International Journal of Scientific Research in Knowledge, 2(5), pp. 241-248, 2014 Available online at http://www.ijsrpub.

com/ijsrk ISSN: 2322-4541; 2014 IJSRPUB http://dx.doi.org/10.12983/ijsrk-2014-p0241-0248

Full Length Research Paper Performance of Soil Embankment Model under Static and Cyclic Loading
Saad Issa Sarsam, Mahmood Diab Alahmad; Hussein Qasim Al-Nadaf
Department of Civil Engineering, College of Engineering, University of Baghdad, Iraq Corresponding Author: saadisasarsam@coeng.uobaghdad.edu.iq
Received 07 March 2014; Accepted 23 April 2014

Abstract The sandy soil with high gypsum content, (usually referred to as gypseous soil), possesses a type of cohesive forces when mixed with optimum amount of water and then compacted, but losses its strength when flooded with water again. The gypseous soil obtained from Al-Fallujah was subjected to laboratory investigation. The physical and chemical properties and then optimum liquid asphalt (emulsion) requirement were determined. A laboratory soil embankment-model of 50x50x30 cm was implemented to study the impact of absorption by capillary rise on the stability of embankment. Load repetitions test was carried out on four gypseous soil models, two of them were pure soil at dry and absorbed conditions, and the other two were stabilized with emulsion at dry and absorbed condition. Another set of four gypseous soil models of the same condition were tested under static load. The impact of changing the hydraulic conductivity of soil due to asphalt stabilization was investigated and the vertical deformation was determined using LVDT. For the pure soil in dry condition the vertical deformation was 7.45 mm at 157 load repetitions, while for pure soil model under absorbed condition, the vertical deformation was 12.5 mm at 29 load cycles. The stabilized soil at dry condition exhibits vertical deformation of 9.75 mm at 911 load cycles, and shows 10.47 mm deformation at 897 load cycles under absorption. When tested under static load, the ultimate sustained pressure was 0.8 MPa with vertical settlement 0.03 mm for pure soil at dry condition, and reduced to 0.3 MPa with vertical settlement 12 mm at absorbed condition. Keywords: Embankment model; Load repetition; Static loading; Emulsion; Stabilization; Capillary rise; Deformation

1. INTRODUCTION The gypseous soils are usually stiff when they are dry, especially because of the cementation of soil particles by gypsum, but great losses in strength and sudden increase in compressibility occur when these soils faces moisture. There will always be a possibility for water to penetrate through the pavement cracks to the soil beneath. Another source of such problem is the absorption of moisture from ground water through the capillary rise action. It may exhibit hazardous situation if it is utilized for construction of road embankment. The problem becomes more complicated when the ground water flows through the gypseous soil causing leaching and movement of gypsum. In addition to softening, a loss in soil solids takes place. This causes a continuous collapse in the gypseous soil; the dissolution of the cementing gypsum causes high softening of the soil (Al-Mufty, 1997). For the construction of any type of structure resting on problematic soils such as gypseous soils, there are many available methods to improve the behavior of soil. Covering the soil particles with thin layer of

asphalt will increase the cohesion, will limit the negative effect of water by blocking the voids, and will reduce the ability of water to traverse the soil layer through capillary action process (Ingles and Metcalf, 1972). Then to prevent the soil from collapsing, the asphalt stabilization could provide a good remedy. Stabilization of such soil with liquid asphalt will furnish waterproof layers with extra particles bond to serve for embankment construction. In gypseous soils, collapse or compression occurs very quickly when the site is flooded with water during heavy rainfall, irrigation or breaking of sewerage and water pipes which may damage the engineering structures because the element of structure cannot follow the sudden deformation that occurs by rearrangement of the inside forces or stresses (Al-Mohammadi et al., 1987). Sarsam and Ibrahim (2008) studied the behavior of shear strength and rebound consolidation of gypseous soil stabilized with asphalt emulsion. The direct shear test results showed that addition of emulsified asphalt significantly increased the cohesion and the angle of internal friction. One dimensional confined compression test results under both dry and saturated

