Advances in Embankment Stability

Chapter 1

1.0 Background Information

Embankment stability is an engineering topic that has been characterized by dramatic progress from the time Proctor proposed the fundamental principles of soil compaction in 1933. In the process of constructing earth dams in California, R.R. proctor postulated that the main goal of compaction is to achieve the highest dry density of the compacted soil. Embankments are normally constructed to elevate the ground level to some height above the existing level so as to support engineering structures such as buildings, dams, roads or railways. The durability and stability of such embankments is largely affected by the shear parameters of the compacted soil, which again depend on how well the soil has been compacted.

Engineers worldwide have been presented with catastrophic cases of embankment failure, for example, the Brumadinho dam disaster in Brazil, the Patel dam failure in Kenya, and the Swar Chaung dam disaster in Myanmar. These have often led to tremendous loss of life and property, especially in highly populated areas. For this reason, scholars have taken to conducting studies on compaction in a bid to prevent similar disasters in the future. Among them are Shima and Imagawa (1980), Escario and Saez (1986), Maatouk et al (1995), and Wheeler and Sivakumar (1995).

Most of these studies dwelled much on the strength properties of unsaturated compacted soils, such as friction angle and cohesion, as opposed to the mechanical behavior of the same. This study, therefore, seeks to investigate the mechanical behavior of compacted soil. It is envisaged that the research findings will be useful in the revision of the earthwork guidelines for the construction of embankments for roads, railways, buildings, and dams. Better guidelines will lead to the reduction of embankment failure cases.

1.1 Context

The research project will first cover the introduction section which provides information on the importance of embankment compaction of the soil. It also looks at the past cases where embankments have failed causing tremendous damage to property and loss of lives. Chapter two is the literature review section which provides available literature about the mechanical behavior of soils, the procedure for soil compaction and the effect of compaction energy on engineering properties.

The methodology chapter outlines the experiments that were used to conduct the study. It involves the application of the proctor test by using a box to test collapse of the open side. A weight was placed on a tray on one side until the embankment moves down. The results and discussion section displays the findings of the entire research project with explanation using figures, tables and written texts. The last chapter provides a summary of the conclusions of the study that have been arrived at from the experiments. It also looks into the recommendations on items that were not covered in this study and can be used for further research studies. At the end of this document is the Appendix page which provides details on available experimental data and results obtained during the research process. The box used in the experimentation process is made from natural wood which is meant to withstand pressure by measuring 40 x 25 x 22 cc. The sliding part of the box can be pulled easily after the process of soil compaction. The aim is to copy the natural environment in which embankment soils are exposed to.

1.2 Justification of study

The purpose of this study was to investigate the mechanical behaviour of compacted soil and its effect on the compaction of embankment stability. In the construction of dams, buildings and roads the compaction of embankment soils is a key process that determines the stability of the whole project. There have been numerous cases where embankments failed leading to collapse of dams and roads. This study seeks to find out some of the effects that are brought about by proper compaction of embankment soils to increase stability of such infrastructure. Analysing the mechanical properties of soil such as the soil type and water content among others enables the experimentation process to be easily analyzed.

1.3 Research Objectives

General objective

The general objective of this study is to investigate the mechanical behavior of compacted soils

Specific Objectives

The specific objectives of this study are:

1. To identify basic procedures of compaction for embankment soils

2. To show the effect of depth and compaction energy on the engineering properties of embankment soils

3. To investigate the behavior of embankment soils during the process of compaction and after the process of compaction

4. To identify the limits beyond which it would be insignificant to further compact embankment soils

1.4 Scope of work

The first step of the research process involved gathering adequate information through collection of data from various sources. The different resources include recent journals, books, educational magazines and authentic informational websites. Such information was obtained from past work of scholars like Escario and Saez (1986) who discussed about compaction effect on embankment soils. This was followed by the process of experimentation. The main part of the experiment will apply the proctor test concept. The proctor test provided information about compaction characteristics of various soils under moisture content variations. Compaction occurs when the OMC becomes denser hence attaining maximum dry density through the elimination of air voids.

Dig deeper into FGM in the UK: Legal Reforms with our selection of articles.

