Citation
Synthetic randomly distributed fiber reinforced compacted sandy clay soil

Material Information

Title:
Synthetic randomly distributed fiber reinforced compacted sandy clay soil
Creator:
Blanchette, David Alexander
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
viii, 147 leaves : illustrations ; 29 cm

Subjects

Subjects / Keywords:
Polypropylene fibers ( lcsh )
Soil stabilization ( lcsh )
Polypropylene fibers ( fast )
Soil stabilization ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 145-147).
Thesis:
Submitted in partial fulfillment of the requirements for the degree, Master of Science, Civil Engineering
General Note:
Department of Civil Engineering
Statement of Responsibility:
by David Alexander Blanchette.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
34872698 ( OCLC )
ocm34872698
Classification:
LD1190.E53 1995m .B53 ( lcc )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

SYNTHETIC RANDOMLY DISTRIBUTED FIBER REINFORCED COMPACTED SANDY CLAY SOIL by David Alexander Blanchette B. S., University of Colorado at Denver, 1979 A thesis submitted to the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering 1995

PAGE 2

This thesis for the Master of Science degree by David Alexander Blanchette has been approved by Peric' Date

PAGE 3

ACKNOWLEDGEMENT I would like to thank Dr. Nien-Yin Chang for his faith and belief me during the preparation of this report and throughout my graduate studies. In addition I would like to thank him for the guidance and opportunity to complete this study.

PAGE 4

Blanchette, David Alexander (M. S., Civil Engineering) Synthetic Fiber Reinforced Sandy Clay Soil Thesis directed by Professor NienYin Chang ABSTRACT In the past decade there has been increasing interest in the study of randomly distributed synthetic fiber reinforced soil. The major emphasis of studies performed have focused on the study of fiber reinforced sandy soil. Therefore this study was undertaken to study fiber reinforced sandy clay soils. The study includes performing a series of common laboratory test to investigate and quantify how sandy clay soil behavior is modified by the addition of randomly distributed synthetic fibers. Laboratory tests performed are compaction, one dimensional consolidation, unconfined compression and consolidated undrained effective stress triaxial test. Reinforcement type used for the study is a thin two inch long collated fibrillated polypropylene fiber, trade name Fibermesh. Two separate soil mixtures are used in this study. One mixture consisted of 25 percent clay and 75 percent sand, the second consisted of 35 percent clay and 65 percent sand. The clay used for the study is a floated kaolinite clay and the sand material is a natural colorado sand formed during concrete aggregate production. iv

PAGE 5

Tests samples were all prepared by static compaction and resulted in producing overconsolidated soil samples for the range of initial effective confining pressures used during triaxial testing. Tests results for both soil mixtures showed that the addition of the fibers increased the soil strength and stiffness. Recorded pore pressure readings taken during triaxial tests showed that fibers had a significant effect on the development of excess pore pressure. This effect is believed to be caused by widening of the shear failure zone during loading. In addition to this finding, tests results indicated that overconsolidated fiber reinforced soil shear strength is a function of the overconsolidation ratio and initial stress conditions similar unreinforced overconsolidated soils. Therefore it is believed that the strength of overconsolidated fiber reinforced sandy clay soils can be represented by the Hyperbolic Stress-Strain model. This Abstract accurately represents the content of the canidate's thesis. I recommend its publication. v

PAGE 6

CONTINI'S Chapter 1. Introduction . . . . . . . . . . . . 1 1.1 Problem Statement 1.2 Research Objective 1.3 Study Scope . . . Engineering Significance 2.0 literature Review . 2.1 Freitag Study 2.2 Gray, et al. Study 2.3 Maher Study. Shrew bridge and Sitar Study. 2.5 Jewell and Wroth Study. . 2.6 Crockford, Grogan, and Chill Study 3. Soil Materials and Material Mixing . . 3.1 Materials 3.1.1 Clay Materials. 3.1.2 Sand Materials. 3.1.3 Reinforcement Fibers 3.2 Material Mixing. . 4. Compaction Charaacteristics vi 3 5 8 9 .10 .11 13 .13 .13 .15 16 16 20

PAGE 7

4.1 Test Equipment 20 4.2 Test Procedure 20 4.3 Test Results. 21 5. Unconfined Compression Tests and Results. 25 5.1 Test Equipment . 25 5.2 Sample Preparation 26 5.3 Test Procedure 27 5.4 Test Results 6. Isotropically Consolidated Undrained Triaxial Test. 37 6.1 Triaxial Test Equipment. . . 37 6.2 Triaxial Test Sample Preparation 38 6.3 Sample Saturation. 39 6.4 Sample Loading 40 6.5 Test Results. 41 6.5.1 Mohr Columb Effective Stress Failure Envelope . 42 6.5.2 Deviator Stress and Pore Pressure Verses Strain Curves. 47 6.5.3 Effect of Test Confining Pressure 58 6.5.4 Discussion of A Coefficient Plot 61 7. One-Dimensional Consolidation Test 70 7.1 Test Equipment 70 7.2 Test Procedure. 70 7.2.1 Sample Preparation. 70 7.2.2 Sample Loading and Unloading. .71 7.3 Consolidation Tests Results . . 72 vii

PAGE 8

8. Hyperbolic Stress-Strain Model 8.1 Hyperbolic Stress-Strain Model Development 8.2 Transformed Hyperbolic Stress-Strain Test Results 8.3 Initial Tangent Modulus Test Results. 8.4 Compressive Strength Test Results 8.5 Hyperbolic Stress-Strain Path . 9. Summary, Conclusions Recommendations 9.1 Summary 9.2 Conclusions 9.3 Recommendations. Appendix A Photographs................. B Mix Al A2 Tests Data. C Mix B1 B2 Tests Data. References . . . . viii .79 .79 .88 .99 101 105 111 111 .111 .112 113 126 136 .145

PAGE 9

1. Introduction 1.1 Problem Statement The use of Randomly Mixed Synthetic Fiber (RMSF) additives to strengthen soils is an emerging technology. This technology has only been studied within a limited scope in the laboratory for the past decade. More recently full scale field test studies have been conducted using RMSF additives in retaining wall backfills and road pavement support base courses. With any new engineering technology there must be a Significant quantity of quality laboratory research performed to insure safe and economical use, therefore this study was undertaken. 1.2 Research Objective This report presents the results of a study on the effects of a RMSF additive on the engineering properties of two compacted sandy clay soils. There were two main objectives of this study. They were to provide a better understanding of the effects RMSF have on sandy clays and to enhance the design of earth structures involving synthetic fibers as strengthening additives. The first objective is obtained in this report by performing and evaluating laboratory test data. The second objective is obtained by suggesting the potential use of a typical soil

PAGE 10

stress-strain model for use in analysis and design of sandy clay soil structures reinforced with RMSF additives. 1.3 Study Scope The scope of this study was to perform a series of laboratory tests to evaluate common soil properties used to design earth structures. These design properties included compressibility, compaction characteristics, unconfined compressive strength, and undrained triaxial shear strength. Standard laboratory tests performed to study these soil properties included one dimensional consolidation, standard compaction, unconfined compression, and isotropically consolidated undrained triaxial shear strength tests. Unconfined compression tests were performed over a range of water contents and undrained triaxial tests were performed over a range of initial effective confining pressures. Two general soil mixes were used for the study. The soil mixes used were chosen to represent two designed soil mixtures which could potentially be used in a landfill cap or landfill liner. The first general soil mix was designated the A mix and consisted of 25% clay and 75% sand. The second general soil mix was designated the B mix and consisted of 35% clay and 65% sand. Each mix was tested with and without a RMSF additive. Mixes designated Al and Bl were tested without fibers and mixes designated A2 and B2 were tested with fibers. 2

PAGE 11

Upon completion of all tests, data was graphed, evaluated, and analyzed for differences in trends in soil behavior for mixes with and without fibers. This evaluation included reviewing current soil stress-strain modeling techniques for applicability to the study test results. 1.4 Engineering Significance The use of RMSF additives for strengthening soils has real potential for use in the engineering and design of foundations and earth structures. Some promising applications are: 1. Local inclusion beneath spread footings to increase soil strength and reduce footing size. 2. Stiffening of soil beneath machine foundations to reduce vibrations and the resulting machine wear. 3. Addition to retaining wall backfIll soil to increase soil strength and reduce requirements of wall strength. 4. Addition to embankment fill soil to increase strength and reduce safe slope requirements of the fill, therefore reducing fill size. The engineering decision to consider RMSF additives for the above applications could and typically would be based on economics. For example would the cost of mixing embankment fill material with synthetic fibers be less costly then placing a greater quantity of fill not mixed with fibers? The answer to this question could be yes, but increasingly engineers must consider other factors when making decisions. These other factors such as land use, visual impact, or 3

PAGE 12

schedule may govern the decision process therefore enhancing the viability for the use of RMSF soil additives. In conclusion with further in-depth research and development and publication on this exciting new technology a new tool could be developed for the Civil Engineer to use in solving engineering problems. 4

PAGE 13

2. Literature Review As stated previously there have been a limited number of studies performed to investigate physical property changes of sand or clay soils from the inclusion of Rl\1SF. The literature search revealed only a few studies on compacted sandy clays with the inclusion of RMSF. The scope of existing studies on compacted sandy clay is limited in that there exist so many variables, such as mixing, compaction method, and clay content. In general current studies have shown that ultimate strength and stiffness of sand or clay increases with the addition of fibers. The following is a review of the more relevant and significant studies as related to this report. Discussion of these important studies is referenced by the leading authors of the reported results. 2.1 Freitag Study Freitag (1986) reported that for a lean sandy clay, liquid limit 42 and plasticity index 22, there was a 25% increase in unconfined compression strenths on samples wet of optimum. Three seperate fiber types were used for the study. Fibers used included a spun nylon string, a polypropylene rope fiber and a 3/4 inch long polypropylene olefin concrete reinforcement fiber, trade name Fibermesh. Freitag found that for fiber types studied their effect on increase in unconfined

PAGE 14

compression strength was relatively constant. Fiber weight content used for the study was equivalent to 1% of the total sample dry weight. 2.2 Gray, et al. Study Gray, et al. (1986) performed a series of Triaxial Compression test on RMSF reinforced samples of dry dune sand from Muskegon, Michigan. Synthetic Fibers used for the study were made of glass, No. 204 filament strand produced by Owings Corning Glass. Lengths of the glass fibers used were, 0.5, 1.0, and 1.5 inchs. Diameter of the glass fibers was 0.012 inch. Triaxial Compression tests on sands reinforced with 0.5 and 1.0 inch long fibers were performed on specimens which were 1.42 inches in diameter and 3.15 inches in height. Tests on sands reinforced with 1.5 inch long fibers were performed on specimens which were 2.8 inches in diameter and 7.1 inches in height. Gray reported that for dry sands, compressive strength as expressed by their major principle stress at failure, for a given fiber and aspect ratio, was proportional to the weight fraction of the fibers. Within this study aspect ratio was defined as the ratio of length to diameter. There study showed that strength varied linearly with aspect ratio for a constant weight fraction and that the slope of the strength verses aspect ratio curve increased as weight fraction increased as shown in figure 2.1. It was also reported that rougher fibers were more efficient in obtaining strength gain. In addition, strength gain for fibers tested

PAGE 15

varied linearly up to a sample fiber content of 2% by weight and thereafter approached an asymtotic upper limit as shown in figure 2.2. 450 wght fraction = 0.5% .... -+-wght fraction = 1.0% Q) 400 wght fraction =2.0% 350 300 Q) 250 '"' ...... aJ 200 0. .... (,) 150 .... Po. 100 '"' 50 0 0 20 40 80 100 120 140 Aspect Ratio (fiber length / diameter) Figure 2.1 Major Principal Stress at Failure for Fiber-Reinforced Sand as Function of Aspect Ratio at Different Fiber Weight Fractions (ref. Gray et al. 1986)

PAGE 16

400 Q) 5 ..... OJ Q) 200 Q) CIt Po 100 or :al o 2 3 4 Weight of Fiber in Sample Figure 2.2 Strength Increase as function of Amount of Internal Reinforcement in Triaxial Compression Test (ref. Gray, et al. 1986) 2.3 Maher Study 7 Maher, (1990) reported results for a series of static load tests for sands reinforced with RMSF for five different fiber types. Maher used these test results to develop a model which predicted increases in strength for sands reinforced with RMFS. The model proposed is a function of sand granulometry and fiber properties. The model is in the form of two equations. Maher determined that for sands a critical confining pressure existed at which point the Mohr Coulomb failure envelope of sands reinforced with RMSF was parallel to the failure 8

PAGE 17

envelope of sands tested without reinforcement. Below the critical confining pressure the failure envelopes diverged with reinforced sands exhibiting higher strength. Maher explained this behavior by stating that the soil fiber interaction below the critical confining pressure was governed by slipping of the fibers. At effective confining pressures above the critical confining pressure the soil fiber interaction was yielding of the fibers. Additional key finding of Maher's study were: (1) increase in fiber aspect ratio, d, resulted in a lower critical confining pressure, and a higher fiber contribution to increased sand shear strength; (2) Fibers with very low modulus contribute little to increased sand strength; (3) A more uniform sand gradation resulted in a lower critical confining pressure and a higher fiber contribution to strength; (4) increase in sand particle sphericity resulted in a higher critical confining pressure and a lower fiber contribution to strength; (5) increase in sand grain size had no effect on critical confining pressure, but reduced fiber contribution to strength. 2.4 Shew bridge and Sitar Study Scott E, Shewbridge and Nicholas Sitar (1989) reported on a series of direct shear tests performed to study the development of the shear zones in reinforced soil composites. They reported that the width and shape of a shear zone in a reinforced soil is altered by the reinforcement, concentration, stiffness ,and reinforcement to soil bond strength. 9

PAGE 18

Their study showed that shear zones are wider in reinforced soils and that the width of shear zones increases with increasing stiffness of the soil composite. Increases in stiffness of the soil composite in their study was obtained by increasing reinforcement concentration or by increased bond strength between the soil and reinforcement. They observed that although reinforcement concentration and reinforcement type affects measured shear strength of a reinforced soil composite the increase was not linear. This test result was explained to be a result of the reinforcement increasing the width of the shear failure zone and therefore distributing dilative strains within the soil. Modifying dilative strains within the soil mass would then change the measured friction angle of the soil and strength increases would not be linear with reinforcement concentration. 2.5 Jewell and Wroth Study Jewell and Wroth (1987) presented data on a series of direct shear tests performed on reinforced sand. Circular polymer reinforcement bars 0.066 inches in radius and 5.0 inches long were used as reinforcement. In addition inextensible steel rod was used. Their test results were: 1) Reinforcement does not increase shearing resistance until that of the sand exceeds the critical state value; 2) Increases in sand shearing resistance for very stiff reinforcement reachs a maximum value after which it is maintained at a constant relative to unreinforced sand as deformation continues; 3) Increases in reinforced 10

