The University of Hong Kong

Department of Civil Engineering

(CIVL2106) Soil Mechanics

Laboratory Report
Particle Size Distribution

Name: Chui Wai Hang Thomas

UID: 3035343035
Group: M
1. Sieving Analysis

Objectives

- To determine the particle size distribution of a coarse-grained soil.

Background Theory

Soil is comprised of a combination of many course and fine particles of varying sizes. It is
very important to understand the particle composition of a soil sample especially in the field
of civil engineering because the particle grading can heavily influence the soil’s strength and
behaviour under loading. For instance, a well graded soil, i.e. a soil with a large range of
particle sizes with little gaps in the grading curve, has in general a higher strength and is less
likely to deform under loading when compared to a soil that is poorly graded.

Sieve analysis is a very traditional and simple technique used to analyse and classify soil by
its grain size. It is done through the process of passing the soil sample through a series of
sieves with decreasing sieve sizes. With this method, the largest soil particles will be trapped
in the upper layers of the sieves and the smallest particles will be trapped in the lower layers.
And thus, by measuring the mass retained on each sieve, percentage of mass finer than the
sieve size can be calculated. And thus, the corresponding particle grading can be plotted to a
graph. By using the graph, the soil sample can be classified using the Unified Soil
Classification System based on its characteristic diameter.

Apparatus

- B.S. Sieves
- sieve shake
- drying oven
- balance
- riffle box
- brush
- pestle
- mortar
Procedure (from lab sheet)

1. Obtain a test sample of around 150 g by riffle box.

2. Weigh the sample to within 0.1 g.

3. Clean the sieves and put the sample through the following series of sieves: 2.0 mm,
1.18 mm, 0.6 mm, 0.425 mm, 0.3 mm, 0.212 mm, 0.15 mm, 0.063 mm

4. Put the sieves on the shake for vibrating the sample for 8 minutes.

5. Collect the fraction retained on each sieve and weigh.

6. Calculate the percentage finer than the sieve and plot the particle size distribution
curve.

Results

Original Mass: 144.2g

Sieve Size (mm) Mass Retained (g) Mass Passing (g) Percent Finer (%)
2.0 36.7 106.9 74.4
1.18 31.4 75.5 52.6
0.6 32.7 42.8 29.8
0.425 11.2 31.6 22.0
0.3 6.5 25.1 17.5
0.212 6.4 18.7 13.0
0.15 3.8 14.9 10.4
0.063 8.3 6.6 4.6
<0.063 6.6 / /
 Characteristic Particle Size Particle Size (mm)
D10 0.15
D30 0.60
D50 1.10
D60 1.43

From these characteristic particle sizes, the following coefficients can be determined.

Coefficient of Uniformity

𝐷%& 1.43
𝐶" = = = 9.53
𝐷'& 0.15

Coefficient of Curvature

(𝐷0& )2 0.602
𝐶" = = = 1.68
𝐷'& ∗ 𝐷%& 0.15 ∗ 1.43
Discussion

From basis sieve analysis, it is possible to categorise the soil particles into their respective
sizes. And based on the grading curve, it is possible to classify the soil by using the Unified
Soil Classification System (USCS).

According to the Unified Soil Classification System, there are two group symbols for
classifying a soil. They are:

Primary Symbols Secondary Symbols

G = Gravel W = Well Graded
S = Sand P = Poorly Graded
M = Silt M = Silty
C = Clay C = Clayey
O = Organic Clay L = Low Plasticity
PT = Peat H = High Plasticity

The first step of classifying the soil is by determining whether it is fine grained or coarse
grained. This can be done by checking whether 50% of the particles are larger or smaller than
0.075mm. Since D50 of this soil sample is 1.10mm, the soil is coarse grained.

After determining that the soil is coarse grained, the soil can be categorised as either a gravel
or sand by considering whether 50% of the particles are larger or smaller than 4.75mm. Once
again, since D50 of this soil sample is 1.10mm, the soil is considered as sand.