The degree of compaction is usually measured depending on its dry density. This is calculated as

= (M / V ) /(1+w) where V= soil volume, M= soil mass, w= water content

After the Proctor Compaction test, the effect of compaction on embankment stability was investigated by first compacting soil in a wooden box measuring 40 cm x 25 cm x 22 cm. The soil was then subjected to pressure, with one of the sides exposed. The mechanical behaviour of the embankment was then observed. After gathering sufficient data from the laboratory experiment, a comprehensive analysis was This was followed by a comparison with already existing research data from other scholars. The acquired data underwent various manipulations such as calculation of correlations, creation of tables and plotting graphs. The final step involved the interpretation of acquired results and drawing meaningful conclusions, finally giving recommendations for future studies.

Chapter 2

Literature review

Compaction of soil is a process that involves the rearrangement of the soil particles while pressing them close together with the use of mechanical techniques. Compaction principle was established by R.R. Proctor back in 1933 when the earth dams in California were being constructed. The main aim of compaction is to attain the highest achievable dry density ratio of the compacted soils. Compaction uses a certain percentage of water content to control the achievable dry density.

The durability and stability of embankments mainly depend on the restriction of deformation and development of shear strength of the soil. The greater the density of the soil then the higher the strength becomes. High-density results in the reduction of soil permeability hence decreased seepage of water. The construction of embankments requires compaction of the soil to increase its density.

2.1 Factors affecting compaction of embankment soils

2.1.1 Water content

Increasing water content for compaction leads to an increase of dry density up to a maximum point. Once the Maximum Dry Density (MDD) of soil is reached, an increase in water content reduces the dry density.

2.1.2 Type of soil

The soil type that is used in the compaction process determines the MDD that is required. Soils that are coarse-grained require compaction to a higher percentage of dry density compared to fine-grained soils. This means that soils which are coarse grained show a higher MDD with a reduced OMC compared to fine-grained.

2.1.3 Soil gradation

A poorly graded soil will have a low MDD. Soil that has been properly graded has various particle sizes enabling finer grains to fill the spaces located between the coarser grains, therefore, decreasing the air voids. When a certain amount of fine-grained soils is added to coarse grains the MDD increases up to a certain level and then starts to decrease when the maximum amount of fines is reached (Cfms-sols.org, 2019).

2.1.4 Compaction Energy

The amount of compaction energy that is applied during the compaction process greatly influences the MDD. The greater the compactive effort the higher the MDD and the lower the OMC. The continued increase in compactive effort after attaining the maximum limit does not increase MDD indefinitely. The MDD reduces in size with each equal increase of energy. The last stage is where the further increase has no significant effect on the increase of MDD.

2.2 Basic procedures of compaction for embankment soils

2.2.1 Selection of fill matter and borrow location

This procedure requires the survey and collection of samples to be tested. The different soils are tested for suitability as a suitable location is chosen depending on the transportation cost, suitable quantity, the suitability of soil and the expenses of practicing on the borrow area. After the determination of granular size and index characteristics, the soil can be categorized depending on IS – 1498-1970 which provides the varying suitability for soil compaction (Soil Management India, 2019). Soil types that have a high level of plasticity or those with a high ratio of shrinkage have a tendency of expanding or becoming too sticky hence not suitable for compaction. Fill materials should not include organic matter unless proper precautions are taken.

2.2.2 Selection of equipment used for compaction

i) Smooth wheel rollers

These type of machines are suitable for the process of proof rolling subgrades as well as completing the finishing of fills. They offer contact pressure ranging from about 30 to 40 t/m2 with a coverage of 100%. They are suitable for soils that lack cohesion.

ii) Pneumatic tired rollers

These are rubber rollers that contain wagons that are heavily loaded and with up to six rows of tightly spaced tires. They offer a contact pressure of about 60 to 70 t/m2 with coverage of up to 85%

iii) Sheep’s foot rollers

These rollers have hollow drums that have numerous projections known as sheep’s foot or shanks. The shanks provide an increased contact pressure of 150 to 750t/m2.

iv) Vibratory rollers

These are suitable for the compaction of granular soils. The vibration effect is produced through the rotation of eccentric weights. The vibratory equipment can be attached to rubber rollers or smooth wheel rollers. Other types of equipment include grid rollers, tampers, and vibrating plate machines (Anon, n.d.).