PAGE 19

sand shear resistance are a function of reinforcement stiffness and shear resistance increases as deformation continues. It is therefore possible that reinforced sand shear strength increases can occur where a unreinforced similar sand might have failed; 4) Reinforcement within a sand composite distributes local high shear strains; 5) The bond angle of friction for rough reinforcement equals the direct shear angle of friction. 2.6 Crockford, Grogan, and Chill Study The Crockford, Grogan, and Chill study was performed as a cooporative effort between the Texas Transportation Institute, U.S. Army Corps of Engineers and Synthetics Industries, Incorporated. The study was undertaken to evaluate the life extension performance of fiber reinforced chemically stabilized silty sand and clay road base soils. Reinforcement type used for the study was a polyolefin fiber that was 1.0 inch long, 0.1 inch wide and .01 inch thick. This fiber is similar to the fiber type used for this study. The study included evaluation of different compaction efforts, confined and unconfined compression tests and road field tests. Clay soil used for the study was classified as CL and had a plasticity index of 22. Within the study results for fiber reinforced soils without chemical stabilization were reported. Only these results are discussed here. Tests on clays showed that there exists an optimum fiber density curve similar to the standard moisture density curve. The optimum

PAGE 20

density curve indicated that there was a peak value of dry unit weight that could be obtained for soils mixed at optimum moisture contents at varing values of fiber content. Strength tests showed that polyolefin fibers increased the modulus of elasticity, strength and strain energy. Strain energy within the study was defined as being the area under the stress strain curve. Changes in soil properties indicated that simultaneously increasing the modulus and strength could delay the onset of failure in road base material while an increase in strain energy could help slow down the failure process once it had started. Plots of strain energy verses fiber content for the clay soil studied showed a linearly decreasing trend as fiber content increased between o and 1 percent of the soil dry weight. Plots of strain energy verses fiber content for the sand soil studied showed a linearly increasing trend as fiber content increased between 0 and 1 percent of the soil dry weight. Clay response was believed to be caused by mixing and or compaction procedures or possibly because the strain energy may be more of a function of the optimum fiber density curve. 12

PAGE 21

3. Soil Materials and Material Mixing 3.1 Materials Soil materials used for this study included a store bought floated Georgian kaolinite clay, a well graded crushed natural Colorado sand and a floated western bentonite clay. Synthetic reinforcement fiber type used for this study was a virgin polypropylene concrete reinforcement fiber. Tap water was used as a saturating fluid for all test. 3.1.1 Clay Materials Two types of clay soils were used for this study. The primary clay fraction of all soil mixes consisted of a standard air floated clay called DB Float. DB Float is a non-hygroscopic, fine particle, hydrated aluminum silicate. It is produced from soft Georgia kaolin by dry processing. Typical manufacturer reported chemical analysis and physical properties are shown in following tables 3.1 and 3.2. Standard DB Float can be formulated in non-aqueous and aqueous systems. It disperses readily in water with common deflocculating agents such as polyphosphates, or sodium silicates. It can serve as a general purpose filler in paper, textiles, linoleum, rubber, as an inert diluent for pesticides and herbicides, etc.

PAGE 22

Table 3.1 DB Float Typical Chemical Analysis Si02 (combined ), % 45.1 Al203 (combined) 38.1 Ti02 1.4 Fe203 0.6 Na20 0.3 MgO 0.2 K20 0.1 CaO Trace Table 3.2 DB Float Typical Physical Properties Free Moisture (maximum), % 79.0-81.0 pH(20% by weight in distilled water) 3.0 Sieve Residue (200 mesh maximum),% Oil Absortion (ASTM 0 381-31) Refractive Index Specific Gravity Bulking Value (U.S. gallon per pounds) Weight per Solid Gallon, pounds Dry Density (lbs per cubic foot, loose pack) Hardness, Mohs' Scale, Soft, non-abrasive 14 0.10 37-41 1.56 2.65 0.0462 22.66 25-30 2

PAGE 23

In order to represent a true landfill liner soil mix design a small quantity of floated bentonite was used as a mix constituent for all soils tested. The quantity of bentonite used was 2% by dry weight and was kept constant for all mixes. There is considerable discussion on how much bentonite should be used for liners based on possible leachate degradation and therefore a minimum amount was chosen for this study. The bentonite used for this study was a standard air floated clay manufactured in Wyoming U. S. Manufacturer physical and chemical properties were not available for the bentonite clay. A specific Gravity of 2.65 was assumed for reporting of test results. 3.1.2 Sand Materials Sand used for all mixes was obtained in Golden Colorado from a Mobile Premix sand and gravel production plant. Physical properties for sand are contained in the following table 3.3. Table 3.3 Sand Phvsical Properties Specific Gravity and Absorption (ASTM C128) 2.63/1.0% Organic Impurities (ASTM C40) Potential Reactivity (ASTM C289) lightweight Pieces (ASTM C133) Plate No.1 Innocuous Trace Clay Lumps and Friable Particles (ASTM C142) Sodium Sulfate Soundness, cycles (ASTM C88) 0.40% 2.29% 15

PAGE 24

3.1.3 Reinforcement Fibers Fibers used for the study were 2 inches in length and approximatly 0.004 inches in diameter. Manufacturer reported fiber physical properties are specific gravity 0.9 and tensile strength ranging between 80 and 110 ksi. Fibers are composed of virgin polypropylene. Standard use for this type of fiber is to mix with concrete to reduce cracking, increase toughness and reduce permeability. During the mixing process the fibrillated fibers open into a grid configuration and mechanically reinforce the soil. 3.2 Material Mixing Samples used for all test were prepared from source containers of premixed soils. Mixed soils were prepared in quantities of 3000 grams to 7000 grams. Prior to batching, water content of each mixed soil was determined. Appropriate weights of each soil were then placed into a 5 gallon paint container. Soil were initialy dry mixed by hand and then remixed after adding 0.2% by dry weight of the sythetic fibers. Water was then added slowly with intermitent mixing and addition of small quantities of water until desired moisture content and a visually homogenous mix were obtained. The containers were then closed and manually shakened for approximately 5 minutes. Containers were reopened and any clumps 16

PAGE 25

which formed were broken up and remixed into the mass. It was noted that mixes prepared with higher water contents and greater clay content formed more clumps and required greater mixing effort. After mixing was completed containers were sealed, wrapped in a plastic bag and stored in a sample preperation room for a minimum period of twenty four hours prior to being used for test. Gradation curves for mixs AI, A2, Bl and B2 are shown in the following figures 3.1 and 3.2. 17

PAGE 26

8 3 Z 1/2 10 211 30 40 o.a, 005 0.1111 0 100 '0 211 >0 (") 0 '3 !50 50" '!1 .. .. ... 7U 10 ., 10 '00 0 '-'EOIIJ" SANO Figure 3.1 Mix Al A2 Gradation Curve 18

PAGE 27

""-R OF MESH PeR N:H. U.S. 5T_0 G_ SIZE .. "UIMI:TERS 8' 3 1/4. 0.05 O.D! O.OS 711 eo IDO SAND Figure 3.2 Mix B 1 B2 Gradation Curve 19 0.111' 100 eo .. 70

PAGE 28

4. Compaction Characteristics The purpose for running compaction tests in additon to evaluating fiber effect on compaction properties was to determine the range of moisture contents to use for preparing samples of uniform density for unconfined compression and triaxial tests. Compaction test results for test samples also provide for a discussion point of reference for how soils with and without fibers behave based on optimum moisture content. 4.1 Test Equipment The equipment used for compaction test consists of the Standard Proctor mold with a measured volume of 945.5 cm3 and a weighted 2.5 kg rammer with a measured drop of 304.5 cm. 4.2 Test Procedure The ordinary compaction method conforming to BS 1377, (1975) procedure was used for all compaction tests. After initial trial tests, it was determined that the mass of material required for filling the test mold without excessive overfill was approximately 2100 grams. Compaction of three equal weights of soil was performed in three layers using 700 grams and 25 blows of the 2.5 kg rammer per layer. The soil for all tests was obtained from source containers which had been mixed

PAGE 29

and cured a minimum of twenty hours prior to performing a test. Compacted samples were then leveled in their molds weighted and extruded. Compaction tests performed on source material stored for two months or more showed no change in densities relative to tests performed one day after source soil preperation. Therefore, length of curing time is not a consideration for evaluating differences in soil behavior for the soils studied. After compacted soil specimens were extruded, they were immediately divided and three portions weighing approximately 150 grams each were taken from the top, middle and bottom of the sample for moisture content determination. Moisture contents were determined by drying samples in a microwave oven for ten minutes. Moisture content determination using a microwave oven proved reliable in that samples remicrowaved for an additional 5 minutes showed no additional loss of weight. In addition the organic content of all soils used in sample preperation was extremely small. 4.3 Test Results A total of 21 compaction tests were performed for this study A minimum of four test were performed for each mix studied. Test results are plotted in Figures 4.1, 4.2, 4.3 and 4.4 for each mix in the form of compaction curves. Test moisture contents for Al and A2 mixes ranged between 8.2% and 13.2%. Test moisture contents for Bl and B2 mixes ranged between 11.0% and 13.2%. Lower bound moisture contents for 21

PAGE 30

13 rr, < 12 12 C Q) 0 11 o 11 13 5 rr, .::: 12 5 ..... C Q) o 11 5 / '" 8 10 11 12 13 14 8 Moisture Content -Figure 4.1 Compaction Curve -Al / 10 11 12 13 14 Moisture Content -Figure 4.2 Compaction Curve -Mix A2 22

PAGE 31

13 < 12 '-.. .c 12 t:: 0 11 11 13 5 5 12 5 '-.. :9 ..... t:: ell o 11 5 8 9 10 11 12 13 14 Moisture Content -% Figure 4.3 Compaction Curve Mix Bl 8 10 11 12 13 14 Moisture Content -Figure 4.4 Compaction Curve Mix B2 23

PAGE 32

the soils tested were established by the inability of easily obtaining reasonably homogenous test samples for the mixes containing fibers using the procedure previously described. The upper bound moisture contents for the soils tested were established based on excess clumping of materials during mixing and soil sticking to the test rammer during compaction. Test results for A mixes showed optimum moisture content occurred at about 10% for Mix Al and 10.5% for Mix A2. Test results for B mixes showed optimum moisture content to occur at about 11.5% for Mix Bl and at about 12% for Mix B2. For both Mix A soils and Mix B soils, sample densities compacted slightly wet of optimum seemed to be uneffected by the addition of fibers within the range of testing performed. For tests performed dry of, and slightly above optimum moisture content soil, mixes with fibers showed a decrease in density for similar compaction effort. Maximum sample density for both Al and A2 mixes is approximately 123 Maximum sample density for Mix Bl is approximately 122 Ibs/ft!l3 and for Mix B2 is approximately 120 Compaction test results for A mix soils show that at similar compaction effort, the maximum density obtained is constant but occurs at a higher water content for samples containing fibers. Compaction test results for B mix soils which contain a higher clay content then A mix soils show that a greater compaction effort is required to obtain similar densities. The indication is that as soil clay content increases the fiber effect on inhibiting compaction increases.

PAGE 33

5. Unconfined Compression Tests and Results The purpose in running unconfined compression tests was to determine the weight fraction of fibers and length of fibers to use for the study prior to performing more time consuming triaxial tests. Preliminary unconfined compression tests using 3/4 inch long fibers at economical weight fractions showed little strength increase for unconfined compression. Tests using 2 inch long fibers exhibited increases in unconfined compressive strength for soils tested ranging up to 50 %, therefore 2 inch long fibers were selected for all tests. In addition preliminary tests indicated that a 0.2% dry weight fraction of fibers would be adequate to increase mix strength to a magnitude high enough to establish comparative trends in soil behaviors. This percentage of fiber content agrees with results from previous studies, Grogan (1994). This percentage of fibers is proportional to about 6 pounds being added to each cubic yard of soil fill material. 5.1 Test Equipment Tests were performed in a hand operated unconfined compression apparatus with a 2000 lb capacity load frame. The load frame consisted of two platens between which the test sample was placed, a calibrated loading ring, a load dial gauge and strain dial gauge. 25

PAGE 34

5.2 Sample Preparation Specimen size used for tests was a nominal 2.8 inches in diameter by 5.8 inches in height. Specimens were prepared in a split mold cylinder placed on a standard triaxial cell base with the mold interior surface protected by a rubber membrane. The rubber membrane was secured to the inside surface of the cylinder with a vacuum prior to soil placement. Soil was placed into the mold in five layers. Each layer consisted of approximately 256 grams for all tests. Each layer was compacted using a 679 gram hand held tamper with a head diameter of 2 inches. The tamper was manually stroked 29 times with a rubber mallet for compaction of each soil layer. The number of strokes used to compact samples was determined by trial and error to obtain sample densities similar to those obtained during compaction testing. During compaction of each layer the tamper was moved around the split cylinder mold after each stroke to insure uniform compaction and to maintain a level surface for placement of the next soil layer. Each soil layer was scarified prior to placement of a succeeding layer. Periodic measurements were made of soil layer thickness to insure uniform density was obtained throughout the sample height. After compaction was completed, samples were removed from the split mold cylinder, the rubber membrane was removed and the sample height and diameter were recorded. 26

PAGE 35

5.3 Test Procedure The standard laboratory unconfined test procedure conforming to ASTM designation 02166 was used for all unconfined compression tests. A total of 13 unconfined compression tests were performed on A mix soils at two preset moisture contents of approximately 10% and 11 %. These moisture contents were chosen to be at about optimum moisture content and slightly wet of optimum as determined from compaction tests. Previous studies by Freitag (1986), had shown that samples of sandy clay compacted on the dry side of optimum exhibited little increase in unconfined compression strength. A minimum of three tests were performed on each soil mix at both moisture contents. A total of 17 tests were performed on mix B soils at preset moisture contents of approximately 11.5 and 13 percent. Again these two values were chosen to be at about and wet of optimum moisture content. Anyone particular set of three tests on a single mix type were tested on the same day and from the same source container of material. This procedure was followed to limit scatter in test results due to potential variability between mixes used to prepare samples and the speed at which testing equipment was manually operated. Upon completion of the sample preparation and after recording sample height, diameter and weight, tests were immediately performed. Tests were performed at a head platen speed which loaded the sample at a strain rate of 0.1 inch per minute. At completion of load testing, 27

