Geotechnical Behavior of Oil-Contaminated Fine-Grained Soils


Habib-ur-Rehman Ahmed

Assistant Professor, National Institute of Transportation, Risalpur, National University of Sciences and Technology (NUST), Pakistan

Sahel N. Abduljauwad

Professor, Civil Engineering Department, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia

Tayyeb Akram

Professor, Dean, National Institute of Transportation, Risalpur, National University of Sciences and Technology (NUST), Pakistan


Humans are, unintentionally or intentionally contaminating soils from different sources. The contaminated soils are not only a challenge for the environmentalists but also for the geotechnical engineers. When contaminated by crude oil, the soil is subjected to a change in its engineering properties. The soil, which is mostly affected by its environment, is clay, being active electro-chemically. So, a comprehensive laboratory-testing program was carried out to compare the engineering properties of an uncontaminated and a contaminated clay. The research was mainly carried out at King Fahd University, Dhahran, while some behavioral aspects were reevaluated at National Institute of Transportation, Risalpur. Laboratory tests included all basic and advanced geotechnical tests along with Scanning Electron Microscope (SEM). Crude oil was chosen as the contaminant. The clay samples were taken from the Al-Qatif area of the Eastern province of Saudi Arabia. The selected soil is considered to be highly expansive in nature. The comparison between uncontaminated and crude oil-contaminated clay showed that there would be a significant change in the engineering behavior of the clay if it were contaminated by crude oil. The contaminated clay behaves more like a cohesionless material, owing to the formation of agglomerates. The coarse-grained soil-like behavior was obvious in the strength behavior of the oil-contaminated clay. The contamination has affected the plasticity and the cation exchange capacity (CEC) of the investigated clay. The swelling pressure of the clay after contamination suffered three times reduction, while no change was observed in the percent swelling of the contaminated clay.

Keywords: Crude Oil Contamination, Clay Soils, Engineering Behavior.


The environment is being polluted by humans, unintentionally or intentionally, for their short-term benefits. In doing so, not only air and water but the land is also being contaminated. Land contamination is not only harmful for the subsurface water aquifers but such actions are a detriment to the buildings and structures standing on it. Any change in engineering properties and behavior of the soil strata may lead to loss of bearing capacity, an increase in total or differential settlements of the foundation systems of the structures. Consequently, structures may undergo functional or structural failure.

Crude oil contamination of soils may occur through a variety of sources such as oil leakage from damaged pipelines, tanker accidents, discharge from coastal facilities, offshore petroleum production facilities, and natural seepage. One typical example is the oil spills in Kuwait during the Gulf war. Another example is the oil spill at Valdez, Alaska resulting from an oil tanker accident. In Saudi Arabia, soils are being contaminated by the leakage of oil from the pipelines. Oil that is washed ashore contaminates the shoreline soils. Oil leakage from damaged pipelines, oil storage tanks, and processing plants may also cause oil contamination in the surrounding soils [1]. Despite the best efforts of both the petroleum industry and regulatory community, releases, leakages and spills of petroleum products occur frequently. It is estimated that, in the United States, 25% of the underground storage tanks used for the storage of petroleum products are leaking [2].

Once a spill or a leakage occurs, the hydrocarbon liquid, under gravity, moves down to the groundwater, partially saturating the soil in its pathway [3]. Upon reaching the groundwater table, this liquid may spread horizontally by migration within the capillary zone, thereby further saturating the soil. Clay particles are chemically active soil particles. Their behavior is always affected by the environment to variable degree depending on the clay particles’ mineralogy. The particular environment includes the pore fluids and their properties and type of ions present therein [4]. Their behavior can be altered substantially by the presence or permeation of different pore fluids. Due to soil contamination by various liquids from different sources, clay behavior may change.