2.3 Soil compaction in the field

The compaction machinery can effectively compact soil to a certain depth limit. For this reason, the embankments are usually compacted in various layers known as lifts. When the thickness of a particular lift is high, then the soil located over the lift will be efficiently compacted while the one under the lift will not be compacted fully. The soil taken from a given borrow area is transferred to the field and then spread in various layers. Each layer is then compacted several times by a roller. This process is repeated to achieve the desired height of the embankment. It is advisable to keep checking the water content during the compaction process. Compaction will be ineffective if the soil is too dry while compaction will be impossible if the soil is too wet (Anon, n.d.).

2.4 Field compaction control

Problems that might arise as a result of improper compaction of soil

i) shear failure of the compacted soil as a result of low shear strength ii) failure of the foundation and support structures due to differential settlements and excess total iii) the failure of slopes located on the embankments iv) collapse of hydraulic structures due to increased permeability of embankment resulting in water loss

2.5 Effect of compaction energy on the engineering properties of embankment soils

i) Dry density

As comparative effort on the soil is increased then a gradual increase in the density of the soil will be realized. An increase in compaction energy reduces the optimum moisture content of a given soil. The reduction in moisture content is due to the decrease of void spaces available in the soil.

ii) Unconfined compressive strength -(UCS)

An increase in compaction energy also increases the UCS of the soil. The compressive strength of a given soil is determined by the content of moisture and packing density. When compaction energy is increased, moisture reduces while density increases.

iii) Swell potential

Increased compaction energy also increases the soil’s swell potential. The compaction of dry soil reduces its permeability, therefore, increasing the swell potential (Water tensions, swelling mechanisms, strength of compacted soil, 1960).

2.6 The behavior of embankment soils during the process of compaction and after the process of compaction

The behavior of embankment soils after compaction is based on the water content expressed in terms of Optimum Moisture Content (OMC) (Zwanzig, 1980). The soils can be described as compacted dry of optimum or as wet of optimum. Regardless of the similarity of the dry density, compacted soil is different on both parallel sides.

i) Soil structure

The wet side of compacted soil has a dispersed structure while the dry side has a flocculated structure. The parallel arrangement of particles on the wet side is perpendicular to the direction where stress is being applied. The orientation degree rises with an increase in water content of the soil. A rapid increase is however characterized by soils that are compacted on the wet side. The shear strain determines the structure of soils that have a greater water content than OMC. The orientation of soil particles increases with increased compaction energy.

ii) Swelling

The dry side has a high water deficiency behavior with water films that are partially developed. This behavior makes dry of optimum compacted soil to imbibe more water compared to wet of optimum soil. This characteristic of water deficiency makes the soil swell when exposed to water (Water tensions, swelling mechanisms, strength of compacted soil, 1960).

ii) Shrinkage

The soils that have been compacted on the wet side display more shrinkage compared to the soils on the dry side. This allows particles to be packed properly due to their arrangement and orientation.

ii) Permeability

The dry side is less permeable compared to the wet side. The permeability rate of the dry side is similar in all directions. On the wet side, permeability is more along the orientation of the particles than across the particles.

iii) Compressibility

When the applied stress is low, then the dry soil becomes less compressible due to the arrangement of particles. The wet soil is however more compressible given similar conditions. Wet of optimum compacted soils display high compressibility when stress is increased therefore causing great deformation. The arrangement of such soils when the structure is dispersed allows reduced resistance to deformation hence increasing compression (Kumor, Kumor and Kopka, 2017).

iv) Stress-strain relationship

Compaction of soils on the dry side has a high stress-strain relationship than soils that are wet of optimum. Soils that have been compacted dry of optimum have less deformation and settlement, therefore, displaying sudden failure. Wet of optimum compacted soils show gradual failure due to the large strains and settlements.

v) Pore water pressure

There exists zero pore water pressure for soils that have been undergone dry optimum compaction. This because of the saturation of local pockets. High pore water pressure is observed in soils with wet of optimum compaction. Such soils (wet of optimum) display reduced frictional component and stress of shear strength (Schanz, 2007).