241

Sarsam et al. Performance of Soil Embankment Model under Static and Cyclic Loading

test conditions indicated that the addition of emulsified asphalt had considerably reduced the void ratio and permeability more than the cutback asphalt did. When the stabilized gypseous soil was tested for rebound consolidation, it indicated that the addition of emulsified asphalt created a type of elastic properties and behavior in the soil at high stress application, and the permanent strain was reduced. The loading type that an embankment experiences in the service life is the repeated load by Vehicles; while the embankment will also experience a static loading due to its self-weight and pavement above. The object of this investigation is to study the effect of change of hydraulic conductivity of gypseous soil by asphalt emulsion stabilization on gypseous soil behavior under static and dynamic (repeated) loading.
Deformation Condition

The soil embankment model was used to assess the hydraulic conductivity under cyclic and monotonic load with dry and absorbed condition. 2. FAILURE CRITERIA EMBANKMENT MODEL FOR SOIL

In several countries, the critical range of permanent vertical deformation of the embankment is 20-30 mm (Sarsam, 1999). In UK, it is recommended to use deflection criterion that should ensure that the deformation is not exceeding (12.5 mm) in depth, (Lister and Addis, 1972). The classification of road surface condition as used in T.R.R.L method (Molenar, 1982) is presented in Table 1.

Table 1: The vertical deformation (Molenar, 1982) Less than 10 mm 10-20 mm Greater than 20 mm Sound Critical Failed

Rut depth of (12 mm) was used as a failure criterion for thickness design in Kentucky (Jain, 1980). This value was based on the depth of rutting made in the top soil due to load repetitions, and is considered in this investigation. The embankment model of pure soil tested in dry condition was considered as a reference, the absorbed pure soil model behavior was compared for variation. While the model of dry stabilized soil was considered as a reference, the absorbed stabilized model with emulsion behavior was compared for variation. 3. MATERIALS PROPERTIES AND TESTING PROGRAM 3.1. Gypseous soil The soil of this investigation was brought from AlFallujah; it was obtained from a depth of 0.5m below

top soil level. Such gypseous soil is usually found at shallow depth starting from top soil and proceeds downward to more than 10 meters deep. It is usually used for road embankment construction and experiences the collapsibility problems when come in touch with water. The soil was sieved through sieve No.4, and the portion passed the sieve was considered for the detail investigation. Table 2 shows the physical properties of the soil, while table 3 illustrates the chemical composition of the soil. Fig. 1 shows the grain size distribution of the soil. The soil is poorly graded sand, non-plastic, with 49% of gypsum content. 3.2. Asphalt emulsion It was obtained from Alzahif Alkabeer plant, table 4 shows that it is cationic emulsion with good abilities of coating and water resistance.

Table 2: physical properties and chemical composition of gypseous soil Physical property Test result Specific gravity according Gs=2.48 Liquid limit (%) 24 Plastic limit (%) Non plastic Plasticity Index (%) Non plastic Max. standard dry density,( standard) 17.17 kN/m3 Optimum moisture content (%) 14% Max. dry density, (modified) 18.67 kN/m3 Optimum moisture content (%) 11% Coefficient of curvature 1.5 Coefficient of uniformity 5.2 Unified Classification System SP-SM

242

International Journal of Scientific Research in Knowledge, 2(5), pp. 241-248, 2014

3.3. Testing program The details of design requirement of optimum moisture content or fluid content of 11% (which is 6% emulsion and 5% water) based on unconfined compressive strength is published elsewhere (Sarsam et al, 2013). A laboratory soil embankment-model of 50x50x30 cm was implemented to study the impact of absorption by capillary rise on the stability of embankment. Load repetitions (cyclic loading) test was carried out on four gypseous soil models, two of

them were pure soil at dry and absorbed conditions, and the other two were stabilized with emulsion at dry and absorbed condition. Such cyclic loading may represent the load repetitions applied by the moving trucks on the embankment. Another set of four gypseous soil models of the same condition were tested under static load (monotonic loading test). The vertical deformation was determined using LVDT, and the impact of changing the hydraulic conductivity of soil due to asphalt stabilization was investigated.