PAGE 36

sample moisture content was taken and dry density was calculated. Reported dry densities are based on mineral content only. Each test sample layer measured approximately 1.2 inches in height while fiber length is 2 inches. Although scarifying each layer after compaction did cause some fibers to pass through the adjacent compacted layer interface a homogeneous fiber distribution in the test samples was not obtained. What effect this has on testing is unknown. Previous testing by Gray (1985), has shown that maximum strength increases occur in samples where reinforcement is orientated in the direction of the maximum principle tensile strain. For samples of dry sand in axial compression, this would be equivalent to fibers being initially inclined 60 degrees to the shear plane. Statistical analysis by Namaan (1988) and laboratory testing Gray ( 1986) using similar fibers has shown for randomly distributed fibers orientation of fibers is approximately 90 degrees with respect to the shear failure surface for sands. 5.4 Test Results Stress verses strain test results for Mixes AI, A2, Bland B2 are plotted and are shown in figures 5.1 through 5.4. Results show for similar soil mixes with fibers and without fibers, that compressive strengths at failure are higher for samples with lower moisture content. Increases in peak strength for Al soils and A2 soils are 50%. Increases in peak 28

PAGE 37

45 40 35 10 5 A -Mix Al Moisture Content = 1 0.2% B -Mix Al Moisture Content =10.0% C -Mix Al Moisture Content = 1 0.1 % Mix A2 Moisture Content =9.7% E -Mix A2 Moisture Content = 1 0.1 % F -Mix A2 Moisture Content = 1 0.0% G -Mix A2 Moisture Content = 10.2% o 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Strain Figure 5.1 Unconfined Compression Strength Mix Al A2 29

PAGE 38

45 40 35 '" 25 I-. 20 S o 15 10 5 A -Mix Al Moisture Content 11.3% B -Mix Al Moisture Content 11.2% C -Mix Al Moisture Content 11.0% D -Mix A2 Moisture Content 11.3% E -Mix A2 Moisture Content = 11.0% F -Mix A2 Moisture Content = 11.0% __ D C B 1 2 3 4 5 10 11 12 13 14 15 Strain % Figure 5.2 Unconfined Compression Strength Al A2

PAGE 39

55 50 45 40 20 15 10 5 A -Bl Moisture Content 11.4% B -Bl Moisture Content 11.5% C -Bl Moisture Content 11.7% D -Mix B2 Moisture Content 11.2% E -Mix B2 Moisture Content 11.3% F -B2 Moisture Content 11.4% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Strain Figure 5.3 Unconfined Compression Strength Mix Bl B2 31

PAGE 40

50.----------------------------------------------, A -Mix Bl Moisture Content =13.1% B -Mix Bl Moisture Content = 13.0% 40 35 C -Mix Bl Moisture Content = 13.2% Mix B2 Moisture Content = 12.9% E -Mix B2 Moisture Content = 12.9% F -B2 Moisture Content =13.0% 25 (Z) I-. 20 .s; o 15 10 5 B C o 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Strain Figure 5.4 Unconfined Compression Strength Mix Bl B2 32

PAGE 41

strength for B1 soils and B2 soils are also 50%. In addition for similar peak strength at failure occurred at about a 50% lower strain for samples with a lower moisture content. The following table 5.1 summarizes average test results. Table 5.1 Unconfined Compression Test Results Mix Moisture Unconfined Undrained Young's Content % Compressive Shear Modulus Strength (psi) Strength (psi) (psi) Al 10.1 27. 13.5 719. A2 10. 43. 21.5 850. Al 11.2 14. 7. 160. A2 11. 22. 11. 250. B1 11.5 39. 19.5 1460. B2 11.3 54. 27. 1460. B1 13.1 17. 8.5 170. B2 12.9 28. 14. 230. Comparison between Al and A2 soil sample test results at a moisture content of 10% shows an increase in peak strength for mix A2 of apprOximately 40%. Comparison between Al and A2 soil sample test results at 11% moisture content shows an increase in peak strength for mix A2 of 50%. Inspection of the unconfined compression stress strain plots shows that at lower moisture contents test results are much more scattered. It is believed this scatter can be attributed to the soil samples 33

PAGE 42

being more sensitive to variability in applied load and sample moisture content. Comparison between Bl and B2 soil sample test results at about 11.5% moisture content shows an increase in peak strength at failure for the B2 mix of about 30%. For Bl and B2 mixes at about a moisture content of 13% there is an increase in peak strength for mix B2 of about 65%. Comparison of test results between soils that are similar in clay content show that for these soil types in an unconfined state, fibers are proportionally more effective in soils with a higher moisture content. In addition this trend is not significantly effected by the range of clay content studied, 25% and 35% by weight, although for soils with more clay the percentage increase is higher. Stress verses strain graphs for Al and A2 soils show that at moisture contents at about optimum, the initial stiffnesses of the soils are similar during loading. At water contents above optimum, the initial stiffness of the A2 soil is greater than the Al soil. This trend is also followed for tests perfonned on Bland B2 soils. As straining increases, soils with fibers at optimum moisture content tend to hold strength longer then those without fibers. Failure strains for B2 soil samples tested at optimum moisture content are 100 percent greater then B 1 soils tested at optimum moisture content. Failure strains for A2 soil samples tested at optimum moisture content are 75 percent greater then Al soils tested at optimum moisture content. Failure strains of B2 soil samples tested wet of optimum are 25 34

PAGE 43

percent greater then BI soils tested wet of optimum. Failure strains of A2 and Al samples tested wet of optimum are proportional. In summary, stress strain graphs show that soil samples, with and without fibers, of similar clay content and at optimum moisture content, have similar stiffness at strains between 0 and 5 percent. But for these soil mixtures at optimum moisture content, those samples with fibers failed at a higher strain. When increasing the moisture content 'of similar soils with and without fibers, the samples with fibers had a higher stiffness between 0 and 5 percent strain during loading. But for similar soil samples tested at higher moisture content, the net differences in strain and stress at failure were less then those tested at optimum moisture content. The indication of the above summary is that soil density and clay content have a significant effect on how fibers strengthen soil. Possible explanation for the results obtained for dense soils is that fibers are compressed during compaction and are not mobilized during initial straining. Increasing water content increases initial void ratio and reduces compression of the fibers during compaction therefore causing them to resist early straining of the test samples. Increases in the net differences in strain at failure for soils with higher clay content indicates that bonding between the fibers and clay soil has occurred resulting in a greater modification to the soil shear failure zone development. It is believed that fibers widen the shear zone and therefore reduce the average shear strain that samples with fibers see 35

PAGE 44

verses that of a similarly loaded sample without fibers. It is not clear if upon additional loading the samples with and without fibers fail at similar average shear zone stresses. These potential fiber effects on strengthening soil will be discussed in more detail in chapter 6. 36

PAGE 45

6. Isotropically Consolidated Undrained Triaxial Test A total of 18 isotropic ally consolidated undrained triaxial tests were performed. Five tests each were perfonned for mixes Al and A2 and four tests each were performed for mixes B1 and B2. At a minimum, each soil mix was tested at three different initial effective confining pressures. Al and A2 mixes were tested at initial effective confining pressures of 10, 25,40 and 60 psi. B1 and B2 mixes were tested at initial effective confining pressures of 15, 25, and 40 psi. Mix B2 was also tested at an initial effective confining pressure of 60 psi. Some tests as noted in the following have not been recorded because it is believed that they were not fully saturated prior to testing. Past research on compacted soil samples has shown that specimen compaction method, curing time, water content, void ratio and loading rate can have significant effect on sample behavior during loading. Therefore these factors were maintained as constant as feasible during testing. 6.1 Triaxial Test Equipment Equipment used for triaxial tests included the use of two standard triaxial cells each equipped with a three way valve/transducer assembly for taking measurements of the sample top and bottom pore 37

PAGE 46

water pressure and triaxial cell confining pressure. Triaxial cells were connected to an adjustable pressure source capable of providing constant sample back pressure and cell confining pressure independently. During testing, pore pressures were recorded using a electronic data recorder at preset intervals. The data recorder was calibrated using standard pressure gauges prior to the start of each test. Loading for samples was applied at a constant rate of .002 inches per minute using a Instron Universal Testing Instrument Model TT-D-L. 6.2 Triaxial Test Sample Preparation Samples used for triaxial tests were prepared from source containers of hand mixed soil material prepared as previously described. Batching of source materials was performed 24 hours prior sample preparation. Samples used for triaxial shear strength testing of anyone mix type were taken from a single source container to eliminate possible errors in test results due to inconsistency in mix preparation and to maintain constant water content and density for the test performed. Test samples were prepared in a 2.8 inch diameter split mold cylinder. Each sample was statically compacted in five 260 gram soil layers using 29 blows per layer with a hand rammer and rubber mallet. This number of blows provided sample densities corresponding to the maximum densities determined during compaction testing. After completion of specimen compaction, samples were removed from the split mold cylinder and placed in a sample miter box to be squared and trimmed to 38

PAGE 47

approximately 2.8 inches in diameter and 5.8 inches in height. After trimming samples were weighed wrapped with filter paper and two rubber membranes, and placed in the triaxial test cell. Filter paper extended from the top of the sample to within 3/4 inches of the bottom of the sample. Filter paper was used for all tests to assist in sample saturation. Double rubber membranes were used to eliminate test failure from punctures within a single membrane caused by aggregate within the sample. In addition to placing filter paper around the sample, a porous stone and a circular piece of filter paper was placed on the top and bottom of each test specimen. The two layers of rubber membranes were secured around the sample top and bottom by use of rubber o-rings and elastic rubber bands. 6.3 Sample Saturation Triaxial test samples were saturated by applying a final net test back pressure of SO psi to each sample prior to loading. The procedure for obtaining saturation for tests was perfonned for a duration of two weeks each. After samples were initially prepared and placed in the triaxial cell, they were subjected to a 5 psi back pressure and a 10 psi confining pressure for a period of one week. During this period, air was expelled out through the top of the sample through a valve connected to the sample top. It is believed this bleeding process at low pressures enhanced the saturation procedure by maximizing the quantity of air expelled from the test samples. Following this one week period, cell 39

PAGE 48

back pressure and confining pressures were raised simultaneously in 10 psi increments and then allowed to reach equilibrium for twelve to twentyfour hours time periods. This procedure of incrementally increasing the cell confining pressure and back pressure was followed until a final back pressure of 50 psi and the test confining pressure were obtained. For tests with effective confining pressures exceeding 10 psi cell, confining pressure increases exceeding 15 psi were limited to a twelve hour period. During this period Skemptons B parameters were taken and typically ranged between 60% and 99% during the two weeks of saturation. The final confining pressure increment was performed with a final B parameter check. Final B parameters checks for all tests ranged between 90% and 100%. Samples were allowed to stabilize for a period of twelve hours prior to testing after applying the final confining pressure increment. Saturation of B mixes appeared to be more consistent then A mixes. 6.4 Sample Loading Test samples were loaded at a constant axial deformation rate of 0.002 inches per minute over periods of time ranging between 6 and 10 hours. Pore pressure readings for tests were taken at 30 second intervals. Pore pressures were measured from the bottom of all test samples. Tests for all Al and A2 mixes were performed to a minimum of 12% strain and a maximum of 25% strain. All tests for Bl and B2 mixes were performed to a minimum 20% strain and a maximum of 30% strain. 40

PAGE 49

At completion of loading samples were removed from the cell, measured, weighed and their moisture contents were taken. Sample moisture content determination consisted of testing three separate 150 gram soil quantities for each specimen from the sample top, middle and bottom. Final water contents reported are the average of the three individual values measured for the soil samples. Maximum differences for moisture contents from anyone sample was 0.4% and they typically were within 0.2% of each other. There appeared to be no trends in water contents taken from top, middle or bottom area of the samples. In addition to taking sample overall measurements and water content each sample circumference was measured at five places along its height. These measurements allowed for quantifying how the samples were straining during loading. In general maximum bulging failure for all samples occurred slightly below the center elevation of the test specimen. Typical maximum bulging diameter was 3.35 inches which is an increase in diameter of 0.45 inches over the test start. Bulging of samples was non-symmetrical in that slightly more bulging occurred in the bottom half of the test specimens. 6.5 Test Results Results for triaxial tests are plotted in several ways and are included in the following. Plots include deviator stress verses strain and pore pressure, q' verses p' and the Mohr Column failure envelope. These 41

PAGE 50

plots general show that mixes with fibers exhibit higher peak failure strength then mixes without fibers. 6.5.1 Mohr Columb Effective Stress Failure Envelope Figures 6.1, 6.2, 6.3 and 6.4 are plots of test data in the form of the Mohr Colomb failure envelope. Figures 6.1 and 6.2 represent results for Al and A2 soil mixes and show that mix A2 has an effective friction angle of 36 degrees and mix Al has an effective friction angle of 28 degrees. These figures indicate a minimal difference in cohesion c for the samples. Mix Al cohesion was 7.0 psi and mix A2 cohesion was 8.0 psi. Figures 6.3 and 6.4 represent results for Bl and B2 soil mixes and show that mix Bl has an effective friction angle of 26.5 degrees and mix B2 has an effective friction angle of 30 degrees. Figures 6.3 and 6.4 also show that the cohesion for mix Bl is about 1.5 psi while that for mix B2 is 6.5 psi. The Mohr Columb curves show there is a definite increase in the soil composite strength for both soil types. For the Al and A2 mixes which contain 25% clay, the primary increase is in the friction angle strength parameter while for the B mixes which contain 35% clay both strength parameters show increases. The net friction angle increase for B mixes is less then that for the A mixes which indicates a decrease in fiber effectiveness as clay content is increased.