To proceed with the classification, the amount of fine content of the soil must be determined.
This is because soil behaviour is heavily influenced by the amount of fine content. Fine
content is defined by particles smaller than 0.075mm. By reading off the grading curve, it can
be seen that around 5% of the soil sample is smaller than 0.075mm in diameter. This means
that the coarse-grained soil is clean with very limited fine content.

The final step of the classification is to determine whether the soil is well graded or poorly
graded. There are two main grading characteristics to be considered are whether it is uniform
and gap-graded. A well graded soil should not have a uniform particle size as a larger
variation of particle sizes can help increase the strength of the soil. In addition, there should
not be significant gaps in the particle grading as it will also heavily affect the strength of the
soil. These two characteristics can be characterised by the coefficient of uniformity and
curvature respectively. The coefficient of uniformity is calculated to be 9.53, whereas the
coefficient of curvature is 1.68. According to the Unified Soil Classification System, a well
graded sand should have a coefficient of uniformity of above 6 and a coefficient of curvature
between 1 and 3. This soil sample satisfy the two criterion and thus is well graded.

The overall classification of the soil sample is: SW

Precautions

- Make sure that the soil is properly dried so that the soil particles do not clump
together.

- Ensure that there is enough shaking time for the sand particles to properly travel
through the sieve layers.

- Make sure that all the soil particles are brushed off from the sieve.

- Be careful when brushing the soil particles to not lose any soil mass.

Assumptions

- In sieve analysis, it is assumed that the particles are perfectly rounded and hence they
can perfectly pass through the sieve at any angle.

- It is assumed that the soil particles are dry and are not clumped together otherwise it
will affect how the particles travel through the sieves.

- The time of shaking is sufficient for all the soil particles to pass through their
respective sieve sizes.

Sources of Error

- While it is assumed that the soil particles are perfectly round, in reality they are
angled, which means that they cannot be perfectly categorised by a single diameter
and they may get stuck on different sieve sizes due to its irregular shape.

- There may be loss of soil particles as some particles may get stuck on the sieve. This
is reflected by the final mass of all soil particles being less than the original mass,
which is 143.6 and 144.2 respectively. This shows that 0.4% loss in mass in the
sample.

- There may be clumping of the soil particles due to moisture content and will affect its
ability to travel through the sieves and thus introducing some random error in the
results.

- There may be wearing down of the sieves especially for the smaller sizes and thus
there may be holes which will cause uncertainty in the measured particle sizes.
Pros and Cons of the Method

Pros

- The experiment is very simple in nature as it is a simple mechanical procedure that

helps separate the soil sample by its particle size. This means that it is very easy to
conduct with very little areas for error.

- Since there is a very direct measurement method, the associated errors are very minor
and thus the data is very accurate and precise.

Cons

- The experiment only provides information of soil particles up to a precision of

0.063mm. This means that the method cannot provide information on the composition
of fine particles of the soil sample and thus it cannot distinguish silt from clay. When
analysing a soil sample composed of mainly fine particles, another method should be
used.

Conclusion

Through sieve analysis, a particle size distribution curve is plotted and using the Unified Soil
Classification System, the soil sample is a well graded sand, SW.
2. Sedimentation Test
Objectives

- To determine the particle size distribution of a fine-grained soil

Background Theory

Sedimentation test is another test to determine the grading of soil particles. However, it is
more effective towards measuring the sizes of finer particles compared to sieve analysis. The
test is based on the phenomenon that settling velocity varies with particle sizes, as larger
particles will in general settle faster than smaller particles. Therefore, by collecting samples
at specific heights and times, it is possible to determine the grading of the soil sample. The
relationship between particle size and settling time can be derived using Stokes’ Law, which
governs the viscous drag forces acting on the soil particles as it settles.

𝐹7 = 3𝜋𝐷𝜂𝑣

At terminal velocity, the drag force is equalled to its weight minus the buoyant force:

𝜋𝐷0 𝜋𝐷0
3𝜋𝐷𝜂𝑣 = 𝐺< 𝜌> 𝑔 − 𝜌> 𝑔
6 6

Rearranging yields:

𝐷2 𝑔𝜌> (𝐺< − 1)
𝑣=
18𝜂

The above equation gives the relationship between settling velocity and the diameter of the
soil particle. As can be seen, a larger particle is expected to have a faster settling velocity.
Finally, substituting v = H/t, where H is the sampling depth and t is the time of measurement.

