THE EFFECTS OF FLUID’S PHYSICOCHEMICAL CHARACTERISTICS ON PIPING EROSION PROGRESSION OF A SILTY SAND

ABSTRACT

This thesis presents an experimental research for studying the absolute and interactive effects of three physicochemical characteristics of permeating fluids (viscosity, pH, and ionic strength) on piping erosion progression of a sandy soil under turbulent flow. Full factorial experimental design was used to produce eight types of fluids of various combinations of the three fluid characteristics. Hole erosion tests were conducted on identically prepared silty clayey sand specimens. Erosion rate index, an important index was used to quantify the relative erosive capacity of the test fluids. The erosion rate index was quantified for two repeat trials for each of the eight test fluids. Regression analysis was conducted on the results to generate a statistic model to describe the effects of the three factors and their interactions. The main findings include: (1) Viscosity, pH and ionic strength were all determined to be significant factors on the rate of erosion. The two-way interactions between viscosity and pH, and between viscosity and ionic strength were also determined to be significant interaction factors, while the interaction between pH and ionic strength did not prove to be statistically significant. (2) Higher pH causes higher erosive capacity; higher ionic strength causes lower erosive capacity of fluid. (3) At low viscosity, ionic strength does not affect the erosion; but when the viscosity is higher (or the fluid temperature is colder), higher ionic strength causes much less erosion. There is almost no interactive effect between pH and ionic strength.

 

 

TABLE OF CONTENTS

LIST OF FIGURES…………………………………………………………………………………………… vi

LIST OF TABLES…………………………………………………………………………………………….. ix

ACKNOWLEDGEMENTS…………………………………………………………………………………. x

CHAPTER 1    INTRODUCTION………………………………………………………………………. 1

1.1 Background……………………………………………………………………………………………… 1

1.1.1 Piping and Backward Erosion……………………………………………………………. 1

1.1.2 Suffusion………………………………………………………………………………………… 3

1.1.3 Remediation of Internal Erosion………………………………………………………… 4

1.2 Research Motivation and Objectives…………………………………………………………… 5

1.3 Methodology……………………………………………………………………………………………. 6

1.4 Organization…………………………………………………………………………………………….. 7

CHAPTER 2   LITERATURE REVIEW…………………………………………………………….. 8

2.1 Review of Erosion Testing…………………………………………………………………………. 8

2.1.1 Dispersivity Test……………………………………………………………………………… 8

2.1.2 Surface Erosion Test………………………………………………………………………. 10

2.1.3 Internal Erosion Test………………………………………………………………………. 10

2.2 Review of Particle’s Fate in Porous Media…………………………………………………. 14

2.2.1 Particle Mobilization………………………………………………………………………. 14

2.2.2 Particle Transport and Deposition……………………………………………………. 16

2.3 Review of Internal Erosion Affected by Permeating Fluids………………………….. 17

CHAPTER 3   RESEARCH MATERIALS AND MATHODOLOGY…………………. 20

3.1 Experimental Setup…………………………………………………………………………………. 20

3.2 Design of Experiments and Fluid Preparation…………………………………………….. 22

3.2.1 Adjust Viscosity…………………………………………………………………………….. 23

3.2.2 Adjust pH……………………………………………………………………………………… 23

3.2.3 Adjust Ionic Strength……………………………………………………………………… 24

3.3 Experimental Procedure…………………………………………………………………………… 25

3.4 HET Data Processing………………………………………………………………………………. 26

3.5 Regression Analysis of HET Results…………………………………………………………. 28

3.6 Determination of Soil’s Shear Strength under Different Temperatures………….. 29

CHAPTER 4    RESULTS AND ANALYSIS……………………………………………………… 32

4.1 Results of Erosion Rate Index Calculation…………………………………………………. 32

4.2 Results of Regression Analysis…………………………………………………………………. 49

CHAPTER 5   CONCLUSIONS………………………………………………………………………… 56

APPENDIX A: FLOW CELL DISIGN………………………………………………………………. 64

APPENDIX B: RAW DATA OF HET TEST……………………………………………………… 69

CHAPTER 1    INTRODUCTION

1.1 Background

1.1.1 Piping and Backward Erosion

Dams     are    considered     as     “installations     containing    dangerous      forces”