Chapter 3

3.1 Introduction

This chapter presents the actual steps that were followed to meet the specific objectives of the study. This was majorly through laboratory experiments. The very first step in the laboratory experiments was to determine the Maximum Dry Density and Optimum Moisture Content of 3 soil samples. This was done through the Standard

3.2 Proctor Compaction Test procedure

Proctor compaction is carried out in order to understand the characteristics of various soils and their moisture content during the process of compaction. The densification of soil through the reduction of air voids is known as compaction. At optimum water content, the dry density is at its highest. A curve can be obtained from the relationship between the dry density and the water content in order to identify the optimum water content and the maximum dry density.

The laboratory requirements for the proctor test include;

Spatula Weighing balance with accuracy of 1g Dessicator Mixing tools IS sieve, 4.75mm Compaction mould, 100ml Oven Collar measuring 60mm high Large mixing pan Rammer of 2.6kg Straight edge Graduated jar Detachable base plate

3.2 Procedure followed for the proctor soil test

20kg of air dried soil was sieved through 20 mm sieve and a 4.7 mm sieve.

The percentage of the retained soil on the 20 mm and 4.75 mm sieve was calculated

The retained percentage was less than 20% therefore enabling the use of 100mm diameter standard mould .

Soil retained on the 4.75mm sieve was mixed and passed through the 4.75mm sieve in limited proportions to obtain a soil specimen of 16 to 18 kg.

The dry mould and the base plate was dried and greased lightly.

The mould was weighed with the base plate to the nearest 1 gram

Soil specimen of 18kg was taken and water was added to it to increase the water content to 8%.

Soil was kept in an airtight container for a period of 20 hours to mature. It was then mixed thoroughly and divided into 8 parts.

A collar was attached to the mould and placed on a solid base.

2.5kg of the processed soil was layered 3 times in the mould. One third of the quantity was first taken and compacted by using the rammer for about 25 blows. The blows were uniformly distributed across the entire surface.

The top surface of first layer was scratched with a spatula before the addition of the second layer which was also compacted with 25 blows of rammer as well as the third layer.

The soil used was sufficient enough to fill the mould while allowing approximately 5 mm above the top part of the mould which was struck off during collar removal.

The collar was removed and excess soil projecting over the mould was trimmed off using a straight edge.

The base plate was cleaned and weighed to the nearest gram.

The soil was removed from the mould.

Water content was determined from the samples obtained from the bottom, middle and top portions.

3% of water was added to a new portion of the processed soil and the previous steps repeated.

After the proctor compaction tests were complete, 3 sets of experiments were done. Below is a description of the general procedure of the experiments.

3.3 General procedure of the experiment

A box measuring 40cm x 25cm x 22cm was assembled in the manner shown in the figure below. The box was made of natural wood so as to be able to withstand pressure. Three sides were fixed firmly with nails, while the 4th side was made such that it could be slid into and out of place. The side is slid into place during the compaction of the soil, and then pulled out before placing heavy weights on the compacted soil. The soil was placed into the box and compacted in a similar fashion to the Proctor compaction test. Once the compaction process was completed, the test box had one of the sides removed to imitate an exposed soil embankment. A weight was then placed on a tray, which was placed on the soil. The weight on the tray was incremented in predetermined values until the embankment failed. This process was repeated for other degrees of compaction. The heavy weights were placed on the box after the soil had been compacted. The weights provide the pressure of the earth similar to that in the natural environment. When the open side collapses, then this acts as a confirmation of the inadequacy of good soil cohesion and enough compressive strength.

3.4 The first experiment

The first experiment was done using Soil Sample A. According to the Proctor Compaction test results, Soil Sample A had the following characteristics:

- Maximum Dry Density (MDD) = 1.544 g/cm3

- Optimum Moisture Content (O.M.C) = 19%

These are fairly good values of compaction. The soil was of limited quantity, such as would not allow me to repeat the experiment with a different degree of compaction.

3.5 The Second Experiment

The second set of experiments was done using Soil Sample B. Soil Sample B was not as good as Soil Sample A. The density and moisture content values of soil sample B are as shown below:

- Maximum Dry Density = 1.42 g/cm3

- Optimum Moisture Content = 21%

This sample was enough quantity and so it was possible to conduct two experiments for comparison.