Adsorption of organic molecules by hydrophobic and hydrophilic clay minerals is strongly governed by their structural aspects [5]. The soil particles coated with activated sludge develop a degree of hydrophobic character [6]. The humic substances or organic compounds similar in chemical structure to the humic substances present in activated sludge can be adsorbed onto clay mineral surfaces as macromolecular complexes. It was also established through a series of long-term Triaxial permeability tests that the permeation of fluids with some organic fluids, such as carbon tetrachloride, caused a significant increase in their hydraulic conductivity [7]. The increase in hydraulic conductivity was attributed to the collapse or shrinkage of the diffuse double layer. Similar studies conducted by [8] and [9] reached the alike conclusions.

The effect of pore fluids on the fabric, hydraulic conductivity and Physico-chemical properties of clays was also studied [10]. The results indicated that changes in hydraulic conductivity correlate well with the point changes in the fabric when the molding pore fluid and the permeation pore fluid are both water. A decrease of 20 percent in dry density at fixed water content may lead to at least an order of magnitude increase in hydraulic conductivity. Tests with organic fluids indicate that the changes in hydraulic conductivity are in response to changes in fabric as well as pore fluid characteristics, interaction with the clay mineral, and confining stress. The pH of the solution is observed to be dominant on the properties of this mineral.

An increase in the percentages of sludge and fly ash cause variation in shear modulus, hydraulic conductivity, and compressibility of the compacted Kaolinite samples [11]. In this study, Fuel oil was also used as one of the contaminants. Fuel oil-contaminated specimens exhibited low strength, high water holding capacity and low unit weight due to the porous and loose structure of these samples as a result of the contamination.

Crude oil affected the physico-chemical nature and microstructure of laboratory prepared marine clays [4]. The crude oil reduced the specific surface area, cation exchange capacity, and double layer thickness of the clay minerals. The addition of crude oil caused the formation of an open structure and increased water holding capacity and decreased strength, stiffness, and permeability for the artificially prepared marine clay sediments. Different types of mechanisms of clay and oil interaction were observed. These include coating of clay particles with oil and oil surfaces with clay particles. In the study, the phenomenon of spherical agglomeration was also observed.


The soil used in this study was an active high plasticity clay obtained from the Al-Qatif area of the Eastern province of Saudi Arabia [12]. For the purpose, undisturbed block samples of clay were carefully cut from the bottom of a test pit. The soil samples were air-dried and pulverized repeatedly until all soil aggregations were reduced to minus 40 sieve size. Due to the low permeability of this soil, it was not possible to inject oil into an undisturbed block. Therefore, the crude oil of grade 35, which was used as the source of contamination, was mixed with the pulverized soil. The oil and soil was mixed in quantities to make the soil fully saturated at its natural dry density of 12 kN/m3. In order to simulate the field environmental conditions, the oil and soil mixture was air-dried for about one week. A comprehensive laboratory testing was then carried out on the conditioned contaminated clay in order to determine its mineralogical composition, and geotechnical and physico-chemical properties. Following laboratory tests were conducted on clean as well as contaminated clay to determine its index properties and compaction, strength, and volume change characteristics:

Determination of the fabric by the scanning electron microscopy (SEM) technique

Cation exchange capacity (CEC), pH, and organic carbon tests were done for the study of the physico-chemical nature of the soil. The CEC was determined by Rhoades's procedure [13].

The consistency behavior was determined by the evaluation of Atterberg limits (ASTM D423, D424 and D427).

Standard Proctor compaction test to investigate the compaction characteristics (ASTM D698)

Unconsolidated Undrained Triaxial compression tests on cylindrical specimens 5.1 cm in diameter and 10.2 cm long, for strength evaluation of soil (ASTM D2850)

One dimensional consolidation test on samples 7.0 cm in diameter and 1.91 cm in thickness, for compressibility analysis (ASTM D2435)

The percentage of swell test was done in a consolidation ring of 7.0 cm diameter and 1.91 cm thick, to determine the volume change behavior of the soil (ASTM D4546).

The swelling pressure test was performed using a constant volume conditions in the oedometer. The volume of the sample was kept constant after flooding the soil with water.