Table 3: Chemical composition of gypseous soil

Chemical Composition Gypsum content (CaSO4) (%) Carbonate content (CaCo3) (%) Total soluble salts (T.S.S.) (%) Total (SO3) (%) pH value Percentage % 49 46 38 22 7.77

Fig. 1: Grain size distribution of the gypseous soil Table 4: Properties of asphalt emulsion
Property Particles charge Viscosity CST Cement Mixing Settling Time (hour) Coating ability and water resistance Coating dry & wet aggregate Test result + ve 45 1.2 19 Good Fair

3.4. Preparation of soil embankment-model box Soil was mixed thoroughly with the predetermined percentage of water, stored into nylon bags for two hours to ensure homogeneity of the mix. For asphalt stabilized soil, Soil was mixed thoroughly with the predetermined percentages of water and liquid asphalt (emulsion), and left for two hours at room temperature of (313) C for aeration. The soil or the stabilized soil was compacted into the model box in five layers of equal thickness of 5cm. The static compaction was implemented for each layer to achieve the required density (95% of modified compaction test). The pure

soil model was left for 24 hours to allow the gypsum to start its cementation action, while the asphalt stabilized soil was left for one week for curing. The decision of aeration and curing periods mentioned above are based on previous work (Prakash and Sarsam, 1980; Prakash and Sarsam, 1981). The model was then subjected to monotonic or cyclic loading test as required. For absorption condition, the compacted soil, or stabilized soil-model box, had inlet holes at the bottom, and sides to allow water to percolate and reaches the soil. The model was subjected to the partial immersion in water for three days to allow the water to rise by capillary action, the elevation of water

243

Sarsam et al. Performance of Soil Embankment Model under Static and Cyclic Loading

covers the bottom 5 cm layer of the soil as demonstrated in fig.2. This procedure may represent the behavior of embankment in the field when water from different sources accumulates at the base surrounding for few days after heavy shower. 3.5. Testing the embankment model Four embankment models were subjected to the load repetitions (cyclic loading) test as shown in fig.3. two models of pure and stabilized soil were subjected to such loading under absorbed condition, while the other two models were tested under dry condition. The model of repeated load system consisted of a load of 10kg attached to shaft; the load has been applied repeatedly by electric motor. The repeated loading

was generally taken as 10 percent of the ultimate static bearing capacity of the pure soil in soaking condition (critical condition). Thus in complete loading cycle, the embankment settlement was recorded at the end of loading cycle by LVDT which was attached to read the settlement as voltage (calibrated as 0.004 V = 1mm) as shown in fig. 4. A digital camera was used in this test for monitoring the repeated loading process and furnishes the possibility of computing the number of blows with each settlement stage. The repeated load mechanism and wave pattern is shown in figure 5, such cyclic loading of 0.1 second followed by a rest period of 0.4 seconds could represent the vehicular loading experienced by the embankment (Sarsam et al., 2013). Fig.6 shows the cyclic loading setup on the model.

Fig. 2: Gypseous soil model under absorbed condition

Fig. 3: Cyclic loading on the soil embankment model

Fig. 4: Vertical deformation measurement accessories LVDT

Fig. 5: Wave pattern of the repeated load

Another four soil embankment models were subjected to static (monotonic) load test, two models of pure and stabilized soil were subjected to such loading under absorbed condition, while the other two models were tested under dry condition. Fig. 7 shows

the model under static load apparatus. Both of the static and cycling loadings were applied through circular metal plate of 50.8 mm diameter and 12.5 mm thickness, which represent the model tire print on the roadway (Sarsam et al., 2013).

244

International Journal of Scientific Research in Knowledge, 2(5), pp. 241-248, 2014

