Effect of crude oil on the geotechnical parameters of sandy clay soil

02028


Introduction
Contamination of the soil by crude oil can be either accidental or intentional, as during conflicts like the invasion of Kuwait [1,2].Contamination can change the characteristics of the soil which can cause the loss of bearing capacity or increase settlement of the foundations of structures.Different methods of remediation have been used to eliminate contamination of the soil, including vacuum extraction, soil washing and biological methods [3].Because of the high cost of remediation, however, especially for large areas, one solution has been to use the contaminated soil in engineering projects.
Such projects include use of the contaminated soil in road construction after mixing it with aggregate or stabilizing the materials.For any type of remediation work and any other possible use of contaminated soil, information about the geotechnical properties of the soil contaminated by crude oil is required [4].
Many researchers have studied the effect of organic contamination on the properties of the soil.Al-sand [5] tested Kuwaiti sand contaminated with crude oil and found a slight decrease in permeability and an increase in compressibility due to contamination.Evgin and Das [6] performed triaxial tests on clean and contaminated sand specimens and reported that the internal friction angle of both loose and dense specimens saturated with oil decreased sharply and that settlement of a footing increased.Meegoda and Ratnaweera [7] suggested that the type, amount and viscosity of the chemicals in the pore fluids affect the compressibility of contaminated soil.Shin and Das [8] investigated the bearing capacity of unsaturated sand contaminated with crude oil.Their specimens had oil contents of 0% to 6% and the results indicated that contamination with crude oil significantly reduced the bearing capacity of the sand.Khamchiyan et al [9] tested clay and sandy soils contaminated with oil to determine the effect of contamination on the geotechnical parameters of the soil.The results showed a decrease in the strength, permeability, maximum dry density, optimum water content and Atterberg limits of the soils.Olgun and Yildiz [10] examined the changes in the geotechnical behavior of clayey soil with high plasticity that had been contaminated with organic pollutants.The results indicated a decrease in the liquid limit and consolidation parameters and an increase in the shear strength with an increase in the contaminant content and a decrease in the dielectric constant of the pore fluid.
Elisha [11] examined the influence of crude oil pollution on the geotechnical parameters of Nigerian clay soil.Their results revealed that an increase in the crude oil increased the Atterberg limits and maximum dry density and decreased the moisture content.Kermani and Ebadi [12] investigated the geotechnical parameters of contaminated soil and reported that the presence of oil increased the friction angle, maximum dry density, compression index, Atterberg limits and optimum moisture content and decreased adhesion in fine-graded soils.
Zhi-bin et al [13] conducted Atterberg limit tests in accordance with ASTM D4318 kaolinite soil contaminated with diesel oil.Their results demonstrated that, as the soil pollution increased, both the liquid limit (LL) and plastic limit (PL) decreased and the plastic index (PI) showed a slight increase.
The influence of contaminants on clay and pure sandy specimens have been studied in previous research.However, because the soil specimens were not from natural sources, study of the influence of contaminants on natural soil is required.Thus, the current study investigated the effect of crude oil on the geotechnical parameters of mixed kaolinite-sand samples in order to simulate the natural soil conditions.

Soil Samples
Kaolinite is the most common mineral in southern Iran.However, because of the establishment of oil refineries across the south, this mineral has been exposed to contamination.For this reason, kaolinite was selected and contaminated with light weight crude oil to study its properties.This soil type is not generally found in pure form in nature; thus, all experiments were performed on mixtures of this kaolinite with 10%, 25% and 40% sand.In accordance with ASTM C778, the sand is classified as Ottawa sand (silica sand).Table 1 lists the geotechnical characteristics of the sand sample.Fig. 1 shows the grain-size distribution of the sand used in the experiments.

Contaminants
Crude oil is classified as light, medium or heavy weight on the basis of the molecular weight of the components.The oil used in this research is light crude oil from Tabriz Refinery and has a sharp odor and dark brown color.Table 3 lists the properties of this crude oil.