PAGE 51

10 90 80 70 ..... 0. 60 '" 50 ... 40 J: V,) 30 20 10 0 A Mohr Circle MIx Al Initial Conf. Press. 10.0 psi B Mohr Circle Mix Al Initial Conf. Press. 39.66 psi C Mohr Circle Mix Al Initial Conf. Press. 61.95 psi q, 28.0 Degrees c = 7.0 psi 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Normal Stress (psi) Figure 6.1 Mix Al Undrained Triaxial Test Mohr Circles of Effective Stress 43

PAGE 52

10 90 80 70 60 '-a 40 c 30 20 10 0 A Mohr Circle Mix A2 Initial Conf. Press. 10.3 psi B Mohr Circle Mix A2 Initial Conf. Press. 25.87 psi C Mohr Circle Mix A2 Initial Conf. Press. 58.48 psi 36.0 Degrees c 8 psi 10 20 30 40 60 70 80 90 100 110 120 130 140 150 Normal Stress (psi) Figure 6.2 Mix A2 Undrained Triaxial Test Mohr Circles of Effective Stress

PAGE 53

10 90 80 70 0.. 60 (lJ 50 C/.) 40 ca (lJ C/.) 30 20 10 0 A Mohr Circle Mix B1 Initial Conf. Press. 14.77 psi B Mohr Circle Mix B1 Initial Conf. Press. 25.75 psi C Mohr Circle Mix B1 Initial Conf. Press. 42.02 psi = 26.5 Degrees c = 1.5 psi 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Normal Stress (psi) Figure 6.3 Mix B 1 Undrained Triaxial Test Mohr Circles of Effective Stress 45

PAGE 54

10 90 80 70 -..... 0.. 60 '" Q) 50 .... tr.l @ 40 Q) tr.l 30 20 10 0 A Mohr Circle Mix B2 Initial Conf. Press. 15.98 psi B Mohr Circle Mix B2 Initial Conf. Press. 25.0 psi C Mohr Circle Mix B2 Initial Conf. Press. 38.55 psi D Mohr Circle Mix B2 Initial Conf. Press. = 60.01 psi ,= 30.0 Degrees c = 6.5 psi o 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Normal Stress (psi) Figure 6.4 Mix B2 Undrained Triaxial Test Mohr Circles of Effective Stress 46

PAGE 55

Comparison of these results to previous studies is similar in that there is a clear increase in apparent friction angle for soils with fibers. Because cohesive soils have not been studied extensively, no data was found on how the cohesion strength parameter is effected by the inclusion of fibers. Maher and Gray (1986) found that there existed a critical confining pressure at which point the failure envelopes of similar soils with and without fibers have parallel failure envelopes. Initial effective confining pressures for their testing on fibers mixed with Muskegon dune sand and several fiber types did not exceed 55 psi. For tests performed in this study a maximum of 60 psi confining pressure was used and no critical confining pressure was found. The indication here is that for compacted sandy clays, the critical confining pressure if it exist occurs at a higher value then normally consolidated sand soils. This observation would appear to be in agreement with Maher's assumption that fibers slip when loaded at confining pressures less then critical in that with clay one would expect less soil bonding with fibers therefore resulting in a greater tendency for fibers to slip. It is also possible that there exist a normal consolidation magnitude factor which causes fibers to yield rather then slip during stressing. In other words, until the confining pressure reaches a magnitude high enough to increase bond strength along the soil fiber interface over the fiber strength, fiber yielding cannot occur. Additional testing on samples with varying overconsolidation values could establish this soil behavior. 47

PAGE 56

6.5.2 Deviator Stress and Pore Pressure Verses Strain Curves Inspection of the test deviator stress verses strain plots, figures 6.5 through 6.12, show for all confining pressures after initial loading the rate of deviator stress increase is greater for soils with fibers. At small strains there is little difference in the slope of the stress verses strain curve. The indication here is that initially stresses within the critical shear zone of the soils with and without fibers are the same. As the soil critical shear zone begins to yield at a faster rate with respect to applied loading the fibers begin to engage through bonding along their length and cause an apparent increase in soil strength. It is suspected that shear strength of the soil in the critical shear zone is not actually increased but that the critical shear zone is widened therefore causing an overall strength increase of the soil composite. This behavior is more clearly suggested for tests where initial pore pressure developments are identical as seen in figures 6.7,6.8, and 6.10. This behavior was suggested by Jewell and Wroth (1987). Yang (1972) has suggested that soil reinforcement fabric tends to create an increase in confining pressure within the soil composite therefore resulting in an increase in the soils apparent strength. Schlosser and Long (1974) have proposed that fabric causes a pseudo cohesion that is a function of fiber spacing and tensile strength. Although both these interpretations have been developed for layered fabric reinforced soil it is believed that the

PAGE 57

75,-----------------------------------------, 70 65 60 -55 45 .... V,) 35 .... B 30 25 20 15 10 5 Mix Al MixA2 10 15 20 Strain % 10,-----------------------------------------, 0. 5 0 -5 6: -10 MixA2 -----___ Mix Al c:: 0 5 10 15 20 Strain % Figure 6.5 Mix Al A2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure 10 psi 49 25

PAGE 58

75 65 60 =55 '" -=50 45 b 40 V,) S 35 .... 30 25 20 15 10 5 Mix Al o 5 10 15 20 25 Strain 5 15 10 MixA2 s.. Q.. ) Q.. 0 10 15 Strain Figure 6.6 20 Mix Al A2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure 25 psi 50 25

PAGE 59

85 80 75 70 65 =60 555 -0 b 45 S40 35 > 30 25 20 15 10 MixA1 MixA2 10 15 20 25 Strain =30,------------------------------------, 52 5 20 = 15 e 10 c.. e 5 MixA2 Mix Al 10 15 20 Strain Figure 6.7 Al A2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure 40 psi 51 25

PAGE 60

125 95 90 85 80 0.. 75 70 65 .... 60 0 .... 55 50 45 40 35 30 25 20 15 10 5 0 0 35 0..2 32015c:t 1050 0 0 10 15 Strain 10 15 Strain Figure 6,8 MixA2 Mix Al 20 MixA2 Mix Al 20 Mix Al A2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure = 60 psi 52 25

PAGE 61

70 65 60 _55 :; 45 If.) 35 30 .;; 25 aJ 020 15 10 MixB2 MixBl o 5 10 15 20 25 Strain S 10 5 0 aJ c.. -5 MixB2 MixBl aJ c.. 0 5 10 15 20 25 Strain Figure 6.9 Mix Bl B2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure 15 psi 53

PAGE 62

70 65 60 -.55 ';' 45 40 CI') 35 .9 30 25 Q) 20 15 10 Mix B2 Mix Bl 5 10 15 20 25 Strain % 0. -20 15 10 Q) :5:: 5 MixBl MixB2 0 10 15 20 Strain % Figure 6.10 Mix Bl B2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain CUIVes Effective Initial Confining Pressure 25 psi 54 25

PAGE 63

75 70 65 60 -.55 --; 45 B 30 ct! .S; 25 IlJ 020 15 10 5 0 0 -=20 IlJ 15 :3 10 IlJ c.. 5 IlJ 0 0 0 MixB2 Mix Bl 5 10 15 20 Strain % MixB2 MixBl 5 10 15 20 Strain % Figure 6.11 Mix B1 B2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure 40 psi 55 25 25

PAGE 64

10 95 MixB2 90 85 80 75 70 -65 Vl 0.. 60 CI) 55 CI) 50 CI:l 45 0 .... 40 35 a 30 25 20 15 10 5 0 0 10 15 20 25 Strain 525 -520-CI) 156:: 10Mix B2 : 5 10 15 20 25 Strain Figure 6.12 Mix B2 Undrained Triaxial Test Stress/Strain Pore Pressure/Strain Curves Effective Initial Confining Pressure 60 psi 56

PAGE 65

physical mechanism for both fabric and fiber reinforced soils could be similar. Triaxial test results for Al and A2 mixes indicates that there is no apparent trend in magnitude of the deviator stress verses strain at failure. Comparison of tests performed at an effective confining pressure of 10 psi show the deviator stress at failure for mix Al to be greater then mix A2. Tests performed at an effective confining pressure of 60 psi show mix A2 have a significant increase in deviator stress at 20% strain. Comparison of test excess pore pressure curves for mix Al and A2 samples show a clear trend in pore pressure development. Positive excess pore pressure for all Al mixes reaches a peak between 1.5 and 2.5 percent strain and then begins to dissipate. The rate of dissipation is greater for samples with a lesser initial effective confining pressure. Also of interest is that dissipation of pore pressure continues until test end although to a lesser rate. Positive excess Pore pressure development for A2 test samples also reach their peak between 1.5 and 2.5 percent strain. After reaching peak positive pore pressures mix A2 soil sample pore pressures decline but at a much reduced rate and magnitude then mix Al soil samples. Differences in dissipation of positive pore pressures between Al and A2 soils indicates that there is more volumetric expansion for Al soils then A2 soils. Comparison of Al and A2 void ratios after completion of test shows that Al test samples tended be more dense then A2 samples. 57

PAGE 66

This could account for the differences in pore pressure development. In addition during testing one test each for Al and A2 samples showed significant variation ultimate strength obtained and these two samples had reduced void ratios. A valid comparison can be made for mix Al and A2 test results plotted in figure 6.8. Both samples had initial effective confining pressures of 60 psi and final test void ratios of 0.28. Pore pressure dissipation for these two test did show a reduced rate and magnitude at 20% strain for the sample with fibers. Comparison of Mix Bl and Mix B2 triaxial test sample excess pore pressure curves show for tests at an initial effective confining pressure of 15 psi a trend similar to that which occurred for Al and A2 mixes. For tests performed at 25 and 40 psi initial effective confining pressures pore pressure development and dissipation are similar. In addition all tests have an initial positive excess pore pressure development with a gradual dissipation occurring until test end. Tests perfonned at higher initial effective confining pressures have increasing positive pore pressure at test end. 6.5.3 Effect of Test Initial Confining Pressure For the mix A 1 test specimen initially consolidated to a pressure of 10 psi the maximum positive pore water pressure developed at about 0.25% strain. For the mix A2 test specimen initially consolidated to 10 psi the maximum positive pore pressure developed at about 0.5% strain. For the 58

PAGE 67

mix B 1 and mix B2 test specimens initially consolidated to 15 psi maximum positive pore pressures developed at 0.75% and 2.0% respectively. For these four test specimens without fibers developed negative pore pressure at 20% strain and specimens with fibers developed positive pore pressure at 20% strain. For mix Al and A2 specimens tested between 25 psi and 60 psi initial effective confining pressures peak positive pore water pressures occurred at about 2.0% strain. For mix Bl and B2 specimens tested between 25 psi and 60 psi initial effective confining pressures peak positive pore pressures occurred between 1.5% and 3.0% strain with the test performed at higher confining pressures peaking at 3.0% strain. At test completion regardless of strain rate there is a clear trend towards specimens with fibers having a higher positive pore water pressure then similar tests performed on samples without fibers.. All tests performed except one had reduced rates of pore water pressure as soils strained with rate of reduction higher for samples without fibers. The above two trends indicate that specimens with fibers which are subjected to low confining pressures do have an affect on how soils react to low strains in the range of 0.5% and 3.0%. This affect is not obvious when reviewing stress strain plots only. In addition as confining pressures increase pore pressure development differences for soils with and without fibers is also seen at strain rates between 4% and 5% and continues until the soils reach failure. These two trends therefore indicate that fiber reinforcement within soils which fail by 59

PAGE 68

bulging have a measurable effect on how the soils dilate during shearing and that this effect is a function of the overconsolidation ratio for compacted sandy clays. The indication is that fibers provide an effective method within the soil of reducing dilation by distributing stresses to adjacent particles. Therefore, for similar soils, except for the presence of fibers, which are loaded at equal constant strain rates and under undrained conditions, it is possible for the soil without fibers to have a higher peak deviator stress at failure than the similar soil with fibers which has higher shear strength parameters. This is shown by test data in figure 6.5 and 6.7. There could be many explanations for the above observed soil behavior. During these triaxial tests the net strain over the length of anyone soil sample tested was constant with respect to time because soils were loaded at a constant strain rate. Development of pore pressure for any soil is based on its dilative characteristics and the strain within its mass. It is then acceptable to assume that strain distribution within the soil mass of those samples discussed above is varied by the presence of fibers. For these soils the mode of failure was bulging with maximum bulging strain occurring near the center. It is within this region of the sample that maximum strains occur and consequently where pore pressure development is initiated. Consider the test results plotted in figure 6.7. Initial maximum pore pressure is developed for both mix A 1 and A2 at the same strain. This indicates that net strain within the shear failure zone is equal and that

PAGE 69

both soils are deforming similarly. Because both soil samples deform similarly to a maximum value it is assumed that the void ratios within the shear failure zone are equal at the test start. This assumption is valid based on pretest moisture content checks where sample moisture contents calculated were within 1%. As constant strain (loading) is continued to be applied to each of these samples there is a noticeable difference in pore pressure development. The fiber reinforced soil (mix A2) pore pressures remain almost constant and maintain at the original positive pore pressure development. The unreinforced soil (mix AI) pore pressures begin to dissipate and continue to dissipate until the end of testing. It is during this later stage of testing that it is believed that the fibers inhibit further maximum straining or dilation within the failure zone and increase the overall width of the maximum shear stress zone by engaging the adj acent soil mass. In comparison to the above, the unreinforced sample shear zone continues to strain causing dilation and the change in pore pressure. 6.5.4 Discussion of A Coefficient Plot The A coefficient plots, figures 6.13 and 6.14 represent the pore pressure parameter at failure (Af) to the effective consolidation pressure just prior to testing. Typically the overconsolidation ratio is used in this plot. Unfortunately consolidation tests were not performed on triaxial test samples and therefore the overconsolidation ratios could not be determined. Regardless, the chart does provide information on 61

PAGE 70

1 0.9 0.8 0.7 0.6 u.. .... 0.5 CIS 0.4 E 0.3 CIS 0.2 0.1 0 -0.1 0 -0.2 ......
PAGE 71

1 0.9 0.8 0.7 0.6 0.5 C1S 0.4 0.3 C1S C1S 0.2 0.1 0 -0.1 0 '7 -0.2 <-0.3 -0.4 -0.5 70 (pore pressure at failure)/( 0 0 3 ) MixB2 MixB1 60 50 40 30 20 10 Initial Test Effective Confining Pressure (psi) Figure 6.14 Mix B 1 B2 Pore Pressure Parameter / Initial Effective Confining Pressure Curve Failure Strain 20% 63

