hence for a spherical particle:
6
s (1.3)
Dp
The same is true for a cubic particle of edge length (a)
Where A = 6a² and V = a³ and hence s = 6 / a.
If the particle is neither spherical nor cubic, then it will be characterized
by a certain equivalent diameter Dp. In this case, the value of s cannot be
calculated from Eq. (1.3) but rather from a modified form:
6
s (1.4)
D p

Where Ф is a correction term that takes in account the shape of the

particle and is termed “The Sphericity, ”. It iss equal to unity for
spherical or cubic particles. Table (1.2) gives the values of Ф for some
solids.
Table (1.2)
Material Sphericity Ф
Spherical beads 1.00
Round sand 0.83
Gravel 0.75
Coal dust 0.73
Angular sand 0.59 – 0.71
Crushed glass 0.65
Mica flakes 0.30

1.4 Particle Size

The “Equivalent diameter” of a particle is the main dimension if the
particle is equi-dimensional (Diameter in a sphere, Edge length in a cube
etc…). For irregular particles, it is usually taken as the 2 nd largest
dimension. Particles are termed coarse if their diameters are in the range
of a few centimeters or millimeters, they are considered fine when their
diameters are in the range of tens or hundreds of microns and very fine if
the particle size is in the range of a few microns or less.

2
A crushed product will usually consist of coarse to fine particles while a
ground product will consist only of fine particles.

1.4.1 Standard screen series (or standard sieves)

Standard screens are used to measure the size and size distribution of
particles in the size range 80 mm to 40 µm. They are made of woven wire
screens, the mesh and dimensions of which are carefully standardized.
The openings are square. Each screen is identified by the “mesh size”
which represents the number of openings per linear inch. The actual
openings are smaller than the mesh size because of the thickness of the
wires. Table (1.3) shows the openings of the most common screens:
Table (1.3)
Mesh size Opening, mm
4 4.760
6 3.327
10 1.651
20 0.833
35 0.417
65 0.208
100 0.147
150 0.104
200 0.074 Fig.(1.1) Set of sieves
325 0.044

1.4.1 Screen analysis

To sort a sample of particles of different sizes, one uses a set of screens
arranged vertically in a decreasing order of opening diameter from top to
bottom. The upper screen is usually chosen in such a way that no particles
will be left on its surface on screening the mixture. Also the bottom
screen is chosen such that a minimum amount of fine powder will drop in
the pan placed beneath the set of screens. (Fig. 1.1)
The mass of particles retained over the screen No. (i), numbered from
top, is mi. The total mass is M = m1 + m2 + …. + mn
The mass fraction of particles retained over this screen is:
xi = mi/M (1.5)

The symbol Dpi will denote the particle diameter corresponding to the
opening of the screen No. (i). A table showing the values of i against

3
those of Dpi is termed “A Differential analysis”. An example of such
analysis is shown in Table (1.4)
Table (1.4)
Mesh size Dpi, mm xi
Over 4 4.760 0.000
4/6 3.327 0.125
6 / 10 1.651 0.143
10 / 20 0.833 0.170
20 / 35 0.417 0.202
35 / 65 0.208 0.183
65 / 100 0.147 0.093
100 / 150 0.104 0.050
150 / 200 0.074 0.034
Pan ------- -------
Another analysis of interest is the “Cumulative Analysis” showing the
cumulative fraction that would rest over a certain screen if all the screens
above this screen were removed. It then gives the fraction of particles that
would be retained on any screen (ci). Table (1.5) shows the cumulative
analysis corresponding to the data of Table (1.4)
It is obvious that :
i
i   x k (1.6)
k 1

Table (1.5)

Mesh size Dpi, mm ci

4 4.76 0
6 3.327 0.125
10 1.651 0.268
20 0.833 0.438
35 0.417 0.640
65 0.208 0.823
100 0.147 0.916
150 0.104 0.966
200 0.074 1.000
Pan ------- 1.000

Figures. (1.2) and (1.3) show the curves corresponding to the differential
analysis and the cumulative analysis respectively

4
0.25

xi
0.2

0.15

0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Dp mm
Fig.(1.2) Differential analysis

1.2

?i
1

0.8

0.6

0.4

0.2

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Dp mm
Fig. (1.3) Cumulative analysis

5
1.4.3 Specific surface of a mixture ( Aw )
This is defined as the total surface area of the particles per unit mass and
can be obtained for spherical particles of equal diameter Dp by:
A A 6
Aw   
M  p .V  p .D p

If the particles are of irregular shape , then:

6
Aw  (1.7)
. p .D p
(
If now the differential screen analysis of the mixture is known, then the
total surface area of the particles resting over the ith screen is:
6 M.x i
Ai  A w  mi  .
. p D avi

Where Davi is the average diameter between the (i-1)th and the ith screens.
The specific surface area is therefore:
6 n xi
Aw  . (1.8)
. p i 1 D pav

If the cumulative analysis is used, the corresponding formula is:

6 1
dc
Aw  . (1.9)
. p 0 D p

This requires plotting (1/Dp) vs. c and performing graphical integration.