under International Humanitarian Law due to the massive impact of a possible destruction on the civilian population and the environment (Baker, 1991). Dam failures are comparatively rare but can cause immense damage and loss of life when they occur. Piping is known to cause catastrophic failures of levees and earthen dams (Seed et al., 2008; Terzaghi, 1943) and has been studied for many years. In the recent years, several disasters in bridge foundations, quarries, levees and earthen dams indicate that soil erosion is a main factor of failure. (Fry et al., 1997; Guiton, 1998; Foser et al., 2000; Fell and Fry, 2007; Briaud, 2008). Two types of erosion can be distinguished: surface erosion that occurs at the soil and water interface and internal (subsurface) erosion, which takes place inside the soil matrix.

Several authors have concluded that soil structures can be divided into two groups:

A primary structure and a secondary structure (Barakat, 1991; Kenney and Lau, 1986; Lafleur et al., 1989; Tomlinson and Vaid, 2000). Basically, the primary structure is a network of grains that form the skeleton of the soil matrix, while the secondary structure is the fines that are in the voids of the primary structure. In internal erosion, there exist two types: one is piping and backwards erosion, which affects both the primary and secondary structures. The other is suffusion that primarily affects only the secondary soil structures.

 

The piping progression process includes piping mobilization, particle transport, and particle deposition. Soil particles are removed from a soil matrix (i.e., dislodging) when the hydraulic shear forces exerted by seepage flow exceed the forces that keep the particles within the soil matrix. The dislodging of these particles can gradually form a continuous, tubular cavity that resembles a pipe.

Backward erosion is an internal erosion mechanism, during which shallow pipes are formed in the direction opposite to the flow underneath water-retaining structures as a result of the gradual removed soil by the action of water. It is an important failure mechanism in both dikes and dams where sandy layers are covered by a cohesive layer; it occurs rapidly leaving little time for reaction (Charles, 1997). Figure 1-1 illustrates the progression of piping and backward erosion.

 

 

Figure 1-1. Progression of piping and backward erosion (Shwiyhat, 2010)

 

Piping and backward erosion are detected at the landside face of a levee or earthen dam. Wet spots, sand boils, and sand heaves are used as the indicators of such activities. Wet spots are areas on the land side of a levee or earthen dam that has become saturated due to seepage. Sand boils and heaves occur when it reaches to the critical hydraulic gradient, which pushes against the downward vertical force of soil until an exit point is created. Figure 1-2 shows a photo of sand boil.

 

Figure 1-2. Sand bags placed around a sand boil (Hickman, KY, 2011)

 

1.1.2 Suffusion

Suffusion is defined as the detachment and migration of fine grained particles through the pores of a soil matrix that is composed of coarse grained particles (Bendahmane et al., 2008). The detachment and migration of coarse grained particles are the result of seepage flow through a hydraulic structure. This type of internal erosion affects the secondary structure of a soil matrix and can increase the potential differential settlement and permeability of the soil. But suffusion is a long-time process that can take up to several years. As for the micro structures in suffusion, grains of the secondary structure are entrained into seepage flow and eroded out from the spaces between large particles. As time goes, the primary structures finally remain in contact and static. The diameter of the fine grains that are removed must be smaller than that of the pore throat, or pore clogging will occur. Figure 1-3 illustrates the progression of suffusion.

 

Figure 1-3. Progression of suffusion (Shwiyhat, 2010)

 

The characteristics of soil greatly contribute to the process of suffusion. This erosion is typical of a gap graded or segregated material. So, careful selection of materials can in most cases help alleviate the problems that lead to suffusion.