Sample preparation
The soil samples first were dried at 105°C for 24 h in the oven to remove any moisture.After combining the kaolinite with 10%, 25% and 40% sand, each mixture was combined with different amounts of crude oil (4%, 8%, 12% and 16% by soil dry weight).One sample of Each soil mixture was prepared as a reference sample without contaminant to determine the effect of crude oil on the kaolinite-sand samples.The combination of the oil and soil was done manually.Because of possible reactions between the oil and soil, the mixture was stored for one week in closed plastic packages at room temperature before use.
Limitations were placed on the amount of crude oil added to the soil samples because it was not possible to reach the peak point in samples containing over 16% oil in the compaction test.The occurred because, even without the addition of water, the samples were considered to be wet.This issue has been addressed by Khamechian et al [1].Furthermore, additional crude oil was released from these samples in the course of the Atterberg limits tests.
In order to determine the effect of light crude oil on the geotechnical parameters of the clay sand samples, the Atterberg limits, standard compaction, direct shear and UCS tests were conducted.At first, the compaction tests were done to find a base density for preparation of samples for other tests.The maximum dry densities and optimum moisture contents of samples containing 16% oil, were selected as the base.With this assumption, all of the sample will be on the dry side of compaction curve and the difference between the samples will be the difference in the amount of crude oil contaminant.

Atterberg limits
The Atterberg limits are used to identify and classify the soil and are a basis for the preliminary assessment of the mechanical properties of soil.Sarsby [14] stated that the distinguishing features of soil relate to its water content and how it influences the performance of the soil.
According to ASTM-D-4959, Relation (1) can be used to calculate the moisture content of the samples as: where is the moisture content, Ww is the water weight and Ws is the dry weight of the ω soil.
One difficulty faced in this research was the inability to evaluate the moisture content using Relation (1) because the evaporation temperature of some oil particles was lower than that of water and these particles evaporated with the water.This relation does not consider the evaporated oil compared with the amount remaining in the soil.Fingas [15] found that the rate of oil evaporation did not relate to the wind speed, turbulence stage, surface, thickness or size.The only effective factors are chemical and physical properties of the crude oil and the time and temperature.
Khamechian et al [9] proposed Relation (2) to determine the moisture content in contaminated soil.They reported that, in contaminated soil, the evaporation of the crude oil and the percentage of materials remaining in the soil relate to the size of the soil particles and the mixture of particles.In addition, the evaporation of crude oil depends on the thermal features of the soil and oil.They also reported that, under similar circumstances, the evaporation of crude oil increased with a decrease in the soil particle size and that, as the oil content increased, the evaporation rate decreased.In Relation (2): where is the moisture content, Wt is the wet weight of the contaminated soil, Ws is ω the dry weight of the contaminated soil, m is the oil remaining in the soil after evaporation and n is the oil content before evaporation.
The use of Relation (2) in the present study did not produce favourable results, because, in some cases, the moisture contents were negative values, which indicates that this relation was impractical in this case.Therefore, Relation (3) has been proposed to produced more rational results: where is the moisture content, Wt is the wet weight of the contaminated soil, Ws is the ω dry weight of the contaminated soil, m is the oil remaining in the soil after evaporation and e is the evaporated oil content.
In order to obtain parameters m and e, soil samples were dried at 105°C for 24 h after contamination.The results in Fig. 2 show that, at the same temperature and time, as the oil content in the soil mixtures increased, the amount of oil evaporation decreased.Moreover, as the sand content in kaolinite-sand mixture increased, the amount of evaporated crude oil increased.After preparing and curing the samples, the Atterberg limits tests were carried out on both contaminated and uncontaminated kaolinite-sand mixtures according to ASTM D4318.The results demonstrated that, as the sand content increased, LL and PL decreased in the three specimens, which indicated a decrease in the absorptivity of the sand compared with kaolinite.Overall, as the crude oil content increased, the LL and PL of all samples increased (Figs. 3, 4 and 5).
The results were similar to those presented by Rahman et al [16], Elisha [11] and Kermani and Ebadi [12] and were contradictory to the results presented by Khamechian et al [11].As the contaminant content increased from 12% to 16%, the LL in the mixtures of 90% kaolinite + 10% sand and 75% kaolinite + 25% sand decreased.For the mixture of 60% kaolinite + 40% sand, it was not possible to perform this experiment because of the release of excess oil from the sample and the consequent adhesion.
The increase in PL and LL could be interpreted according to double-layer theory.As is known, molecules of water are bipolar; hence, a molecule of water has a positive charge on one side and a negative charge on the other side.Bipolar water is absorbed both by means of the negative charge on the surface of the clay particles and the cations in the double-layer.Another mechanism by which water is absorbed into the clay particles is the hydrogen bond in which hydrogen atoms in the water molecules combine with oxygen molecules on the surface of the clay.The water retained in the clay particles by the force of gravity is known as double-layer water.The most internal layer of water is known as absorbed water and is stored by the clay.The absorption of water by the clay particles generates plastic performance in fine-particle soil.The water in the soil pores which is not absorbed by clay particles moves easily through the soil and is called free water.Free water determines the soil liquidity [17].However, because oil molecules are not bipolar like water molecules, the soil particles in the present study were covered with oil as it entered the soil, which did not allow the water molecules to form a disperse double-layer.The increase in PL indicates that an increase in the oil content changed the chemical structure of the pore fluid, which caused a chemical-physical interaction among the soil particles.
The addition of crude oil to the soil particles can cause cation exchange.It could be asserted that the increase in the contaminant content increased the ability of the crude oil to form larger masses from adjacent soil mineral particles.The water available between the pores had been absorbed into the larger surface of the flocculated structure and the amount of free water decreased.Therefore, in order to provide sufficient free water between the pores, more water was required to reach a state of liquidity in the soil.