The fabric of a soil refers to the geometric arrangement of particles [14]. Besides the initial density and the stress level, the fabric is considered as the important parameter controlling the engineering behavior of soils. The scanning electron microscopy technique was used to study the fabric of the soil. The following types of fabric were studied for the uncontaminated and the contaminated clays, and were given specific symbols for the future reference.

Uncontaminated clay (F-1)

Fig. 1(a) shows the SEM micrograph of clean clay at 10,000 magnifications. Mineralogically, Smectite flakes and Palygorskite tubes are visible. At this level of magnifications, a deflocculated fabric was observed. There are no distinct flocs visible in these micrographs.

(a) Uncontaminated Clay(b) Contaminated Clay

Figure 1. SEM Micrographs of contaminated and uncontaminated clay


Contaminated Clay (F-2)

Fig. 1(b) is the SEM micrograph of the contaminated clay. The size of the Palygorskite tubes in the contaminated clay is sufficiently larger than the ones in uncontaminated clay. This is due to the oil coating on the individual clay particles and the clay groups. The flocs so formed seem to be silt- or even sand-sized particles. The models for this oil-clay interaction mechanism are shown in Fig. 2.


Figure 2. Schematic Model for Fabric F-2


Contaminated Clay and Distilled Water (F-3)

When water was added to the loose mix of oil-contaminated clay, it was observed that the water dissociated the 'oil-clay' bond. A model for this mechanism is shown in Fig. 3. The distilled water, used in the study, has the capability to dissociate the ionic compounds owing to its highest dielectric constant (80) of all the liquids. The fabric can be regarded as a dispersed fabric.


Figure 3. Schematic Models for Fabrics F-3 and F-4


Contaminated Clay and Brine (F-4)

Brine is the high salt content water and was obtained from the groundwater existing in saline soils, locally known as Sabkha soils [15]. When brine was added to oil mixed clay, it was seen that the clay particles behaved like a cohesionless and non-plastic material. The fabric under these conditions can be regarded as single-grained, dispersed and cohesionless. The probable mechanism and the resulting fabric are shown in Fig. 3.

Physico-chemical Properties

These properties are related to the physical and chemical interaction of the soil particles with each other and with their environment such as the pore fluid and dissolved salts. For fine-grained soils, the physical interaction is of little importance. However, the behavior of fine-grained soils is entirely dependent on how the particles interact chemically with each other or with their environment. Various physico-chemical properties determined for uncontaminated and contaminated clay are tabulated in Table 1.


Cation Exchange Capacity (CEC)

CEC is defined in milliequivalents per 100 g of the dry soil and it was measured by [13]. The comparison of CEC for uncontaminated as well contaminated clays is shown in Table 1. There is a marked reduction in the CEC of the oil-contaminated clay.


The pH was measured by the potentiometer method. The results showed that there was a slight reduction of pH for oil contaminated clays, showing the acidic nature of the crude oil.


Table 1. Summary of Various properties of Clay
before and after contamination


Geotechnical Properties

The geotechnical properties were determined for both the uncontaminated and contaminated clay. The geotechnical properties were used to investigate the influence of microstructure and physico-chemical changes on the physical and mechanical behavior of the clay under investigation.


Atterberg limits

The measured index properties are shown in Table 1. The addition of crude oil caused an increase in the Atterberg limits and the plasticity index. The most probable reason for the increase in Atterberg limits is the extra cohesion provided to the clay particles by the oil. Therefore, additional water is required to cause sufficient layer thickness to change soil from one consistency to the other consistency.


Moisture-Density Relationship

The compaction characteristics of soils need to be evaluated when soils are intended to be used as fills or to be compacted in-situ for roads. Standard Proctor test has been used for the comparison. The contaminated soil showed a marked increase in maximum dry density at relatively low optimum moisture content. Since the soil particles were coated by oil, it acted as an excellent lubricant to achieve such a high density at such lower moisture content, as shown in Fig. 4 and Table 1.