18𝜂𝐻
𝐷=A
𝑔𝜌> (𝐺< − 1)𝑡

Where:

D = corresponding particle size, in mm

𝜂 = dynamic viscosity of water at test temperature, 0.0091g/cm*s
H = sampling depth, in cm
g = acceleration due to gravity, 9.81m/s2
𝜌> = density of water, 1.0g/cm3
Gs = specific gravity of soil particle, 2.65
t = sampling time, in seconds

By using the above equation, it is then possible to determine the particle size grading curve
from measurements of varying sampling times.

Apparatus

- Sampling Pipette
- Soil Mixer
- Stop Watch
- Balance
- Oven
- Temperature Controlled Water Bath
- 500ml Sedimentation Tube
- Dispersion Cup
- Sodium Pyrophosphate (SP)

Procedures (from lab sheet)

1. Weigh out about 10 g of oven-dried powder soil. Put in a dispersion cup, add 200 ml
distilled water and 25 ml Sodium Pyrophosphate (Na4P2O7·10H2O) and stir the
mixture for 10 minutes by soil mixer.

2. Transfer the suspension to the sedimentation tube and wash the dispersion cup with
water. Make up the tube to 500 ml with water (caution should be paid). Leave the
tube in the water bath for 10 minutes. Shaking the tube and the contents by inverting
several times and then replace in the water bath and start timing by stop-watch.

3. Obtain a sample of 10 ml by the pipette from the depth of 10 cm below the surface at
216 seconds after the shaking. Weigh the empty dried weighing bottle and transfer the
sample to it, wash down any suspension left on the inner walls of the pipette. Dry off
the water by placing the bottle in the oven. Weigh the dried sample.

4. Repeat step (3) and take samples at times of 14 minutes, 40 minutes and 6 hours after
shaking.
Results

Mass of Dried
Bottle No. Time (s) Mass of Bottle (g)
Sample & Bottle (g)
1 216 16.7725 16.9865
2 840 16.751 16.9572
3 2400 16.9625 17.1527
4 21600 16.6185 16.7535

To determine the mass of soil in the sample, the total mass of sample and bottle is subtracted
by the mass of bottle and the mass of Sodium Pyrophosphate. This is because the SP added to
the solution for the sedimentation test has not been evaporated in the oven and thus is
accumulated in the dried mass. The mass of SP in the 10ml soil sample is 0.018g.

The particle size, i.e. diameter, is determined through the equation derived before by
substituting the sampling times:

18𝜂𝐻
𝐷=A
𝑔𝜌> (𝐺< − 1)𝑡

Lastly, the percentage finer is given by the following equation:

𝑤/𝑣
𝑁= ∗ 100%
𝑊/𝑉

Where,
w = net mass of dried sample (the mass of SP should be excluded)
v = volume of the collected sample, 10ml
W = total mass of soil sample used, 10g
V = total volume of original suspension, 500ml

Mass of Dried Soil

Bottle No. Diameter (mm) Percentage Finer (%)
Sample (g)

1 0.196 0.02164 98
2 0.1882 0.01098 94.1
3 0.1722 0.00649 86.1
4 0.117 0.00216 58.5
 Particle Size Distribution Curve from Sedimentation Test
100

80
Percentage Finer (%)

60

40

20

0
0 0.005 0.01 0.015 0.02 0.025
Particle Size (mm)

Discussions

After plotting the particle size distribution curve, the soil can be classified using the United
Soil Classification System in a similar manner to the case of using sieve analysis. However,
one point to note is that the range of particle sizes determined in this sedimentation test is
0.022mm to 0.002mm compared to that of sieve analysis, which is 0.063mm to 2mm. This
comparison shows that sedimentation test is much more suitable and precise at measuring
smaller particle sizes, whereas sieve analysis is more suitable for measuring larger particle
sizes.