1.4.4 Average particle size

This may be estimated in different ways. One of the mostly used
diameters is “The mean volume-surface diameter, Dvs” defined as the
diameter of uniform particles that would have the same specific surface
area of the mixture. Hence:
6
D vs  (1.10)
. p .A w

Another mean diameter is the weighed mean diameter simply obtained

using the following formula:

6
 1
D m = ∫ Dp. dφi (1.11)
0

On the other hand, the median diameter is that particle size corresponding
to a cumulative fraction φi = 0.5

Example 1.1
Estimate the specific surface area of the mixture whose screen analysis is
shown in Tables (1.4) and (1.5). Take the particle density 2600 kg/m3 and
the sphericity = 0.67

Solution:
1- By differential screen analysis:

Dp, mm Dpav, mm xi xi/Dpav, m-1

4.76 - 0 0
3.327 4.0435 0.125 30.91381
1.651 2.489 0.143 57.45279
0.833 1.24 0.17 136.876
0.417 0.625 0.202 323.2
0.208 0.3125 0.183 585.6
0.147 0.1775 0.093 523.9437
0.104 0.1255 0.05 398.4064
0.074 0.089 0.034 382.0225
 2438.415

Hence Aw = 6(2438.415) = 8.44 m2/ kg

0.672600
2- By cumulative analysis:

Dp, mm c 1/Dp, m-1

4.76 0 210.08
3.327 0.125 300.57
1.651 0.268 605.69
0.833 0.438 1200.5
0.417 0.64 2398.1
0.208 0.823 4807.7
0.147 0.916 6802.7
0.104 0.966 9615.4
0.074 1 13514

Performing graphical integration using the trapezoidal rule, one gets:

7
Area under the 1/Dp - c = 2616 m-1, see Fig. (1.3)
Hence Aw = 6(2616) = 9.055 m2/ kg

0.672600
Example 1.2

A feed of particles having the screen analysis given in Table (1.4) was
screened over a 35 mesh screen. Estimate the specific surface area of both
the overflow and the underflow to the screen.

Solution:

In order to obtain the screen analysis of, say the overflow to the screen
we first get the value of ci corresponding to the screen opening of 0.417
mm. According to Table (1.5), this is 0.64.
Now, divide the values of ci for the overscreen particles by 0.64. One
obtains the following table:

Dp, mm ci (modified)

4.76 0.000
3.327 0.195
1.651 0.419
0.833 0.684
0.417 1.000

Again, graphical integration can be performed to give an area of 959 m-1

This gives a specific surface area of:

Aw = 6 (959) = 3.32 m2/ kg

0.672600

To get the specific surface area of the underflow from the screen, we
perform a total area balance.

The fraction of mass retained above the screen is 0.64 and that passing
through the screen is 1 - 0.64 = 0.36.
Hence, 1  9.055 = 0.64  3.32 + 0.36 (Awunderflow), from which, the
specific surface area of the underflow from the screen is:
Awunderflow = 19.25 m2/ kg

8
1.4.5 Some special distributions
Some particle agglomerates can follow some particular distributions
defined by some analytical relation between the cumulative fraction i
and the particle size Dp. This is especially true in case of fine powders.
Some of these distributions are given in what follows:
(a) The Gaudin distribution:
This is one of the earliest distributions. Its characteristic equation is:
1 – φi = (Dp/Dmax)m (1.12)

The value of m usually varies from 1.5 to 2.5, although values out of this
range have often been reported.
Some important characteristics of this distribution can be deduced:
 If we set φi = 1, we get Dp = 0. This means that the smallest
particle is assumed to have an almost zero size.
 This distribution verifies the condition: φi = 0 when get Dp = Dmax
 The surface mean volume diameter can be prove to equal:
Dvs = Dmax.(n – 1)/n (1.13)
 The mean diameter equals:
Dm = Dmax. n /(n + 1) (1.14)

 To test whether a cumulative distribution follows the Gaudin

model, ln(1 – φi) is drawn against ln Dp. A straight line should be
obtained of slope = m. The value of Dmax can then be obtained by
setting φi = 0 in the equation of this line.