Fig. 6: Cyclic loading setup on the soil model

Fig. 7: Static load setup apparatus

4. DISCUSSION OF TEST RESULTS 4.1. Monotonic (static) Loading Test Vertical deformation was recorded using digital dial gauges as shown in fig.6; the stress- deformation behavior of pure soil is presented in fig.8, it indicates the collapsible behavior of the soil when tested under absorbed condition. At 0.3 MPa, the soil exhibit 12mm of vertical deformation, while it was stable when tested under dry condition, and shows 0.03mm of deformation at 0.8 MPa of applied stress. The reduction in load sustaining capability of the soil due to water absorption was 62.5% at failure. Fig.9 shows the impact of changing the hydraulic conductivity of gypseous soil by asphalt stabilization on load sustaining capacity. The variation in stabilized soil behavior between the dry and absorbed test condition was not significant up to 3.5 MPa of applied stress. Both testing conditions shows low deformation of 0.07mm. The embankment model was able to sustain an applied stress of (4 - 4.5 MPa) before failure and shows 0.12 mm of vertical deformation. This finding indicates that asphalt stabilization has positive impact on gypseous soil by adding more strength, blocking the voids and limiting the capillary rise of water. The collapsible nature of gypseous has changed to more stable condition. The load carrying capacity increases

more than four folds due to asphalt stabilization, on the other hand, the vertical deformation decreases by one fold when testing the soil or stabilized soil at absorbed condition. 4.2. Load repetition (Cyclic Loading) Test Fig.10 shows the collapsible behavior of gypseous soil under cyclic loading. The soil is able to sustain 157 load repetitions when tested at dry condition, and experience total vertical deformation of 7.45mm. When the soil is subjected to capillary rise of water, the load sustained capacity is reduced by 81%, and the soil can sustain 29 load repetitions and exhibit 12.5mm of vertical deformation at failure. When the soil is stabilized with emulsion, the stabilized soil is able to sustain 911 and 897 load repetitions at failure for dry and absorbed test conditions respectively. The impact of water absorption was not significant; this may be attributed to the change of hydraulic conductivity of the soil when emulsion was introduced. Fig. 11 shows such behavior of the stabilized soil under cyclic loading condition. The variation in vertical deformation at failure between both testing conditions is also not significant indicating the positive impact of emulsion on the properties of gypseous soil. Table 4 summarizes the soil and stabilized soil behavior under cyclic loading.

245

Sarsam et al. Performance of Soil Embankment Model under Static and Cyclic Loading

Fig. 8: Applied stress - vertical deformation for pure soil

Fig. 9: Applied stress - vertical deformation for stabilized soil

The vertical deformation behavior of pure soil was almost similar and within a range of 12 14 mm between static and cyclic load at absorbed condition, while it is variable at dry test condition. This may be attributed to the fact that cyclic loading process consist of loading and rest periods, the stress relaxation process will gradually introduce and increase the deformation by breaking the cementation

action and rearrangement of the soil particles orientation, while the static load will not give a chance for such particles rearrangement process. When the soil was stabilized with asphalt emulsion, the asphalt will act as a lubricant in the particles rearrangement process and introduce a type of elastic behavior. This will furnish similar behavior of stabilized soil embankment under cyclic or static load.

Fig. 10: Vertical deformation of pure soil under cyclic loading Fig. 11: Vertical deformation of stabilized soil under cyclic loading Table 4: Results of deformation and load repetitions
Embankment Model Type Pure soil with dry condition Pure soil under absorbed condition Emulsion stabilized soil with dry condition Emulsion stabilized soil under Absorbed condition Load repetitions 157 29 911 897 Total vertical deformation (mm) 7.45 12.55 9.75 10.47

5. CONCLUSIONS Based on the testing program, the following conclusions could be drawn.

1. Gypseous soil embankment model show collapsible behavior when tested by monotonic under absorbed condition. The load sustaining capacity was 0.8MPa with 0.03 mm of deformation at dry test condition, and reduced to 0.3MPa with 12mm of

246

International Journal of Scientific Research in Knowledge, 2(5), pp. 241-248, 2014