In this case, it is actually impossible to accurately determine D50 by interpolating the graph as
the minimum percentage finer is at 58.5%. However, by considering the fact that over 58.5%
of the soil sample is finer than 0.002mm, it can be concluded using the United Soil
Classification System that the sample is a clay.

To proceed with the classification system, the liquid limit of the soil sample must be
determined to determine whether it is high plasticity or low plasticity. In general, a liquid
limit of over 50% corresponds to a high plasticity and vice versa. However, since no
plasticity tests were conducted on the soil sample, there is insufficient data on the plasticity
of soil and thus the sample can only be categorised as clay.
Precautions

- The soil sample should be given ample amount of time to be completely mixed.

- Make sure that the soil sample is completely dry before the experiment to have an
accurate measurement of the total dry soil mass.

- The extraction process must be done slowly and carefully in order to not disrupt the
sedimentation process and also the ensure that the extracted volume is exactly 10ml.

- Rinse and clean the pipette completely the ensure that there is no residue affecting the
later measurements.

- Be careful when transferring the sample to avoid any spilling.

Assumptions

- Stokes’ law assumes that the particles experiencing the drag force is rounded.

- Stokes’ law only encompasses the viscous drag forces acting on the particle and does
not consider the inertial forces acting on the particle. This assumption holds if the
flow is laminar which takes place when the particle is small and flows at a low
velocity.

- It is assumed that the particles are well mixed in the solution beforehand such that
they are evenly distributed throughout the sample and there is no clustering of
samples that may affect the settling speed.

- The extraction of solution will not cause a huge disturbance in the fluid and will not
affect the sedimentation process of the remaining soil particles.

Sources of Error

- Stokes’ Law assumes that the particles falling in the tube is round. However, this is
definitely not true for the case of soil particles as some of them might be angular and
they are in general irregular in shape. Therefore, there may be errors in the equations
used to determine the relationship between particle size and time of sampling.

- There may be clustering of the soil particles due to uneven mixing or they may be
formed during the sedimentation process. This clustering will lead to a faster
sedimentation times for the soil particles and will affect the grading results by making
 the soil sample seem to have more larger soil particles than it really has.

- There will be human errors involved with in many stages of the experiment such as
the measurements of volumes, weights, and the extraction and transferring of solution
and so on. This will introduce random uncertainty in the results.

- There is insufficient data determined in the experiment. For instance, there is a very
limited range of data as there is no information on the particle sizes below the 58.5%
finer threshold as stated before. This can be remedied by continuing the experiment
for a longer duration as a longer sedimentation time can help measure smaller particle
sizes. In addition, there are only 4 data points, which is very insufficient in terms of
generating a smooth grading curve.

Pros and Cons of the Method

Pros

- This method is much more suitable for measuring smaller particle sizes when
compared to the sieve analysis, due to the sedimentation process being much slower
and more precise for small particles. Therefore, the sedimentation test can accurately
categorise soil samples with small particle sizes.

- Due to the delicate nature of the test, not a large amount of soil sample is needed, as
only 10g of soil is used in this test compared to that of 144.2g of the sieving analysis.
This means that the test is less demanding on the soil sample needed and may be more
applicable in situations where samples are hard to collect.

Cons

- The duration of experiment is significantly longer than the sieve analysis as

sedimentation is a very slow process. For instance, the final sampling time is 6 hours.
The relationship between particle size and sampling time is inverse square root
related. This means that to measure even finer particle sizes, the sampling time will be
increased by an impractically long amount of time and so there is a practical limit to
how small the particles can be measured by the sedimentation test.

- The test is also very ineffective at measuring large particle sizes as they tend to settle
very quickly and any measurement from the sedimentation test will be inaccurate.

- Since the sedimentation test is an indirect measurement of the particle sizes, there are
many assumptions to be made regarding the fluid flow and also the forces acting on
the particles. For instance, the fluid is assumed to be static and the particles are
 assumed to be rounded and experience no inertial forces. However, this is not true in
reality and they will introduce uncertainties in the measurement.

Conclusion

With the sedimentation test, it can be found that the soil sample is clay, but further analysis
on its plasticity must be done to completely classify the soil based on the Unified Soil
Classification System.