For the first experiment, 20 kilograms of soil was used. This was compacted in four layers, each of approximately 5 kilograms. For each layer, 200 milligrams of water was added.

For the second experiment, the soil mass was increased to 24 kg. The water ratio was also increased. The same number of weights were placed on the soil as was done in the first experiment.

A - The box used for the experiment was made of natural wood. Its overall dimensions were 40 cm x 25 cm x 22 cm. One of the sides has restrains to allow the sliding edge removal after compaction.

B - This shows the compacted soil still in the box. The sliding part of the box has been removed to expose the embankment as in the real situation.

C - Weights were incrementally placed on top of the soil, which came to a total of 83 kg. The weights provide the required amount of pressure that would be expected in the real situation. Observation is made after addition of each weight to check whether the embankment collapses or withstands the pressure. The weights were placed on top of each other beginning with the heaviest. Two concrete cubes of stones are placed at the base of the weights for support and even distribution of pressure.

D - Shows the arrangement of weights from a top angle. The weights have been properly aligned to provide even distribution of pressure

A - This shows the second experimental setup. When the weights were placed on the soil, the embankment collapsed. This is the soil that had achieved a dry density of 77 %.

B - The embankment collapses when the weight is increased to 83 kg. The soil in the first experimental setup was able to withstand the same weight of 83 kg. The scattering of the weights with the collapse depicts the kind of accidents that may occur when an embankment collapses.

C - This image shows the top of the collapsed embankment. The failure can be categorized as shear failure. The exposed side of the embankment collapses when it could no longer sustain the pressure.

D - This image shows the intact top of soil for the first experimental setup. By mere eye observation, this soil looks better compacted than the soil in image C. The difference in compaction is what led one to collapse and the other one to remain intact.

Chapter 4

4.1 Introduction

This chapter presents the key experimental results for all the laboratory experiments done. The first stage followed the Proctor Compaction Procedure for determination of Maximum Dry Density and Optimum Moisture Content. The second stage involved compaction of the soil in a similar fashion to the proctor compaction test, and then loading the compacted soil incrementally with various weights. For the second stage, the results are more observational. No alphanumeric data can be presented for this second set of experiments. We shall use images to present the results for this stage of the experiments.

4.2 Calculating dry density achieved for Experiment Set Up 2

Density = Mass/Volume

Volume = 40 x 25 x 17

Mass = 24,000 K

Bulk Density = 24,000/1,700 = 1.411 g/cm3

Percentage %wc = 29%

Dry Density = Bulk Density/(1+w%)

= 1.411/(1+0.29)

= 1.094 g/cm3

Maximum dry density from proctor test = 1.42g/cm3

Dry density from test = 1.094 g/cm3

Max dry = 1.42 g/cm3

% compaction = 1.094/1.42*100 = 77%

4.3 Calculating dry density achieved for Experiment Set Up 1

Mass of soil = 20,000 g

Volume of Soil = 18.5 x 40 x 25 = 18,500

Bulk Density = 20,000/18,500 = 1.081 g/cm3

Percentage moisture content = 19%

Dry Density = 1.081/(1+0.19) x 100 = 90.84% = 91%

Chapter 5 Discussion and Data Analysis

5.1 Introduction

This chapter presents an analysis of the results obtained from the experiment. This is mostly in form of graphs. Below is a plot of the Dry Density vs Moisture Content for Soil Sample A as obtained from the Proctor Compaction Test.

The graph shows a distinct peak, as expected. This peak shows the maximum dry density that can be achieved for the soil, and also the optimum water content required to achieve that density. For most soils, when compacted, the density of the soil increases up to a certain point, as the moisture content is increased. When the water content is increased beyond the optimum water content level, the density of the compacted soil does not increase any further. Instead, the density starts decreasing with further increase in the water content. Each compactive effort for a given soil has its own Optimum moisture content. When the compactive effort is increased, the maximum dry density likewise increases, but the optimum moisture content decreases.

From the graph above, the following soil characteristics can be deduced.