Shear Strength

The shear strength of soil is a crucial property since it controls bearing capacity as well as stability of the foundation system of a civil engineering structure. The strength chosen for comparison purposes is the unconsolidated undrained triaxial compression test (UU CTC). The size of the sample used for the strength tests was 5.1 cm in diameter and 10.2 cm long. The samples were compacted at the field dry density of 12 kN/m3. Results shown in Fig. 5 indicate that at low confining pressures, the strength of the contaminated soil was less than that of the uncontaminated one, while at high confining pressures; the strength was a little more than that of the uncontaminated one. The increase in the strength of the oil mixed clays could be attributed to the agglomeration of particles in the presence of oil. Thus the fabric for which the strength tests were conducted was F-2 (contaminated clay fabric). At low confining pressures, the strength was found reduced due to the formation of large-sized particles, the specific surface area decrease less, consequently there is less cohesion, which is obvious from the Mohr-Coulomb envelope (Fig. 5). At high confining pressures, the strength is relatively high, showing the dependency of the strength of contaminated clay on confining pressure.


Figure 4. Moisture-Density relationship for
the Uncontaminated and Contaminated Clay


Figure 5. Mohr-Coulomb Envelopes for
Uncontaminated and Contaminated Clay


Compression and Swelling characteristics

One dimensional consolidation testing was used to study the volume change behavior of the uncontaminated as well as contaminated clay. A consolidation ring of 7.0 cm diameter and 1.91 cm height was used in the testing. Samples for consolidation, swelling and swelling pressure tests were prepared at the field dry density of 12 kN/m3.

The uncontaminated as well as the contaminated soil samples were prepared at the field dry density given above. Samples were allowed to swell under a surcharge pressure of 6.9 kPa. Results are shown in Fig. 6. For uncontaminated clay, the rate of swelling in the initial stages was very high. This behavior was probably due to the dispersed soil fabric (F-1). Individual particle surfaces were open to adsorb water. For the contaminated clay, the water around the sample came in contact with the top and bottom surfaces of the sample. This interaction caused the fabric to change slowly from F-2 to F-3. F-2 fabric should show less swelling due to the presence of oil coated particles. But, once this oil dissociates from particles in the presence of distilled water, the fabric changes its characteristics from F-2 to F-3. In fabric F-3, the clay particles' surfaces are exposed and adsorb water and consequently swell. Therefore, while the rate of swelling in the case of oil-contaminated clay was much less than the uncontaminated sample, the total percentage swell remained the same. The curve at the bottom of Fig. 6 is the swelling curve for the contaminated clay when 100% saturated brine was used as the molding fluid. The fabric of the sample applicable for this phenomenon is F-4. The sample was tested with brine as the inundation fluid. It showed a remarkable reduction in the percentage of swelling. The most probable reason is that, in brine, Na+ ions get into the oil phase making it more and more positively charged. Therefore, the coating of the positively charged oil on the negatively charged clay particles becomes more adherent. The bond becomes relatively stronger and does not dissociate as it does in the presence of distilled water. Since the clay particles are oil-wet rather than water-wet, the oil coated particles behave like relatively chemically inactive granular materials.


Figure 6. Comparison of Percentage Swelling for
Uncontaminated and Contaminated Clay

After the height of the sample in the percent swell test became constant, the samples were loaded and unloaded in the load increment ratio of one. In Fig. 7 and Table 1, the compression index, Cc, is higher for the contaminated soil. This can be attributed to the presence of organic matter in the crude oil.


Figure 7. Comparison of consolidation behavior of
Uncontaminated and Contaminated Clay

Swelling pressure test

Swelling pressure (constant volume) tests were conducted on the uncontaminated as well as the contaminated clays. Swelling pressure is the pressure applied by the swelling clays when their volume change is prevented. A load cell was used to record the increase in pressure after the addition of water. The sample was then inundated by water. Since the sample was being prevented from changing its volume, in reaction, it applies pressure to the load cell. This pressure is a direct measure of the swelling pressure. The tests were run at the field density of the soil. As shown in Fig. 8, the swelling pressure exerted by the contaminated clay was about 1/3rd of that exerted by the uncontaminated clay.