Standard compaction
Standard compaction tests were performed on all three mixtures of kaolinite-sand according to ASTM D698.As indicated in Figs. 6 and 7, when the sand content in the kaolinite-sand mixtures increased, the maximum dry density of the soil increased and the optimum moisture decreased.This reduction in the optimum moisture content could be due to low water absorption by the sand compared to the kaolinite.In general, as the amount of contamination increased, the curves of the contaminated soil moved to the left side of the compaction graph of the uncontaminated soil.Fig. 8 shows the effect of the crude oil content on the maximum dry density of the samples.In all three mixtures, the dry density increased as the crude oil content increased.This increasing trend has been reported by other researchers [5,18,19] and contradicts the results of Khamechian [9].Fig. 7 also shows the relation between the optimum moisture content and the crude oil content in the soil specimens.This figure reveals an increase in the optimum moisture content with a decrease in the crude oil content in three kaolinite-sand mixtures.A decrease in the optimum moisture content along with an increase in the crude oil content demonstrates the lubricating effect the crude oil.This reduces the amount of water required to reach maximum compaction and facilitates compaction.A change in the compaction parameters occurred as the crude oil occupied the pores between the particles and the lubricating effect of the crude oil made the contaminated soil flow more easily compared to the uncontaminated soil.In other words, as the crude oil penetrated into the soil, it moved toward the pores and filled them in place of water.This decreased the optimum moisture content.It is known that, when crude oil penetrates into the soil, the contact zone of the clay molecules is overlaid and encased in water molecules and charged ions.This influences the engineering performance of the soil.For contamination values of over 16%, the crude oil content was sufficient to attain an effective compaction value such as that for 20% crude oil; thus, compaction was not possible.