PAGE 72

how samples with and without fibers act relative to past consolidation history. Based on the mean principle vector stress path curves figures 6.15,6.16,6.17 and 6.18 it is clear that all samples tested have to some degree been overconsolidated during sample compaction. Comparison of the A coefficient plot figure 6.14 at 20% strain for mix B1 and B2 samples shows that values of net pore pressure developed at failure when ratioed with the failure deviator stress converge at higher test initial effective confining pressures. At lower values of test initial confining pressures this ratio is significantly different. Soils without fibers have reduced pore pressure parameter coefficients. Past research has shown that pore pressure parameters values less then zero indicate a highly overconsolidated soil, and values between 0.0 and 1.0 indicate a lightly overconsolidated soil. For the special case where the pore pressure parameter is equal to zero a degree of overconsolidation proportional to 4 is indicated, Oakland (1982). Assuming these ranges hold constant for soils with and without fibers it is possible that specimens compacted with fibers resisted the development of overconsolidation and at the same time are compacted to similar void ratios. This behavior can partially be explained by previous discussions where it has been shown that soils reinforced with fibers resist the dissipation of positive pore pressure. It is also known that highly overconsolidated soils tend to develop higher negative pore pressures at failure. Review of the figures shows that specimens without fibers have a preconsolidation ratio of 4 when initial confining pressures are

PAGE 73

0 0 ---0' 15 145 75 70 65 60 55 50 45 40 35 3025 20 15 10 5 0 A -Mix Al Initial Fffeet. Conf. Press.=10 psi B -Mix Al Initial Fffeet. Conf. Press.=25 psi C -Mix Al Initial Fffeet. Conf. Press.=40 psi Mix Al Initial Fffeet. Conf. Press.=60 psi p' (psi) 01' 203' 3 Figure 6.15 Mix Al -p' : Fffeetive Stress Path Curves 65

PAGE 74

0 0 .v.; c.. 105 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 A-Mix A1. Initial Effect. Conf. Press= 10 psi B-Mix A2 Initial Effect. Conf. Press=25 psi C-Mix A1. Initial Effect. Conf. Press=40 psi D-Mix A2 Initial Effect. Conf. Press=60 si p' (psi) aI' + 3 Figure 6.16 Mix A2 -p' : q Effective Stress Path Curves 66

PAGE 75

-A Mix Bl Initial Effect. Conf. Press =15 psi B Mix Bl Initial Effect. Conf. Press.=25 psi C -Mix Bl Initial Effect. Conf. Press.=40 psi p' (psi) 3 6.17 Mix Bl -p' : q Effective Stress Path Curves 67

PAGE 76

...... 0.. 15 50 45 40 35 30 25 20 15 10 5 A -B2 Initial Effect. Conf. Press.= 15 psi B -B2 Initial Effect. Conf. Press.=25 psi C -B2 Initial Effect. Conf. Press.=40 psi B2 Initial Effect. Conf. Press.=60 psi p' (psi) 01'+203' 3 Figure 6.18 B2 -p': q Effective Stress Path Curves 68

PAGE 77

between 20 psi and 30 psi and specimen with fibers at initial confining pressure at about 10 psi. In conclusion then, all else being equal, fiber reinforced compacted sandy clay soils that are heavily overconsolidated have a much increased pore pressure parameter effect then those that are lighted overconsolidated. 69

PAGE 78

7. One-Dimensional Consolidation Test A total of four one-dimensional consolidation tests were performed, one test for each mix type. Testing for A mixes and B mixes were performed simultaneously. The purpose of performing consolidation tests was to determine if properties for mixes with fibers differed from mixes without fibers. In addition tests were performed to estimate the preconsolidation pressure at which triaxial tests were run. 7.1 Test Equipment One-dimensional consolidation test were performed using a standard fixed ring oedometer equipped with a loading yoke. The fixed ring oedometers for all test were firmly attached to a rigid table. Samples were placed between porous stones without the use of filter paper during loading. 7.2 Test Procedure Test were performed generally accordance with the procedures as outlined in ASTM 02435. 7.2.1 Sample Preparation Consolidation samples were prepared from proctor compaction mold samples. The proctor mold samples were compacted slightly wet of 70

PAGE 79

optimum to obtain soil densities similar to the soil densities used for triaxial test samples. After compaction of the proctor mold samples densities of the samples were verified and consolidation samples were prepared. Consolidation samples were prepared by slowly working the oedometer cell cuning ring into the compaction sample. Samples were then trimmed and leveled to match the cuning ring thickness. Samples were then weighed within the ring and the remaining portions of the compaction mold sample were tested for water content to establish an initial void ratio for the oedometer sample. Sample size for all test was approximately 63.5 mm in diameter and 19.0 mm in height. During testing samples were kept saturated by submerging them in tap water. 7.2.2 Sample Loading and Unloading A total of seven loading and four unloading stages were used for each test. For each loading increment the sample was allowed to consolidate for approximately twenty four hours. Loading increment pressures consisted of 500 psf, 1000 psf, 2000 psf, 4000 psf 8000 psf and 16000 psf. Consolidation readings were taken at 6, 15, and 30 seconds, 1,2,4,8, 15, 30, 60, and 120 minutes and at the the start of the next loading increment. Unloading increments consisted of 8000 psf, 4000 psf, 2000 psf, and 1000 psf. Consolidation readings for unloading were only taken at the start and end of the loading increment. 71

PAGE 80

7.3 Consolidation Test Results Test results for all four samples have been plotted in figures 7.1 through 7.4 in the form of void ratio verses log(p) were p represents the loading increment pressure. Comparison of figures 7.1 and 7.2, consolidation loading curves for mix Al and A2 test samples, can not be made. During preparation of the mix Al sample significant disturbance occurred. The percentage void ratio difference between the test sample and compaction source sample for mix Al differed by 30% with the test sample having a void ratio of 0.27 and the compaction mold sample having a void ratio of 0.36. The void ratios of the A2 test sample and compaction mold source sample were 0.29 and 0.32. It is not known what effect this may have had on the rate of consolidation for the mix Al sample. The consolidation curve suggest that the reinforcement does inhibit consolidation in that the final void ratio of the mix A2 sample is greater then mix Al for the same final consolidation pressure. Comparison of the unloading curves for figures 7.1 and 7.2 can be made. These two figures show that the rebound properties of sandy clays with 25% clay content are not Significantly effected by the reinforcement used in the study. This statement can only be made for the range of void ratios which these test were performed. Significant disturbance of the mix B 1 and B2 one-dimensional consolidation test samples did not occur. The mix BI test sample and source compaction mold sample void ratios were 0.39 and 0.36. The mix 72

PAGE 81

o .... "0 > LoadCurve Unload Curve 0.35 Cc 0.067 0.3 0.25 0.2 10 100 1000 Effective Consolidation Pressure (psi) 7.1 Mix AlOne Dimensional Consolidation Test 73

PAGE 82

.s .... tU r::.::: "0 0.4.--------------------------------------, LoadCurve Unload Curve 0.35 Cc 0.11 0.3 0.25 0.2 10 100 1000 Effective Consolidation Pressure (psi) 7.2 Mix A2 One Dimensional Consolidation Test 74

PAGE 83

.9 ..... cz::: > 0.4.--------------------------------------. -.Load Curve Unload Curve 0.35 Cc 0.097 0.3 0.25 0.2 10 100 1000 Effective Consolidation Pressure (psi) Figure 7.3 Mix Bl One Dimensional Consolidation Test 75

PAGE 84

o > ---.Load Curve Unload Curve 0.35 Cc 0.096 0.3 0.25 0.2 1 10 100 1000 Effective Consolidation Pressure (psi) 7.4 B2 One Dimensional Consolidation Test 76

PAGE 85

B2 test sample and source compaction mold sample void ratios were 0.37 and 0.36. Comparison of figures 7.3 and 7.4 show that both unloading curves and loading curves are similar. The test sample void ratio for mix Bl was 7% higher then the mix B2 test sample at the start of testing. By the end of the fourth loading increment both samples had consolidated to the same void ratio and for the remaining loading increments, void ratios remained similar. In addition both samples rebounded to a similar void ratio during unloading. Results therefore indicate for mix A 1 and A2 soils under similar boundary and test conditions there may be a difference in consolidation while for mix Bl and B2 soils there is no difference. It is possible then as clay content increases the effect of reinforcement on inhibiting consolidation lessens. This could be explained by a red uced bond between soil and reinforcement with higher clay contents. For both soils there is no apparent change in rebound properties. It was initially intended that results from consolidation test would be used for establishing the overconsolidation ratio of triaxial test samples. As a result of the apparent sample disturbance during preparation and the differences between preparation of consolidation and triaxial test samples a valid interpretation is not feasible. In addition interpretation of consolidation test on compacted samples with low void ratios and with considerable sand content is difficult. It is believed that the pre consolidation pressures for the consolidation test specimens ranges between 25 and 80 psi. Test on triaxial specimens at an initial effective

PAGE 86

confining pressure of 60 psi indicates at a minimum preconsolidation pressures for triaxial samples are at or above 60 psi. It is recommended that future studies consider using traxial consolidation test to obtain preconsolidation pressures for compacted samples of this type. 78

PAGE 87

8. Hyperbolic Stress-Strain Model The Hyperbolic Stress-Strain Model is a simplified nonlinear stressstrain relationship for soils. The model as discussed in this report describes the relationship between soil compressive strength, stress history and preloading initial stress state. The model has been adapted for use in the finite element method of analysis for soil structures. Application of the hyperbolic stress-strain relationship within the finite element method has provided a powerful method for predicting deformation characteristic of soil structures. Should fiber reinforced soil stress-strain characteristics be of the hyperbolic form their potential design use in earth structures could better be evaluated. 8.1 Hyperbolic Stress-Strain Model Development Kondner (1963) showed that the nonlinear stress-strain response of many soils, including sandy clays tested in consolidated undrained triaxial conditions could be represented by a hyperbolic stress-strain formula. The hyperbolic stress-strain formula Kondner developed was a two constant hyperbolic equation of the following form. a bE 79 . . . . . .

PAGE 88

Within this equation is the normal stress difference, is the axial strain, and a and b are the constants with values that depend on the soil initial state of stress and test conditions. Initial state of stress parameters include soil preconsolidation pressure, and soil initial effective confining pressure. Test boundry conditions include applied axial strain rate and drainage. Figure 8.1 is a typical test specimen stress-strain curve in the form of equation 8.1. From this figure and by mathematical evaluation of the limits of equation 8.1 it can be determined that the constant a is the reciprocal of the initial test tangent modulus (Ei) and b is the reciprocal of the upper limt of the stress difference which the curve approachs at infinite strain. Kondner showed that by rearranging equation 8.1 into its linear form of equation 8.2 the values of constants a and b could be determined. {8.2) Figure 8.2 is the transformed plot of Figure 8.1 in the form of equation 8.2 where parameter a is the y-intercept and parameter b is the slope of the best fit straight line through the data points. Typically, the analysis of the transformed hyperbolic stress-strain curve shows that the asymptote value of the stress difference is greater then the actual test principal stress difference at failure. To accomodate this difference within the soil model a factor called the failure ratio, is introduced. The failure ratio is the ratio of the actual test specimen

PAGE 89

o M"'Asymptote (0'1-0'3) ult lib Strain, % Figure 8.1 Hyperbolic Stress-Strain CUIVe Strain, Figure 8.2 Transformed Hyperbolic Stress Strain CUIVe 81

PAGE 90

principal stress at failure to the hyperbolic model predicted ultimate principal stress difference as indicated by equation 8.3. (01-03) f Rf (01-03)ult . . . . . '8.3) In equation 8.3 (01-03) f represents the principal stress difference at failure from the actual test and (01-03) ult represents the asymptote value from figure 8.1. Past research, as compiled by Duncan (1980), shows that soil failure ratios typically range between 0.7 and 0.95 and are independent of the soil initial confining pressure. The resulting expression for the hyperbolic model principal stress difference, after rearranging terms and substitution of the initial tangent modulus, ultimate stress difference, and failure stress difference into equation 8.1 is: (01-03) . . . (8.4) ERf ] (01-03f Within equation 8.4 both the initial tangent modulus (Ei), and the compressive strength at failure (01-03)f, are dependent on the state of stress in the soil prior to loading. For consolidated undrained triaxial test past research by Janbu (1963), as referenced by Duncan (1970), has 82

PAGE 91

shown that the initial tangent modulus is a function of the initial minor principal stress (oc). Equation 8.5 below expresses this relationship. Ei=KPa(;:) ............ In equation B.S Ei is the initial tangent modulus and oC is the test initial confining pressure. Pa is atmospheric pressure and was introduced into the equation to accommodate the use of different pressure units. Both and n are coefficients which are determined by plotting a series of test data for Ei (fig. and oC on a log-log scale plot as shown in figure On figure the coefficient is the y-intercept value and the coefficient n is the slope of the best fit straight line. For consolidated undrained triaxial tests performed at a constant rate of strain, past research, Perloff has shown that the compressive strength of soil is a function of the initial confining pressure and preconsolidation pressure prior to loading. Perloff showed that plots of a series of test data in the fonn of the over-consolidation ratio and the ratio of compressive strength at failure to the initial confining pressure were of the power function fonn. s (Ol-03)f=r(OP) oC oC

PAGE 92

......... -; "0 n= log (Ei/Pa) log 10 s:: 00 s:: cu s:: Initial Confining Figure 8.3 Effect of Initial Confining Pressure on Initial Tangent Modulus 84 10

PAGE 93

Within equation 8.6 the parameters r, s, and t are constants for a constant rate of applied strain. These three constants can be detennined by plotting the test data used to develop figure 8.4 on a log log scale plot as shown in figure 8.5 and by using linear equation 8.7. Perloff suggested evaluating the constants r, s, and t by a trial and error method. The first step in the procedure is to plot actual test data as shown in figure 8.5 for several different constant t values, until a straight line can be drawn through the data points. The fact that the data points fonn a straight line on the log-log plot is an indication that the soil behavior can be expressed by equations 8.6 and 8.7. The second step in the procedure is to detennine the constant r. The constant r is solved for by substitution of the value of 1.0 for the over-consolidation ratio into equation 8.7. The solution to equation 8.7 then shows that r is equivalent to f / oc -t. The third step is to evaluate s, which is the slope of the straight line in figure 8.5. Combination of equations 8.3, 8.4, 8.5, and 8.6 along with their related constants describe the behavior of overconsolidated soils, loaded in compression in the consolidated undrained state. The importance of these equations and constants comes from their incorporation into the 85