Figure 8. Comparison of Swelling Pressure Rate of
Uncontaminated and Contaminated Clay


The following conclusions can be drawn from the experimental study carried out on uncontaminated and oil-contaminated clay.



The authors would like to express their appreciation to the King Fahd University of Petroleum and Minerals for providing their support and laboratory facilities for the successful execution of this research and also to Hassan Zakriyah for his help in the performance of laboratory work.



  1. Evgin, E., and B. M. Das (1992) “Mechanical Behavior of Oil Contaminated Sand,” Environmental Geotechnology, Usmen & Acar (eds), Balkema, Rotterdam, 1992.
  2. EPA (1989) “Volumetric Tank Testing: An Overview” US Environmental Protection Agency, EPA/625/9-89/009
  3. Pamukcu, S., and H. Hijazi (1992) “Grouting / Soil Improvement and Geosynthetics,” Geotechnical Special Publication No. 30, ASCE 345 East 47th Street New York
  4. Tuncan, A., and S. Pamukcu (1992) “Predicted Mechanism of Crude Oil and Marine Clay Interactions,” Environmental Geotechnology, Usmen & Acar (eds), Balkema, Rotterdam
  5. Lagaly, G. (1987) "Clay-Organic Interactions: Problems and Recent Results," Proceedings, International Clay Conference, pp. 343-351, The Clay Minerals Society, Bloomington, Indiana, USA
  6. Gauffreau, P.E. (1988) "Hydrophobic Soil: A Low Cost Alternative to Clay Lining Materials," Thesis, Lehigh University, Bethlehem, Pennsylvania, USA
  7. Evans, J.C., H.Y. Fang, and I.J. Kugelman (1985) "Organic Fluid Effects on the Permeability of Soil-Bentonite Slurry Walls," Proceedings, National Conference on Hazardous Waste and Environmental Emergencies, Cincinnati, OH, pp. 267-271, 14-16
  8. Acar, Y.B., and A. Gosh (1986) "The Role of Activity in the Hydraulic Conductivity of Compacted Soils Permeated with Acetone," Proceedings, Environmental Geotechnology Conference, Vol. 1, pp. 403-410, ENVO, Bethlehem, Pennsylvania, USA
  9. Bowerds, J.J. (1985) "The Influence of Various Concentrations of Organic Liquids on the Hydraulic Conductivity of Compacted Clay," PhD dissertation, University of Texas at Austin, USA
  10. Acar, Y. B., and I. Olivieri (1989) "Pore Fluid Effects on the Fabric and Hydraulic Conductivity of Laboratory Compacted Clay," Transportation Research Record, Vol. 1219, pp. 144-159, USA
  11. Pamukcu S., T. Mustafa, and H. Y. Fang (1990) "Influence of Some Environmental Activities on Physical and Mechanical Behavior of Clays," Physico-chemical Aspects of Soils and related Materials, ASTM STP 1095, K. B. Hoddinott and R. O. Lamb, Eds., American Society of Testing and Materials, pp. 91-107
  12. Abduljauwad, S. N. (1993) “Study of Performance of Calcareous Expansive Clays,” Bulletin of the Association of Engineering Geologists, Vol. XXX, No. 4, pp. 481-498
  13. Rhoades, J. D., “Soil pH. In. Methods of Soil Analysis, Part A, Chemical and Microbiological Properties,” Agronomy Monograph 9, 1982.
  14. Holtz, R. D., and W. D. Kovacs (1981) “An Introduction to Geotechnical Engineering,” Prentice-Hall, Inc., Englewood Cliffs, New Jersey
  15. Al-Amoudi, O.S.B., S.N. Abduljauwad, Z.R. El-Naggar, and Rasheeduzzafar (1992) “The Response of Sabkha to Laboratory Tests: a Case Study,” Engineering Geology, Vol. 33, pp. 111-125


© 2007 ejge