(b) The Rosin – Rammler distribution:

It usually applies for narrow distributions and is defined by the following
equation:
φi = exp – (Dp/Dm) n (1.15)
The value of the parameter n usually ranges from 0.5 to 2. The greater the
value of n, the broader will be the distribution.
In the above equation, Dm represents a particle diameter corresponding to
a value of φi = e-1 = 0.368
This distribution possesses the following characteristics:
 If we set φi = 1, we get Dp = 0. This means that the smallest
particle is assumed to have an almost zero size.
 If we set φi = 0, we get Dp = ∞. That is why this model cannot
predict the whole distribution and is usually valid down to φi = 0.2

9
  If ln |ln φi| is plotted against Dp, a straight line should be obtained
of slope = n. The value of Dm can then be obtained as the abscissa
corresponding to φi = e-1 = 0.368

(c) The normal distribution

This distribution is commonly encountered in diverse fields. In the
present case, the cumulative function is given by:
Dp ( Dp Dm )2
1 
i  .e 2. 2
.dD p (1.16)
. 2 0

In this equation, σ is the standard deviation of the distribution which

designates its broadness. The above integration cannot be solved
analytically and has to be performed using tables of error function.

(d) The lognormal distribution

This is the distribution that is usually followed by fine powders. Its
basic equation resembles that of the normal distribution except that ln
Dp is substituted for Dp.

Example 1.3
The following table shows the cumulative distribution of crushed
stone. Prove that these data fit the Gaudin model and find the value of
the parameter m as well as the probable maximum particle size.

Dpi, i
mm
4.76 0.01495
3.327 0.483035
1.651 0.853543
0.833 0.957251
0.417 0.987697
0.208 0.996482
0.147 0.998117
0.104 0.99899
0.074 0.999453
pan 1

Solution:
We prepare a plot of ln (1 – φi) against ln Dp. This is shown in the
following figure

11
 The equation of this line was found to be:
ln(1 – φi) = 1.76. ln D – 2.798
Thus, the parameter m = 1.76
To get the expected maximum size, we set: φi = 0, we get:
ln D = 2.798 / 1.76 = 1.589, from which: Dmax = 4.9 mm

0
-3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2

-1 ln D mm

-2

-3
ln(1 -φ)

-4

-5

-6

-7

-8

Fig.(1.3): Gaudin plot

1.5 Angle of repose

When a mass of particulate solids is spilled over a horizontal surface, it

tends to form a cone. The angle of repose is the angle between the slant
edge of the cone and the surface. Fig. (1.4).

The value of the angle of repose depends on

several factors among which are the particle
size and shape and the moisture content of the
solids. r
Fig. (1.4) Angle of Repose
1.6 Angle of friction
If a particle is placed on a rough surface, a force of friction will develop
between the particle and the surface that acts in a direction opposed to
any potential motion of the particle on this surface. The force of reaction
due to the pressure exerted by the particle on the surface is perpendicular
to that surface. The angle between the reaction normal to the surface and

11
the resultant force of the reaction and the force of friction, as the particle
moves (or is about to slide) on this surface is termed the angle of friction
(). Fig. (1.5).

The ratio between the force of friction as the Normal reaction

particle moves (or is about to slide) on the
surface and the normal reaction is termed the

coefficient of friction between the solid and the
surface ().
Force of Friction
Fig. (1.5) Angle of Friction
It is clear that  = tan  (1.11)

We have to differentiate between two types of angles of friction: The

angle of friction (m) which is the angle of friction between the surfaces
of two particles, and the angle of friction of particles and the surface of
their container (simply written, ).

For a mass of totally homogeneous particles (in size and packing), the
angle of internal friction would be equal to the angle of repose. In general
since the particles on the outer surface are usually loosely packed, r<m.
Table (1.6) gives the values of these angles for some common materials.

Table (1.6)
Angle of Angle of
Angle of
Material Internal Friction on
Repose
Friction Steel
Dry Clay 40 42 32-37
Gravel 45 50 40-45
Limestone Lumps 35 40 25-30
Dry Sand 45 50 35-39
Dry Cement 50 53 28-33
Gypsum Lumps 40 44 35-38
Coke 38-40 45-48 40-45

12