PAGE 94

5 ::l 4.5 e.. 4 s:: S c 3.5 s:: 0 : 3 s:: .-.. 0 .......... 0 2.5 ::l 2 ... tIS s:: 1.5 =a 0.5 / 6 16 Overconsolidation Ratio, ap/oc Figure 8.4 Effect of Overconsolidation Ratio Initial Confining Pressure on Compressive Strength 86 21

PAGE 95

c <: c 10,--------------------------------------. t Constant detennined by trial error. r y -intercept s slope Overconsolidation Ratio, op/ Figure 8.5 Trial and Error Plot for Determining Constants for Equations 8.6 8.7 87 10

PAGE 96

finite element method of analysis for the prediction of stresses or strains within a soil mass. 8.2 Transformed Hyperbolic Stress-Strain Test Results Triaxial test results for mixes AI, A2, Bl, and B2 have been evaluated and plotted in the form of figures 8.1 and 8.2. The procedure for evaluation of the test data involved plotting actual test data in the form of the ratio of axial strain verses stress difference against the axial strain. These plots are shown in figures 8.6, 8.7, 8.8, and 8.9. Review of these figures shows that there is a slight convex curve downward overall for the data. This response is typical for soft clays as noted by Konder and Duncan. The figures also show that at low strains the plotted data curves downward sharply towards the ordinate. This behavior was also noted by Kondner and Duncan. This inconsistency relative to the solution of the hyperbolic equation was believed to be caused by the difficulty and accuracy of controling and reading initial test strains and stresses. To establish a consistant method for plotting the transformed hyperbolic equation, many test results from a variety of soils have been evaluated in past research. Kondner and Duncan found that by plotting a straight line through the 70% and 95% compressive strength data points the best representation of the data was obtained. A similar procedure was also followed for the development of the transformed hyperbolic plots in this report. 88

PAGE 97

C""\ o ..!. 0.004 o '" Q) 0.003 0.001 40 psi 60 psi 0.05 0.1 0.15 Strain % Figure 8.6 Mix Al 0.2 Transformed Hyperbolic Stress-Strain Test Data Plot Transformed Straight line Plot 89 0.25

PAGE 98

0. '-.. o 0 004 :::::: C \l) 0.003 C5 0.001 oc 10 psi oc 25 psi oc = 60 psi 0.05 0.1 0.15 Strain, % Figure 8.7 MixA2 0.2 Transformed Hyperbolic Stress-Strain Test Data Plot Transformed Straight line Plot 0.25

PAGE 99

0.007.-------------------------. _0.00 0. '" !"") o '0 0.005 Col a.f C Q) 0.004 5 .... C/:I '" c 0.003 .... C/:I -a .S( 0.002 0.001 15 psi 25 psi 40 psi 0.05 0.1 0.15 Strain, % Figure 8.8 MixBl 0.2 Transformed Hyperbolic Stress-Strain Test Data Plot Transformed Straight line Plot 91 0.25

PAGE 100

0. f") o :. 0.004 af t:: 0.003 0.001 oc 15 psi oc 2S psi oc 40 psi oc 60 psi 0.05 0.1 0.15 Strain, % Figure 8.9 0.2 Transformed Hyperbolic Stress-Strain Test Data Plot Transformed Straight line Plot 0.25

PAGE 101

For soft sandy clay soils failure of soil samples occurs at very large strains. As a result during this study no soil samples actually reached failure. But all tests were carried out to a strain where the normal stress in the test specimen was increasing at a very slow rate as strain increased. Triaxial test samples for mixes Al and A2 were all tested to a minimum 16% strain and some tests were tested to 20% strain. In order to maintain a consistent reference point for data comparison, the stress value corresponding to 16% strain was chosen to represent the 95% failure point for plotting the transformed hyperbolic best fit straight line. These plots are shown in figures 8.10 and 8.11. The second point for these plots was set at the 70% stress value in relation the the 95% stress value. For mixes B1 and B2 all tests were carried to a minimum of 20% strain. The stress which occurred at 20% strain was chosen to represent the failure point of the tests. The transformed hyperbolic plots were then formed by drawing a straight line through points of stress which corresponded to 70% and 95% of the 20% strain stresses. These plots are shown in figures 8.12 and 8.13. Review of the four transformed hyperbolic figures shows that the data obtained for the mix B soils is more consistent and of greater scope then the mix A soils. The lack of consistency of calculated values for the mix A soil initial tangent moduli are believed to be caused by the use of the 16% strain stress point as the 95% compressive strength. Specifically, for the mix A2, 40 psi initial confining pressure test the 93

PAGE 102

en 0. --('0') 0 0 :::::::-en C/.) 0 c '@ ..-C/.) c; '-0 0 (01-cr3)ult (01-03) 03 a Et=l/a 0.005 (psi) ( l/psi) (psi) ( l/(psi) (psi) (psi) 10 0.00030 3341 0.0143 70 62 0.88 40 0.00027 3708 0.0105 95 82 0.86 60 0.00015 6480 0.0107 94 86 0.92 0.004 Average 0.89 0.003 03 = 10psi 0.002 03 = 40psi 03 = 60psi 0.001 0.05 0.1 0.15 Strain, % Figure 8.10 Al 0.2 Transformed Hyperbolic Stress-Strain Plot 0.25

PAGE 103

0. -('f'l c ,3 --CJ.) '--'--0 'OJ 0 0 .... '-.... 0 0 .... c:::: 0.006-r------------------------, (01-03)ult (01-03)f 0.005 (psi) ( l/psi) (psi) (l/(psi) (psi) (psi) 10 0.00037 2715 0.0156 64 56 0.88 40 0.00019 5171 0.0129 78 71 0.91 60 0.00020 4910 0.0073 136 116 0.85 0.004 Average Rf 0.88 0.003 10psi 40psi 0.002 60psi 0.001 0.05 0.1 0.15 Strain, % Figure 8.11 MixA2 0.2 Transformed Hyperbolic Stress-Strain Plot 95 0.25

PAGE 104

en 0. --,.... 0 0 -Co> '" '" IJ) '-.... 0 S IJ) 0 0 .... c '-(/.) ...... 0 0 .... 0.007 (01-03)ult (01-03)[ (psi) (l/ psi) (psi) (l/ (psi) (psi) (psi) 15 0.00078 1280 0.0230 43 38 0.88 25 0.00047 2113 0.0224 45 41 0.91 0.006 40 0.00030 3354 0.0169 59 56 0.95 Average Rf 0 .91 15psi 0.005 25psi 0.004 40psi 0.003 0.002 0.001 0.05 0.1 0.15 Strain, % Figure 8.12 0 2 Transformed Hyperbolic Stress-Strain Plot 0.25

PAGE 105

0. --a .-.. -"" .... en 0 .... 0 c .... en -; 0 .S c:::: 0.007,------------------------, 0.006 oC a Et=l!a (ol-03)ult (ol-03)f (psi) ( lIpsi) (psi) ( lI(psi) (psi) (psi) 15 0.00101 991 0.0156 64 49 0.77 25 0.00065 1527 0.0136 73 60 0.82 40 0.00052 1914 0.0121 82 69 0.84 0.005 60 0.00033 3007 0.0094 107 93 0.87 Average 0.83 0 .004 15 psi OC 25psi 0.003 OC 40psi oC 60 psi 0.002 0.001 0.05 0.1 0.15 Strain, % Figure 8.13 MixB2 Transformed Hyperbolic Stress-Strain Plot 97 0.2 0.25

PAGE 106

70% stress value is low. As seen on figure 6.9 of section 6.5.2, this point ends up on the lower portion of the stress-strain curve and has a higher stress-strain rate then the other mix A2 test. Although there is inconsistency in the mix A data, the trend in increasing initial tangent modulus with increasing confining pressure is evident. Also the value of for the test on anyone mix is constant and is within 5% of the average value which shows is independent of the confining pressure. For consideration of how mix A soils with and without fibers behave, it is interesting to note that the intitial tangent modulus of soil without fibers is higher then soil with fibers. Review of figures 8.12 and 8.13, which are mix B test results, shows the data to be very consistent. For all tests perfonned for soil samples with and without fibers the initial tangent modulus increases with increasing confining pressure. Also the initial tangent modulus for soils with fibers is consistently less then similar tests on soils without fibers. The average ratio of failure compressive strength to ultimate compressive strength are generally within 5% of individual test values, although it is noted that for mix B2 there is an increasing trend. This trend is believed to be caused by the varying slopes of the actual test stress-strain curves at 20% strain. For comparison of soils with and without fibers, the test data shows that specimens with fibers have a lower initial tangent modulus. Also the coefficient for specimens with fibers is lower then those without fibers. This could be expected because the rate of stress increase at the 98

PAGE 107

20% strain was greater for mix B2 specimens when compared to mix B1 specimens. This can be seen in figures 6.13 thru 6.16 as the slope of the stress-strain curve is greater for samples with fibers. The agreement of the transformed hyperbolic stress-strain plots with actual test data is acceptable within the limits of test performed for this study. Therefore it is believed that the hyperbolic stress-strain relationship of equation 8.1 can be used to evalute the stress-strain characteristics of fiber reinforced soils. It is also felt that with additional research on a broader scope, the agreement between actual test and the model could be enhanced. 8.3 Initial Tangent Modulus Test Results Because of the limited amount of test data involving mix A soils, the following discussion will be limited to the mix B soils. Figures 8.14 and 8.15 are log-log plots of the ratio of the initial tangent modulus to atmospheric pressure verses the ratio of initial effective confining pressure to atmospheric pressure. Both plots for mix Bland mix B2 soils show that the data points form a straight line. Therefore equation 8.5 of the hyperbolic model is adequate to describe the relationship between the initial tangent modulus and the initial effective confining pressure. These two plots show, as in the previous discussion, that the initial tangent modulus of soils with fibers is less then soils without fibers. 99

PAGE 108

"'"5 "'0 o ..... Q) OJ:) K 66.9 n 0.72 Ei 66.9(ac/Pa)"72 "'-.2 "'0 ..... Q) OJ:) 10 ..... :s 1 1 1 Initial Confining Pressure/Pa Figure 8.14 MixB2 K 85.53 n .98 .98 Ei/Pa = 85.S3pc/Pa) Initial ConfIning Pressure/Pa Figure 8.15 MixB1 100 10 10

PAGE 109

8.4 Compressive Strength Test Results The final component of the hyperbolic stress-strain model remaining to be discussed is whether the failure compressive strength of soils with fibers studied in this report, can be described by the non-linear power function of equations 8.6 and 8.7. These two equations relate soil compressive strength to the initial state of stress in the soil. The initial soil stress properties of concern are the initial confining pressure and preconsolidation pressure prior to soil loading. Because for this study the preconsolidation pressure has not been directly determined, a value must be extrapolated using the pore water pressure parameter at failure, Af. Head (1986) reported a range of Af values for sandy clays with a overconsolidation ratio of 1 to be 0.25 to 0.75. Also Perloff (1976) noted that when the overconsolidation ratio is equivalent to 4, Af O. Review of figure 6.18 shows that Af 0.25 for tests performed at initial effective confining pressures of 60 psi and Af 0 for test performed at initial effective confining pressures of 15 psi. This data indicates that the preconsolidation pressure for the soils tested is 60 psi or greater. To facilitate this discussion, a preconsolidation pressure of 60 psi has been assumed for all mix B tests performed. The lack of a 60 psi confining pressure data point for the mix B1 set of tests precludes the development of a resonable power function curve. Therefore only the mix B2 power function curve has been developed herein. Figure 8.16 is a plot of the mix B2 data in the generic power function form of figure 8.4. Figure 8.17 is a plot of the mix B2 data in 101

PAGE 110

I-, I-, 0... 00 C C C C 0 :.......... -"-0 1-, ..... .... v u C I-, I-, .... C/:) 5 4 3 u 0 o / 357 Overconsolidation Ratio, Figure 8.16 MixB2 Effect of Overconsolidation Ratio Initial Confming Pressure on Compressive Strength 9

PAGE 111

OJ ::l OJ e.. .5 c c '"' '-' .... OJ r"\ 0 ::l 0 0 OJ C OJ OJ V'.l 10 1 1 =-2 r=3.46 s .282 op 60 psi Overconsolidation Ratio, op/ oc Figure 8.17 MixB2 Trial and Error Plot for Determining Constants for Equations 8.6 8.7 10

PAGE 112

the linear form of equation 8.7. The fact that the data can be fitted to a straight line in figure 8.17 is an indication that the power function form of equation 8.6 is valid for describing the relationship between the compressive strength, overconsolidation ratio, and initial effective confining pressure of the mix B2 fiber reinforced soil. The solutions for the parameters t, r, and s were determined as outlined in section 8.1. To better visualize the accuracy of equation 8.6, the calculated values for t, r, and s, along with the overconsolidation ratios and confining pressures were substituted into equation 8.6. By rearranging equation 8.6, the compressive strengths for the mix B2 test were then evaluated and are as reported in table 8.1. Table 8.1 Mix B2 Comparison of Test Compressive Strength to Power Function (Eq 8.6) Calculated Compressive Strength Power Function Parameters, t=-2, r=3.46, Test Specimen Preconsolidation Pressure = 60psi Initial Effective Test Compressive Eq. 8.6 Calculated Confining Pressure Strength Compressive Strength oc (psi) (01-03) f (psi) (01-03)f (psi) 15.98 49.36 48.31 25.0 60.34 60.72 38.55 69.02 74.01 60.10 92.26 87.60 104

PAGE 113

Comparison of calculated values and test values for compressive strengths in table 8.1 are within 5% or less of each other. Therefore equation 8.6 is acceptable for determining failure compressive strength. 8.5 Hyperbolic Stress-Strain Path The hyperbolic stress-strain path can be determined by combining equations 8.4, 8.5 and 8.6. Equation 8.5 is substituted directly into equation 8.4 while equation 8.6 is solved for the compressive strength and then substituted into equation 8.4. The resulting equation is: . (8.8) Rf The applicability of the hyperbolic model to fiber reinforced soils can be established by comparing test stress-strain paths to the predicted model stress-strain paths. Figures 8.18 and 8.19 are such plots of two mix B2 tests at 25 psi and 60 psi initial effective confining pressures. Comparison of the test stress-strain paths and predicted model stressstrain paths in figures 8.18 and 8.19 shows that at low strains the stiffness of the actual test sample is greater then the predicted stiffness. Beyond the 4% strain the two stress paths converge to within 10% of 105

PAGE 114

10 90 80 0. ("') c 70 (1)-60 s:: (1) '"' 50 a 40 '"' tZl 30 20 10 0 0 Eq. 8,8 Parameter values 25 psi 66.9 n 0.72 Pa 14.7 psi r=3.46 =-2 s = 0.282 ap 60 psi 0.83 ---Test Data -Hyperbolic Model 0.05 0.1 0.15 Axial Strain, E % Figure 8.18 0.2 Comparison of Test Stress-Strain Curve vs. Hyperbolic Model Stress-Strain Curve (Eq. 8.8) 106 0.25

PAGE 115

10 90 80 '" 0. C'<') 0 70 a.r 60 t::
PAGE 116

each other. The stress-strain path convergence for the test performed at the lower initial effective pressure is somewhat better then at the higher value. To evaluate the effect of the developed initial tangent modulus determined from figure 8.13, a second set of moduli were developed for the mix B2 test. For this second set of moduli the lower stress point on the transfonned hyperbolic plots were set at 3% strain and new moduli were calculated. This set of initial tangent moduli were then used to redevelop equation 8.S and its associated constants. Results from equation 8.S and the previous results determined for equation 8.6 were then used in equation 8.8 fonn a second set of stress-strain paths. These two plots are shown in figures 8.20 and 8.21. Review of figures 8.20 and 8.21 shows that the initial stiffness of the soil for both the test and the overall stress-strain curves to be much closer. It is interesting to note that the initial tangent moduli determined for the second set of curves were very similar to those for the mix B1 test. These results again indicate that the use of the hyperbolic stressstrain fonnula for modeling the behavior of these soils, for the range of tests perfonned, is acceptable.