Maximum Dry Density for soil sample A is 1.544 g/cm3

Optimum Moisture Content for soil sample A is 0.190 %

The figure below shows the Dry Density vs Moisture Content Graph for Soil Sample B

From the graph, we can get the following soil parameter values:

Maximum Dry Density for Soil Sample A is 1.42 g/cm3

Optimum Moisture content for soil sample A is 21 %

5.2 Dry Density versus Slope Stability

From the results of the experiment, the dry density achieved under Experiment 1 was 91%. When the weights were placed on the soil, the soil did not collapse even when the total weight reached 83 kg. The soil, having been compacted well, resulted in a stable embankment that could withstand the pressure from above. On the other hand, the dry density achieved under Experiment 2 was 77%. After adding the about 83 kg of weights, the compacted soil collapsed. In other words, the embankment collapsed.

Chapter 6 Conclusion

In this research, a number of laboratory experiments were conducted. The experiments revolved around finding the various density types of the different soil samples. These included dry density, bulk density, and wet density. Other soil parameters investigated included moisture content and compressibility. From the analysis of results presented above, the following conclusions can be made.

The maximum density achievable through compaction depends on the amount of water added to the soil. There exists an optimum moisture content for which the density becomes maximum. Any amount of water below this this optimum value will result in a lower dry density for the compacted soil. Likewise, any amount of water above the optimum level yields a compacted soil with less density than the maximum achievable dry density.

The compressibility of a given soil depends on its dry density. A greater dry density corresponds to less compressibility. This is the main reason the soil in Experiment Setup A did not collapse. On the other hand, a lower dry density corresponds to a higher degree of compressibility. That is the reason the soil in Experiment Setup B failed as the weight on it continued increasing.

Slope stability is affected by the degree of compaction of the soil. A well compacted soil results in a more stable slope. Stable slopes will not fall unless the weight on it is overly excessive. A poorly compacted soil will lead to a less stable slope. Even small pressure can cause its collapse

Chapter 7 Future work

The current study concentrated on the effect of compaction on embankment stability. The experiment set up depicted a purely vertical embankment. This may not usually be the case in the real life situation. Embankments are not usually purely vertical, be it in dam construction, railway construction, or road construction. Embankments normally slope out. The degree of the slope may either be gentle or steep. This is one factor that this study did not consider. For this reason, the following recommendations can be made.

A possible future study should include an experiment set up that simulates an embankment with a steep slope, rather than a purely vertical slope. This will better represent the real situation in the field.

Another possible advancement of this study should involve different types of soil with varying parameters such as cohesion and angle of internal friction.

References

Bel'fer, S. (1964). The influence of soil condition on deformation of the base of an embankment. Soil Mechanics and Foundation Engineering, 1(6), pp.359-361.

Belloc, H. (1967). On. Freeport, N.Y.: Books for Libraries Press.

Escario, V. and S´ez, J. (1987). Discussion: The shear strength of partly saturated soils. Géotechnique, 37(4), pp.523-524.

Gallipoli, D. (n.d.). Partial Saturation in Compacted Soils.

Ghazireh, N. and Billam, J. (1989). Elasto-plastic unsaturated soil subject to impact. Géotechnique, 39(3), pp.525-526.

886347 High embankment dams and collapse settlements due to reservoir filling. (1988). International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 25(6), p.304.

Kumor, Ł., Kumor, M. and Kopka, M. (2017). Geotechnical Parameters of Soil, Considering the Effect of Additional Compaction of Embankment. Procedia Engineering, 189, pp.291-297.

Schanz, T. (2007). Experimental Unsaturated Soil Mechanics. Berlin, Heidelberg: Springer-Verlag.

Seed, H., Lundgren, R. and Chan, C. (1954). The effect of compaction method on the stability and swell pressure characteristics of soils. [Place of publication not identified]: [publisher not identified].

The factors influencing collapse settlement in compacted soils. (1975). International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 12(9), p.133.

Wheeler, S. and Sivakumar, V. (1995). An elasto-plastic critical state framework for unsaturated soil. Géotechnique, 45(1), pp.35-53.

Zwanzig, F. (1980). Embankment stabilization and soil mechanics. Washington, D.C.: National Academy of Sciences.