vertical deformation after absorption. The reduction in load sustaining capability of the soil due to water absorption was 62.5% at failure. 2. Implementation of cyclic loading on gypseous soil embankment model exhibit similar collapsible behavior, the soil is able to sustain 157 load repetitions when tested at dry condition, and experience total vertical deformation of 7.45mm. When the soil is subjected to capillary rise of water, the load-sustained capacity is reduced by 81%, and the soil can sustain 29 load repetitions and exhibit 12.5mm of vertical deformation at failure. 3. Addition of asphalt emulsion has positive impact on soil behavior; it changes the hydraulic conductivity of the soil, the embankment model was able to sustain an applied stress of (4 - 4.5 MPa) before failure and shows 0.12 mm of vertical deformation at dry and absorbed test conditions under monotonic loading. 4. Under repeated load condition, the stabilized soil is able to sustain 911 and 897 load repetitions at failure for dry and absorbed test conditions respectively. The impact of water absorption was not significant. REFERENCES Al-Mohammadi N, Nashat I, Bake G (1987). Compressibility and Collapse of Gypsiferous Soils. 6th Asian Conference on Soil Mechanics, Tokyo. Al- Mufty A (1997). Effect of Gypsum Dissolution on the Mechanical Behavior of Gypseous Soils. Ph. D. Thesis, Dep. of Civil Engineering, University of Baghdad. Ingles O, Metcalf J (1972). Soil Stabilization, Principles and Practice. Butter Worth Limited, Australia.

Jain S (1980). Prediction of Rut Depth in Flexible Pavement. I.R.C Journal Vol, 41-2-December India. Lister N, Addis R (1972). Field Observation of Rutting, their Practical Implication. T.R.R. No 640, (P 28-38). Molenar A (1982). Visual condition surveys, A Tool in the planning of maintenance. Proceeding, seminar on maintenance and drainage aspects of road pavements Barigalore, India. Olutaiwo A, Adedimila A, Sidiq U (2008). An Examination of the Use of Liquid Asphalt Binders in Road Works in Nigeria. Journal of Engineering and Applied Sciences, 3(1): 134142. Prakash S, Sarsam S (1980) Effect of Binder and Moisture Contents on Soil Cutback Mix. IRC, Indian Highways, 8(3): 55-69. Prakash S, Sarsam S (1981). Effect of Curing Time on Soil Cutback Mix for Faloga Soil. IRC, Indian Highways, 9(10): 39-44. Sarsam S (1999). Prediction of rut depth in asphalt concrete pavement. Scientific Journal of Tikrit University, Engineering Science, 6(4):116-135. Sarsam S, Alsaidi A, Mukhlef O (2013). Behavior of Reinforced Gypseous Soil Embankment Model under Cyclic Loading. Journal of Engineering, 19 (7). Sarsam S, Ibrahim S (2008). Contribution of Liquid Asphalt in Shear Strength and Rebound Consolidation of Gypseous Soil. Engineering and Technology, 26(4). Sarsam S, Alsaidi A, AL-Khayat B (2011). Implementation of gypseous soil-asphalt stabilization technique for base course construction. Journal of Engineering, 17(5): 1066- 1076.

247

Sarsam et al. Performance of Soil Embankment Model under Static and Cyclic Loading

Prof. Saad Issa Sarsam was born in Baghdad (1955), got his BSc. In Civil Engineering (1977), Post graduate diploma in Transportation Engineering (1978); MSc in Transportation Engineering (1980). He worked as senior material Engineer for NCCL (1982-1992); He joined the academic staff at University of Mosul (1992-2005) and got the Assistant Professor degree at (2002); He joined the academic staff at University of Baghdad (2005 until now) and got the Professor degree at (2007). Areas of specialization and interest: (Roller compacted concrete; modified asphalt concrete; Asphalt stabilized embankment models; Road user characteristics).

Assist. Professor Dr. Mahmood Diab Alahmad was born in Baghdad (1975), got his BSc. in Building & Construction Department, university of technology (1997), Baghdad. Got his - M.Sc. in Geotechnical Engineering (1999), university of technology. Got his PhD. In geotechnics (2002). He joined the academic staff at University of Baghdad (2006 until now) and got the Assist. Professor degree at (2014). Areas of specialization and interest: Grouting, embankment models, soil stabilization, soil contamination.

Hussein Qasim Abdul-Ameer Al- Nadaf was born in Thiqar, (1983). Got his B.Sc. in Building & Construction Department, university of technology (2005), Baghdad. Got his - M.Sc. in Geotechnical Engineering (2013), university of Baghdad. He got a job as Site coordinator at Huawei Co. Baghdad. Areas of specialization and interest: -Soil stabilization; Soil contamination; Grouting of Soil with Cement.

248