PAGE 117

10 90 80 P.. 0 70 ,S. 60 c 50 40 30 20 10 0 0 Eq. 8.8 Parameter values crC = 25 psi = 93.3 = 0.72 Pa = 14.7 psi r=3.46 =-2 s = 0.282 crp = 60 psi 0.93 Ei = 2004.4 psi -Test Data Hyperbolic Model 0.05 0.1 0.15 Strain, % Figure 8.20 0.2 Comparison of Test Stress-Strain Curve vs. Hyperbolic Model Stress-Strain Curve (Eq. 8.8) Revised for Ei (3% strain) 109 0.25

PAGE 118

10 90 80 0. --. 0 70 60 Cl) Cl) 1-0 SO i5 Cl) 40 1-0 ..... 30 20 10 0 0 0.05 -Test Data Hyperbolic Model Eq. 8.8 Parameter values = 60 psi K = 93.3 n = 0.72 Pa = 14.7 psi r=3.46 t =-2 s = 0.282 ap = 60 psi = 0.93 = 3765. psi 0.1 0.15 Axisl Strain, Figure 8.21 MixB2 0.2 % Comparison of Test Stress-Strain Curves vs. Hyperbolic Model Stress-Strain Curve (Eq. 8.8) Revised for (3% strain) 110 0.25

PAGE 119

9. Summary, Conclusions and Recommendations 9.1 Summary A series of standard laboratory tests were perfonned on compacted sandy clay soils mixed with 2.0 inch long synthetic fibers for this study. Mix A soils consisted of 25% clay and 75% sand and Mix B soils consisted of 35% clay and 65% sand. Tests performed included one dimensional consolidation, standard compaction, unconfined compression, and undrained triaxial shear strength tests. Triaxial shear strength tests were perfonned with a range of initial effective confining pressures. Soil sample responses to compaction and loading were measured and graphed in standard fonns to provide a basis for evaluating how sandy clays with and without fibers react. 9.2 Conclusions The tests results showed that for the type of synthetic fibers used, significant response effects occured on sandy clay soil during loading and at failure. The following are the main conclusions of this study. 1. The fibers cause changes in sandy clay soil behavior during loading. One of the key effects being pore water pressure development during loading. 2. Unconfined compression strength tests showed a 50% increase in undrained shear strength for soil mixed with fibers as compared to soil without fibers. Also these tests showed an 111

PAGE 120

increase in stiffness for fiber reinforced soil at low strains for moisture contents above optimum. 3. The presence of fibers increased the Mohr Coulomb shear strength parameters. 4. Common soil analysis methods can be used for evaluating fiber reinforced soil response to loading. 5. The hyperbolic model can be used to predict fiber reinforced soil response to static loading. 9.3 Recommendations This study, although broad in scope is limited in the number and range of tests perfonned. Future research should include the following: 1. Focus in on the use of fibers in specific earth structures. 2. Simultaneous, combined laboratory tests and full scale field studies. 3. Apply higher back pressures to triaxial tests to enhance saturation of tests samples. 4. Triaxial consolidation tests should be perfonned to establish overconsolidation ratios. 5. Triaxial tests should be perfonned over a wider range of initial effective confining pressures. 6. Carefully monitor triaxial test sample void ratios 7. Perfonn dynamic soil tests to determine effects pore pressure differences have on fiber reinforced soil shear strength.

PAGE 121

A 113

PAGE 122

Photograph # 1 Compaction Test Equipment

PAGE 123

Photograph #2 Mix A2 Compaction Test Sample Moisture Content 11 115

PAGE 124

Photograph #3 Mix A2 Compaction Test Sample Moisture Content 12%

PAGE 125

Photograph #4 Mix A2 Unconfined Compression Test Sample in Loading Test Equipment

PAGE 126

Photograph #5 Mix A2 Unconfined Compression Test Moisture Content 10% Strain at Test End 10%

PAGE 127

Photograph #6 Triaxial Test Loading and Data Recording Equipment 119

PAGE 128

Photograph #7 Triaxial Test Load Cell 120

PAGE 129

Photograph #8 Mix B2 Triaxial Test Sample in Load Cell during Test 121

PAGE 130

Photograph #9 Mix B2 Triaxial Test Sample after removal from Cell Moisture Content 13.11% Initial Confining Pressure 15 psi Strain 21% 122

PAGE 131

Photograph # 1 0 B2 Triaxial Test Sample after removal from Cell Moisture Content 13.11% Initial Confining Pressure 15 psi Strain 21% 123

PAGE 132

Photograph #11 Dryed Triaxial Tests and Compaction Tests Samples 124

PAGE 133

Photograph #12 Consolidation Test Equipment and Lab Technician (Jeff) 125

PAGE 134

Appendix B Mix Al A2 Tests Data 126

PAGE 135

Mix Al Triaxial Test #2 Pata Sample back pressure at start of test 46.82 psi Sample confining pressure at start of test 71.81 psi Sample initial effective confining pressure at start of test 25.01 psi Sample diameter at start of test 2.80" Sample height at start of test 5.83" Sample weight at start of test 2.90 lbs Sample moisture content at start of test 11.02% B parameter at start of test 0.92 Sample diameter A at end of test 2.89" Sample diameter B at end of test 2.99" Sample diameter C at end of test 3.07" Sample diameter P at end of test 3.05" Sample diameter E at end of test 2.95" Sample diameter F at end of test 2.87" Sample height at end of test 5.11" Sample weight at end of test 2.89 lbs Sample moisture content at end of test 10.26% Sample void ratio at end of test 0.271 Net change in pore pressure at end of test 7.09 psi Sample strain at end of test 12.3% Test Sample Bevation at end of Test 127 1.11" 1" 1" 1" 1" Pia. A Pia. B Pia. C Pia. P Pia. E Pia. F

PAGE 136

Mix A 1 Triaxial Test #3 Data Sample back pressure at start of test 48.21 psi Sample confining pressure at start of test 58.21 psi Sample initial effective confining pressure at start of test 10.0 psi Sample diameter at start of test 2.81" Sample height at start of test 5.59" Sample weight at start of test 2.79 lbs Sample moisture content at start of test 10.32% B parameter at start of test .64 Sample diameter A at end of test 2.85" Sample diameter B at end of test 2.99" Sample diameter C at end of test 3.23" Sample diameter D at end of test 3.35" Sample diameter E at end of test 3.17" Sample diameter F at end of test 2.95" Sample height at end of test 4.61" Sample weight at end of test 2.82 lbs Sample moisture content at end of test 11.44% Sample void ratio at end of test 0.3009 Net change in pore pressure at end of test -12.87 psi Sample strain at end of test 18.1 % Test Sample Elevation at end of Test 128 0.61" 1" 1" 1" 1" Dia. A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 137

Mix A 1 Triaxial Test #4 Pata Sample back pressure at start oftest 47.89 psi Sample confining pressure at start of test 109.84 psi Sample initial effective confining pressure at start of test 61.95 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.772 lbs Sample moisture content at start of test 9.9% parameter at start of test 0.82 Sample diameter A at end of test 2.95" Sample diameter B at end of test = not recorded Sample diameter C at end of test 3.35" Sample diameter P at end of test 3.54" Sample diameter E at end of test 3.31" Sample diameter F at end of test 2.97" Sample height at end of test 4.17" Sample weight at end of test 2.7841bs Sample moisture content at end of test 10.95% Sample void ratio at end of test 0.289 Net change in pore pressure at end of test 21.48 psi Sample strain at end of test 25.8% Test Sample Bevation at end of Test 129 0.17" 1" 1" 1" 1" Pia. A Pia. B Pia. C Pia. P Pia. E Pia. F

PAGE 138

Mix A 1 Triaxial Test #5 Data Sample back pressure at start of test 44.70 psi Sample confining pressure at start of test 84.36 psi Sample initial effective confining pressure at start of test 39.66 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.783 lbs Sample moisture content at start of test 9.9% B parameter at start of test 0.88 Sample diameter A at end of test 2.91" Sample diameter B at end of test 3.11" Sample diameter C at end of test 3.35" Sample diameter D at end of test 3.29" Sample diameter E at end of test 3.05" Sample diameter F at end of test 2.87" Sample height at end of test 4.53" Sample weight at end of test 2.806 lbs Sample moisture content at end of test 10.90% Sample void ratio at end of test 0.287 Net change in pore pressure at end of test 7.71 psi Sample strain at end of test = 20.2% Test Sample Bevation at end of Test 130 0.53" I" I" I" 1" Dia. A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 139

Mix A2 Triaxial Test #1 Data Sample back pressure at start of test 52.45 psi Sample confining pressure at start of test 62.75 psi Sample initial effective confining pressure at start of test 10.3 psi Sample diameter at start of test 2.81" Sample height at start of test 5.83" Sample weight at start of test 2.811 lbs Sample moisture content at start of test 9.94% B parameter at start of test 0.84 Sample diameter A at end of test not recorded Sample diameter B at end of test not recorded Sample diameter C at end of test not recorded Sample diameter P at end of test not recorded Sample diameter E at end of test not recorded Sample diameter F at end of test not recorded Sample height at end of test 4.88 Sample weight at end of test 2.862 lbs Sample moisture content at end of test 11.91 Sample void ratio at end of test 0.314 Net change in pore pressure at end of test 0.60 psi Sample strain at end of test = 16.3% Test Sample Bevation at end of Test 131 0.88" I" Pia. A Pia. Dia. C Dia. D Pia. E Dia. F

PAGE 140

Sample back pressure at start of test 53.49 psi Sample confining pressure at start of test 79.38 psi Sample initial effective confining pressure at start of test 25.87 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.74 lbs Sample moisture content at start of test 10.54% B parameter at start of test 0.82 Sample diameter A at end of test not recorded Sample diameter B at end of test not recorded Sample diameter C at end of test not recorded Sample diameter 0 at end of test not recorded Sample diameter E at end of test not recorded Sample diameter F at end of test not recorded Sample height at end oftest 4.43" Sample weight at end of test 2.780 lbs Sample moisture content at end of test 11.95% Sample void ratio at end of test 0.315 Net change in pore pressure at end of test 12.49 psi Sample strain at end of test 21.3% Test Sample Bevation at end of Test 0.43" I" I" I" I" Dia. A Dia. Dia. C Dia.D Dia. E Dia. F

PAGE 141

Sample back pressure at start oftest 49.50 psi Sample confining pressure at start of test 88.50 psi Sample initial effective confining pressure at start of test 39.0 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.709 lbs Sample moisture content at start of test 9.98% B parameter at start of test 0.84 Sample diameter A at end of test not recorded Sample diameter B at end of test 2.99" Sample diameter C at end of test 3.23" Sample diameter D at end of test 3.23" Sample diameter E at end of test 3.19" Sample diameter F at end of test 2.95 Sample height at end oftest 4.65" Sample weight at end oftest 2.755 lbs Sample moisture content at end of test 11.92% Sample void ratio at end of test 0.314 Net change in pore pressure at end of test 25.17 psi Sample strain at end oftest 19.7% Test Sample Bevation at end of Test 133 0.65" I" I" I" I" Dia. A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 142

Mix A2 Triaxial Test #5 Data Sample back pressure at start of test 38.42 psi Sample confining pressure at start of test 96.80 psi Sample initial effective confining pressure at start of test 58.48 psi Sample diameter at start of test 2.81" Sample height at start of test 5.67" Sample weight at start of test 2.779 lbs Sample moisture content at start of test 9.37% B parameter at start of test 0.87 Sample diameter A at end of test 3.01" Sample diameter B at end of test 3.17" Sample diameter C at end of test 3.37" Sample diameter D at end of test 3.35" Sample diameter E at end of test 3.19" Sample diameter F at end of test 2.99 Sample height at end of test 4.33" Sample weight at end of test 2.806 lbs Sample moisture content at end of test 10.89% Sample void ratio at end of test 0.287 Net change in pore pressure at end of test 27.6 psi Sample strain at end of test 24.7% Test Sample Bevation at end of Test 134 0.33" 1" 1" 1" 1" Dia.A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 143

Unconfined Compression Tests Mix Al A2 Sample Data Sample Height Weight Area Moisture (in) (lbs) (in 2 ) Content A1#1 5.77 2.815 6.13 10.17 A1#2 5.73 2.804 6.14 10.01 A1#3 5.79 2.815 6.01 10.11 A1#4 5.71 2.816 6.19 11.31 A1#5 5.70 2.809 6.19 11.21 A1#6 5.72 2.818 6.22 10.99 A2#1 5.84 2.799 6.22 10.41 A2#2 5.84 2.813 6.14 10.40 A2#3 5.89 2.816 6.17 10.2 A2#4 5.79 2.813 6.14 9.67 A2#5 5.77 2.808 6.19 11.27 A2#6 5.74 2.817 6.19 11.03 A2#7 5.75 2.825 6.20 10.98 135