Direct shear
The direct shear test was performed in a 6×6 cm circular shear box at a rate of shear equal to 1 mm/min and for three vertical stresses (50, 100 and 150 kPa).The data indicated that the change in cohesion coefficient and the internal friction angle occurred with an increase in the contamination content.The influence of the different crude oil contents on the cohesion coefficient and internal friction angle of the samples are shown in Figs. 8 and 9, respectively.Fig. 8 shows the cohesion of the kaolinite-sand mixtures versus the crude oil content.It is evident that, for the 90% kaolinite + 10% sand and 75% kaolinite + 25% sand mixtures, the addition of 0% to 16% crude oil decreased the cohesion coefficient.Similar results have been reported by Khamechian et al [1,9] on CL soil.As the crude oil penetrated the samples, it lubricated the soil.This caused the formation of large particles or a flocculated structure in the soil and decreased the area of the soil particles, which decreased the cohesion coefficient.An increase in contaminant decreased the thickness of the double-layer around the clay particles.Because the thickness of the double-layer decreased, the surface potential on the clay particles decreased and decreased the adhesion in the samples.Fig.8 shows that a mixture of 60% kaolinite + 40% sand performed differently than the two other mixtures.As the crude oil content increased, the cohesion coefficient of the soil increased.
Lamb [20] and Sridharan and Rao [21] found that the adhesively of the clay particles increased as the dielectric constant of the pore fluids decreased.
It can be said in this case that an increase in the sand content and decrease in the kaolinite content will decrease the double-layer compared to the other mixtures.The sand content had a significant influence and the increase in the cohesion coefficient was caused by increased adhesion produced by the crude oil between the sand particles.The curves for mixtures with 90% kaolinite + 10% sand and 75% kaolinite + 25% sand are shown in Fig. 14.They indicate that an increase in the crude oil content generally increased the internal friction angle of the samples.In the mixture containing 10% sand, an increase in the oil content from 0% to 16% increased the internal friction angle 26%.In the mixture containing 25% sand, an increase in oil from 0% to 12% increased the internal friction angle 10%.However, when the crude oil content increased to over 12%, the internal friction angle decreased.This could be due to the high level of lubrication caused by the oil.In general, because the internal friction angle of the kaolinite-sand mixture is related to the sand content and the presence of crude oil has no significant effect on sand, the internal friction angle of the kaolinite-sand mixture would not normally be influenced by the presence of contaminants.The development of this parameter could be related to the flocculation of the clay structure due to the presence of contaminants.
It was observed that the cohesion coefficient changed as the sand content in the kaolinite-sand mixture increased to 40%.For this mixture, as the crude oil content increased, the internal friction angle decreased.It is possible that, as the sand content increased, its effect on the kaolinite-sand mixture increased.With an increase in the oil content, the sand particles became surrounded by oil and caused the internal friction angle of the soil to decrease.
Mohammadi et al [23] investigated a bentonite-sand mixture and observed a decrease in the internal friction angle.They reported that the sand particles covered by crude oil experienced a decrease in friction between them caused by increased lubrication between the particles.The particles of sand that combined with crude oil became larger than the uncontaminated particles and the contamination with crude oil decreased the internal friction angle of these particles.
As stated, internal friction angle is relevant to the sand in the kaolinite-sand mixture.A decrease in the internal friction angle of the sand particles decreased the internal friction angle in the 60% kaolinite + 40% sand mixture.

Conclusion
In the current study, three kaolinite-sand mixtures were contaminated with 4%, 8%, 12% and 16% crude oil (by soil dry weight) in order to observe the effect of crude oil on the geotechnical parameters of sandy clay soil and the following results were obtained:

•
The LL for all three kaolinite-sand mixtures increased as the contaminant content increased up to 12%.From 12% to 16% contamination, LL decreased in the samples.This could be due the effects of oil lubrication.It was observed that, at high sand contents in the presence of oil, the LL decreased.In all three kaolinite-sand mixtures, the PL increased as the crude oil content increased.
• An increase in the sand content in the kaolinite-sand mixtures decreased the optimim moisture content and increased the maximum dry density.An increase in the crude oil content in all three kaolinite-sand mixtures caused an increase in the maximum dry density and a decrease in the optimum moisture content.This occurred because the crude oil filled some of the pores in the soil which would be expected to fill with water, which reduced the optimum moisture content.In addition, oil lubrication of particles caused an increase in the maximum dry density of the soil.
• An increase in the sand content of the kaolinite-sand mixtures increased the internal friction angle and decreased the cohesion coefficient in the samples.The mixture of 60% kaolinite + 40% sand behaved differently from the other mixtures in that, as the oil content increased, the cohesion coefficient increased.In the 90% kaolinite + 10% sand and 75% kaolinite + 25% sand mixtures, an increase in crude oil content from 0% to 16% increased the internal friction angle.On the other hand, as the sand content in the kaolinite-sand mixture increased to 40%, a change was observed in the soil performance.In this sample, an increase in the crude oil content decreased the internal friction angle.

Fig. 4 .
Fig. 4. PL in soil samples vs. sand and crude oil contents.

Fig. 5 .
Fig. 5. PI in soil samples vs. sand and crude oil contents.

Fig 6 .
Fig 6.Relationship between maximum dry density and crude oil content in specimens.

Fig 7 .
Fig 7. Relationship between optimum moisture and crude oil contents in specimens.

Fig. 8 .
Fig. 8. Effect of crude oil content on cohesion coefficient of soil samples.

Fig. 9 .
Fig. 9. Effect of crude oil content on friction angle of soil samples.

Table 1 .
Geotechnical properties of Ottawa sand used in the experiments

Table 2 .
Basic properties of soil samples

Table 3 .
Properties of crude oil