1.1.3 Remediation of Internal Erosion

Figure 1-4 shows a slurry cut-off wall, which is an impermeable barrier that is installed in an earthen embankment and penetrates the underlying permeable foundation soil. It is considered the most effective means of eliminating seepage and subsurface erosion problems and is widely used. Cut-off walls include slurry walls and sheet pile walls. A slurry wall is installed by first excavating a trench that is backfilled with slurry. There are various types of slurry materials, such as cement-bentonite slurry, soil-cementbentonite slurry, and soil-bentonite slurry. The slurry solidifies into an impermeable wall along the longitudinal direction of the levee or earthen dam. A sheet pile wall is made of interlocking steel sheet piles that are driven through the embankment and into the foundation soil using a vibratory hammer. A cut-off wall should penetrate most (such as 95%) of the permeable soil stratum. If the pervious foundation soil has significant depth, installing cut-off walls may not be economical.

 

Figure 1-4. Cross-section view of a slurry cut-off wall in a levee (After USACE, 2000)

1.2 Research Motivation and Objectives

In the case of water-retaining embankments such as dams and levees, which are essential for the safety and quality of life for people throughout the world, subsurface erosion (internal erosion) is of particular concern and many catastrophic failures have been attributed to its potentially destructive consequences, including the 1972 failure of the

Buffalo Creak Dam in West Virginia (Davies et al., 1972), the 1976 Teton Dam failure in Idaho (Penman, 1987; Sherard, 1987), the 1990 Cyanide Dam failure in North Carolina

(Leonards and Deschamps, 1998), and three levee breaches during Hurricane Katrina in 2005 (Seed et al., 2008a; Sills et al., 2008b).

In previous experimental studies on internal erosion, tap water or de-ionized water have often been used as a permeating fluid. However, when the fluid permeates through soil and interacts with the environment, its properties are altered from those of pure water (Hillel, 1998). Significant research has been conducted in the past to investigate the effects of physicochemical characteristics of fluids (with focus on pH and ionic strength) on the incipient motions of colloidal-size (<1 μm) particles, such as glass microspheres (Sharma et al. 1992) and colloidal titanium hydrous oxide spheres (Hubbe, 1985). These studies generally concluded that higher pH and lower ionic strength result in easier particle mobilization. Limited research has been conducted on the effects of permeating fluid’s physicochemical characteristics on soil’s erosion behaviors. Past research preliminarily revealed the individual effects of a fluid’s viscosity, pH, and ionic strength on the mobilization of colloidal particles and the erosion of clayey soils. Whether and how these physicochemical characteristics affect the piping progression on sandy soil is unknown. Therefore, the main objective of this research is to reveal the relative and interactive of fluid’s physicochemical characteristics (viscosity, pH and ionic strength) on piping erosion progression of a sandy soil.

1.3 Methodology

The erosion characteristics are often described by the erosion rate index and critical shear stress. Critical shear stress is the minimum hydraulic shear stress required to mobilize a particle, also known as incipient motion. Piping erosion progression is often indicated by erosion rate index, which is quantified in this research for fluids with various combinations of physicochemical characteristics. In this research, three physicochemical characteristics

(factors) of fluid were studied: pH, ionic strength, and viscosity. Distilled water was used as the base of each test fluid. Each fluid parameter had a target low and high levels. The experimental design approach followed three-factor, two-level, full factorial design. Hole erosion test (HET) was used to simulate piping erosion process and the method given by

Wan and Fell (2004) was used to calculate the erosion rate index of each permeating fluid. Regression analysis, ANOVA (analysis of variance), was used to show the relative and interactive effects of viscosity, pH and ionic strength on piping progression of a sandy soil.

1.4 Organization

This thesis consists of five chapters. Following the introduction in this chapter, Chapter 2 presents a synthesis of published studies on erosion testing, particle fate in porous media, and internal erosion influenced by permeating fluids.

Chapter 3 presents the research materials and methodology, including the experimental setup of HET tests, fluids preparation, direct shear testing of soils in different temperatures, HET data processing, regression analysis of HET results and determination of shear strength under different temperatures.

Chapter 4 presents the results and analysis, including the results of erosion rate index calculations and the results of regression analysis.

Chapter 5 presents the conclusions from this research.

THE EFFECTS OF FLUID’S PHYSICOCHEMICAL CHARACTERISTICS ON PIPING EROSION PROGRESSION OF A SILTY SAND

Sharing is caring!