PAGE 144

Appendix C Mix B2 Tests Data 136

PAGE 145

Mix Bl Triaxial Test #1 Data Sample back pressure at start of test 49.93 psi Sample confining pressure at start of test 64.70 psi Sample initial effective confining pressure at start of test 14.77 psi Sample diameter at start of test 2.81" Sample height at start of test 5.67" Sample weight at start oftest 2.763 Ibs Sample moisture content at start of test 12.33% B parameter at start of test 1.0 Sample diameter A at end of test not recorded Sample diameter at end of test 2.91" Sample diameter C at end of test 3.13" Sample diameter D at end of test 3.33" Sample diameter E at end of test 3.25" Sample diameter F at end of test = not recorded Sample height at end of test 4.59" Sample weight at end of test 2.782 lbs Sample moisture content at end of test 12.95% Sample void ratio at end of test 0.342 Net change pore pressure at end of test -4.05 psi Sample strain at end of test 21.2% Test Sample Bevation at end of Test 137 0.59" 1" 1" 1" 1" Dia. A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 146

Mix Bl Triaxial Test #2 Data Sample back pressure at start of test 50.6 psi Sample confining pressure at start of test 76.35 psi Sample initial effective confining pressure at start of test 25.75 psi Sample diameter at start of test 2.81" Sample height at start oftest 5.65" Sample weight at start oftest 2.737 lbs Sample moisture content at start of test 12.39% B parameter at start of test = 0.87 Sample diameter A at end of test 2.85" Sample diameter B at end oft est 3.01" Sample diameter C at end of test 3.21" Sample diameter D at end of test 3.33" Sample diameter E at end of test 3.19" Sample diameter F at end of test 2.99" Sample height at end of test 4.54" Sample weight at end of test 2.760 lbs Sample moisture content at end of test 13.1% Sample void ratio at end of test 0.345 Net change in pore pressure at end of test 5.32 psi Sample strain at end of test 21.2% Test Sample Bevation at end of Test 138 0.54" I" 1" I" I" Dia. A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 147

Mix Bl Triaxial Test #4 Data Sample back pressure at start of test 44.3 psi Sample confining pressure at start of test 86.32 psi Sample initial effective confining pressure at start of test 42.02 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.738 lbs Sample moisture content at start of test 12.0% B parameter at start of test 1.0 Sample diameter A at end of test 2.87" Sample diameter B at end of test 2.95" Sample diameter C at end of test 3.25" Sample diameter D at end oftest 3.33" Sample diameter E at end of test 3.10" Sample diameter F at end of test 2.93" Sample height at end oftest 4.53" Sample weight at end of test 2.748 lbs Sample moisture content at end of test 12.71 % Sample void ratio at end of test 0.335 Net change in pore pressure at end of test 11.15 psi Sample strain at end of test 21.3% Test Sample Bevation at end of Test 139 0.53" 1" 1" 1" 1" Dia. A Dia. B Dia. C Dia. D Dia.E Dia. F

PAGE 148

Mix Triaxial Test #1 Data Sample back pressure at start of test 50.29 psi Sample confining pressure at start of test 66.27 psi Sample initial effective confining pressure at start of test 15.98 psi Sample diameter at start of test 2.81" Sample height at start of test 5.65" Sample weight at start of test 2.736 lbs Sample moisture content at start of test 12.78% B parameter at start of test 1.0 Sample diameter A at end of test not recorded Sample diameter B at end of test 3.07" Sample diameter C at end of test 3.19" Sample diameter D at end of test 3.26" Sample diameter E at end of test 3.21" Sample diameter F at end of test not recorded Sample height at end of test 4.51" Sample weight at end of test 2.752 lbs Sample moisture content at end of test 13.11 % Sample void ratio at end of test 0.346 Net change in pore pressure at end of test 3.22 psi Sample strain at end of test 21.2% Test Sample Bevation at end of Test 140 0.51" 1" 1" 1" 1" Dia. A Dia. Dia. C Dia. D Dia. E Dia. F

PAGE 149

Mix Triaxial Test Data Sample back pressure at start of test 50.2 psi Sample confining pressure at start of test 75.2 psi Sample initial effective confining pressure at start of test 25.0 psi Sample diameter at start of test 2.81" Sample height at start of test 5.65" Sample weight at start of test 2.759 lbs Sample moisture content at start of test 12.66% parameter at start of test = 0.94 Sample diameter A at end of test not recorded Sample diameter B at end of test 3.09" Sample diameter C at end of test 3.23" Sample diameter D at end of test 3.31" Sample diameter E at end of test 3.17" Sample diameter F at end of test = not recorded Sample height at end of test 4.53" Sample weight at end of test 2.768 lbs Sample moisture content at end of test 12.79% Sample void ratio at end of test 0.337 Net change in pore pressure at end of test 5.25 psi Sample strain at end of test 21.4% Test Sample Bevation at end of Test 141 0.53" I" I" Dia. A Dia. Dia. C Dia. D Dia. E Dia. F

PAGE 150

Mix Triaxial Test #3 Data Sample back pressure at start of test 50.33 psi Sample confining pressure at start of test 88.88 psi Sample initial effective confining pressure at start of test 38.55 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.742 lbs Sample moisture content at start of test 12.74% B parameter at start of test not recorded Sample diameter A at end of test 2.91" Sample diameter B at end of test 3.09" Sample diameter C at end of test 3.23" Sample diameter D at end of test 3.23" Sample diameter E at end of test 3.13" Sample diameter F at end of test 2.95" Sample height at end of test 4.57" Sample weight at end of test 2.748 lbs Sample moisture content at end of test 12.70% Sample void ratio at end of test 0.334 Net change in pore pressure at end of test 16.32 psi Sample strain at end of test 21.2% Test Sample Bevation at end of Test 142 0.57" 1" 1" 1" 1" Dia. A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 151

Mix Triaxial Test #4 Data Sample back pressure at start of test 38.45 psi Sample confining pressure at start of test 98.44 psi Sample initial effective confining pressure at start of test 60.01 psi Sample diameter at start of test 2.81" Sample height at start of test 5.63" Sample weight at start of test 2.732 lbs Sample moisture content at start of test 12.52% B parameter at start of test 0.95 Sample diameter A at end of test -sample height 3.98" Sample diameter B at end of test 3.11" Sample diameter C at end of test 3.44" Sample diameter D at end of test 3.60" Sample diameter E at end of test 3.41" Sample diameter F at end of test 3.07" Sample height at end of test 3.98" Sample weight at end of test 2.729 lbs Sample moisture content at end of test 12.3% Sample void ratio at end of test 0.324 Net change in pore pressure at end of test 24.68 psi Sample strain at end of test 21.2% Test Sample Bevation at end of Test 143 0." 0.98" I" I" I" Dia.A Dia. B Dia. C Dia. D Dia. E Dia. F

PAGE 152

Unconfined Compression Tests Mix B1 B2 Sample Data Sample Height Weight Area Moisture (in) (lbs) (in2 ) Content B1#1 5.79 2.799 6.19 11.56 B1#2 5.78 2.799 6.19 11.70 B1#3 5.7 2.80 6.19 11.79 B1#4 5.83 2.804 6.16 13.18 B1#5 5.84 2.816 6.25 12.96 B1#6 5.86 2.823 6.18 13.09 B1#7 5.93 2.80 6.18 11.40 B1#8 5.93 2.807 6.18 11.47 B1#9 5.93 2.78 6.18 11.75 B2#1 5.84 2.799 6.18 13.02 B2#2 5.81 2.805 6.17 12.93 B2#3 5.80 2.797 6.16 12.50 B2#4 5.85 2.789 6.21 11.39 B2#5 5.84 2.788 6.20 11.29 B2#6 5.81 6.23 11.14 B2#7 5.78 2.797 6.20 11.41 B2#8 5.80 2.806 6.19 11.30 144

PAGE 153

References 1. Andersland, O. B., and Khattak, A. S., (1979), "Shear stength of Kaolinite/fiber soil mixtures", R. Int. Renforcement des Sols., Paris 2. Arenicz, R. M., and Chowdhury, R. N., ( 1988), "Laboratory Investigation of Earth Walls Simultaneously Reinforced by Strips and Random Reinforcement," Geotechnical Testing Journal, GTJODJ, Vol. 11, No.4, Dec., pp. 241-247. 3. Austin, D. N., Shrader, S. L, and Chill, D. S., (1993), "Soil Stabilization with discrete fibrillated polypropylene fibers," Geotechnical Fabrics Report, Oct. 4. Crockford, W. W., Grogan, W. P., and Chill, D. S., (1993), "Strength and life of Stabilized Pavement Layers Containing Fibrillated Polyolefin" Paper Presented at 1993 meeting of Transportation Research Board, Texas Transportation Institute, College, Tx. 5. Ducan, M., and Chang, (1970), "Nonlinear Analysis of Stress and Strain In soils," Journal of the Soil Mechanics and Foundation Division,ASCE, Vol. 94, No. SM5, Proc. Paper 7413, Sept., pp.7513-1653. 6. Duncan, M., Byrne, P., Wong, K. S., and Nabry, P., (1980), "Strength, Stress-Strain and Bulk Modulus Parameters for Finite Bement Analysis of Stresses and Movements in Soil Masses," Report No. UCB/GT/80-01, Department of Civil Engineering, University of California, Berkley, Aug. 7. Bias, V., and Swanson, p., (1980), "Cautions of Reinforced Earth with Residual Soils," Transportation Research Record 919, pp. 21-26. 8. Freitag, D. R., (1986), "Soil Randomly Reinforced with Fibers," Journal of Geotechnical Engineering, ASCE, Vol. 112, No.8, Paper No. 20810, Aug., pp. 823-826. 9. Gray, D. R., and Ohashi, R., (1983), "Mechanics of Fiber Reinforcement in Sand," Journal of Geotechnical Engineering, ASCE, Vol. 109, No.3, Proc. Paper 17780, March, pp. 335-353. 145

PAGE 154

10. Gray, D. H., and Al-Refeai, T., (1986), "Behavior of Fabric-versus Fiber-Reinforced Sand," Journal of Geotechnical Engineering, ASCE, Vol. 112, No.8, Paper No. 20860, Aug., pp. 804-82l. 11. Gray, D. H., Shewbridge, S. E. and Sitar, N., (1991), "Defonnation Characteristics of Reinforced Sand in Direct Shear, Journal of Geotechnical Engineering, ASCE, Vol. 117, No. 11, Discussion and Closure, Nov., pp. 1810-1817. 12. Grogan, W. P., and Johnson, W. G., (1994), "Stabilization of High Plasticity Clay and Silty Sand by Inclusion of Discrete Fibrillated Polypropylene fibers (fibergrids) for use in Pavement Subgrades," Technical Report CPAR-GL-94-2, U. S. Army Engineers Waterways Experiment Station, Vicksberg, MS. 13. Head, K. H., (1982), "Manual of Soil Laboratory Testing," Volume 2, Permeability Shear Atrength and Compressibility Tests. 14. Head, K. H., (1986), "Manual of Soil Laboratory Testing, Volume 3, Effective Stress Test. 15. Heywood, P., (1986), Thread and Sand Tangle togeather in Retaining Wall," Engineering News Record, Sept. 18, pp 86. 16. Hoare, D. J., (1979), "Laboratory Study of Granular Soils Reinforced with Randomly Oriented Discrete Fibers," C. R. Int. Renforcement des Sols., Paris. 17. Jewell, R. A., and Wroth, C. P., (1987), Direct Shear Tests on Reinforced Sand," Geotechnique 37, No.1, pp 53-68. 18. Janbu, Nilmar, (1963), "Soil Compressibility as Detennined by Oedometer and Triaxial Test," European Conference on Soil Mechanics Foundation Engineering, Wiesbaden, Gennany, Vol. 1, pp. 19-25. 19. Kondner, R. L., (1963), "Hyperbolic Stress-Strain Response: Cohesive Soils," Journal of the Soil Mechanics and Foundation Division, ASCE, Vol. 89, No. SM 1, Proe. Paper 3429, February, pp. 115-142. 20. Lawton, E. C., Khire, M. V., and Fox, N. S.,(1993) "Reinforcement of Soils by Multioriented Geosynthetic Inclusions," Journal of Geotechnical Engineering, ASCE, Vol. 119, No.2, Proe. Paper 2132, February, pp. 257-275. 21. Maher, M. H., and Woods, R. D., (1990), "Dynamics Response of Sand Reinforced with Randomlv Distributed Fibers," Journal of 146

PAGE 155

Geotechnical Engineering, ASCE, Vol. 116, No.7, Proc. Paper 24840, July, pp. 1116-1131. 22. Maher, M. H., and Gray, D. H., (1990) "Static Response of Sands Reinforced with Randomly Distributed Fibers," Journal of Geotechnical Engineering, ASCE, Vol. 116, No. 11, Proc. Paper 25205, Nov.,pp. 1661-1677. 23. Namaan, T., Moavenzadah, F., and McGarry, F., (1974), "Probalistic Analysis of Fiber Reinforced Concrete," Journal of Engineering Mechanics Division, ASCE, Vol. 100, No. EM2, Feb., pp. 397-413. 24. Oakland, M. W., and Lovell, C. W., (1983), "Standardized Tests for Compacted Shale Highway Embank ments," Transportation Research Record 873, pp. 15-21. 25. Perloff, W. H., and Baron, W., (1976), "Soil Mechanics Principles and Applications," 26. Perloff, W. H., and Osterberg, ]. 0., (1964), "Effect of Stress History on Strength of Cohesive Soils," Transportation Research Record 47, pp. 49-71. 27. Schlosser, F., and Long, N. T., (1974). Recent Results in French Research on Reinforced Earth," Journal of the Construction Division, ASCE, Vol. 100, No. C03, Mar., pp. 223-237. 28. Seed, H. B., and Chan, C. K., (1959), "Structure and Strength Characteristics of Compacted Clays," Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 85, No. SM 5, Proc. Paper 2216, Oct., pp. 87-127. 29. Shewbridge, S. E., and Sitar, N., (1989), "Deformation Characteristics of Reinforced Sand in Direct Shear," Journal of Geotechnical Engineering, ASCE, Vol. 115, No.8, Paper No. 23785, Aug., pp. 1134-1147. 30. Yang. Z., (1972), "Strength Deformation Characteristics of Reinforced Sand," Dissertation presented to the University of California, at Los Angeles, Calif., in partial fulfillment of the requirements for the degree of Doctor of Philosophy. 147