Use of the Dynamic Cone Penetrometer in Compacted Clay–Sand Layers

Abstract

The ASTM D6951 suggests a test method using the dynamic cone penetrometer (DCP) for the assessment of shallow pavement strength. This study is dedicated to evaluating a similar test method to be used for fine-grained clay–sand mixtures, which are commonly used to construct liners and fluid barriers. The liner layers include clay, which is water-sensitive. The shear strength of the liners depends mainly on the composition and proportions of the mixtures adopted. A series of tests were conducted in the laboratory for compacted clay–sand mixtures, including 5% and 10% bentonite material. The responses to advancing a dynamic probe into compacted material at three various moisture conditions—namely, the optimum, dry of optimum, and wet of optimum water contents—are investigated. The penetration profiles are compared for the two clay content levels selected and for three moisture state conditions. The clay–sand liners support vehicles and vibratory compactors during construction and when waste material is placed. The shear strengths of the mixtures are evaluated in the laboratory using direct shear tests. From the results obtained in this investigation, general guidelines and recommendations are presented to help with the efficient assessment of liners consisting of clay–sand mixtures. A cheap and quick approach to assess the density and shear strength of landfill liners or clay–sand layers can be achieved using a hand-held dynamic cone penetrometer. The state of moisture during compaction was found to influence the DCP results. The compaction of layers in a wet of optimum state is not expected to be an ideal alternative.

1. Introduction

The standard test method designated as ASTM D6951/D6951M-09 [1] (entitled the Standard Test Method for Use of the Dynamic Cone Penetrometer in Shallow Pavement Applications) is initiated to assess pavements for strength evaluation. It is claimed that the penetration rate may be related to the in situ strength, such as the California bearing ratio (CBR) and the soil density when the type of soil and water content are known. The test is carried out with an 8 kg hammer and the option of a 4.6 kg hammer to suit fine-grained soils or when excessive penetration is expected with heavy hammers. A schematic diagram of the DCP device is presented in Figure 1.

This tool can be utilized in the assessment and control of clay–sand mixture layers used in liners and the waste disposal process. It is handy, quick, and more reliable when used in homogenous soil. The compacted clay liners are generally assumed to have a uniform moisture content and be free from gravel and boulders. This dynamic penetration test is ideal for assessing the shear strength and dry density of such layers. The heavy version (8 kg hammer) may not be needed and only the lightweight hammers can be utilized. Clay–sand liners often include bentonite as a clay additive, which is known for its high expansion. The field density tests normally conducted at the time of construction can be subjected to serious changes due to the swelling nature of the material. The expansion will cause the density and shear strength to drop significantly when wetting occurs as a result of possible changes in the environmental conditions [2].

The clay used in liners and water barriers is commonly highly plastic or processed commercial bentonite. The cost of processed clay and bentonite has led many researchers to investigate the use of local materials. Rawas et al. (2005) investigated the use of Oman shale in liners [3]. Reference [4] Obrike et al. (2009) investigated the use of Auchi shale and Imo shale in waste disposal landfills in Nigeria. Langdon et al. (2008) [5] studied the permeability of clay liners of some geological formations and different depositional basins in Turkey. These are only some examples, and the research is continuous.

Compacted clay–sand mixtures have frequently been introduced to act as barriers in waste disposal and containment.

The failure of a landfill liner can occur when the hydraulic conductivity is increased beyond a specified limit or when a failure occurs as a result of a chemical attack or cracks developing within the clay–sand mixture. General specifications of soil liners can vary from one project to another. Some tolerance for an amount of water to be allowed to pass can be considered. Daniel (1993) [6] suggested that the coefficient of hydraulic conductivity for liner materials is ≤10⁻⁷. (cm/s). Dafalla [7] demonstrated the successful use of natural clays with moderate to high plasticity in Saudi Arabia.

The cone penetration concept is used in determining the liquid limit for clays and has been standardized for a long time in practice (BS 1377, ISO 17892–6) [8,9] often called the fall cone test procedure. The definition of the liquid limit is the moisture content at which a clay material starts to behave like a liquid. This occurs at very low shear strength. The fall cone can be designed to measure the shear strength (S_u) in kPa using the Hansbo formula [10]:

$$S_u=k(\frac{m\times g}{d^2})$$

(1)

where

Su = shear strength;

k = a cone coefficient;

m = mass of the cone;

g = acceleration of gravity;

d = the cone penetration measured in mm.

The mass of a 30° cone is 80 g, and k is the cone coefficient, which was taken as 0.8 for the cone size and geometry for the fall cone used in the laboratory to determine the liquid limit. g is the acceleration of gravity (m/s²), and d is the cone penetration depth measured in mm.

Modern penetration devices have become very sophisticated in recent years and are supplied with state-of-the-art sensors that are capable of measuring water pressure, moisture content, seismic parameters, and others. Many studies have confirmed direct and indirect correlations between these measurements and some geotechnical parameters [11,12,13]. These parameters include moisture content, plasticity, index properties, and shear strength.

The stiffness of the ground in coastal zones or swamp areas was assessed in early times using the feel or resistance to the penetration of a steel rod or wood. This can either be dynamic or by continuous pushing. The initial penetrometers were dynamic until recent technologies introduced control units to measure the response to pressure or the rate of penetration. For gravelly and coarse formation, a cone attached to a heavy hammer is recommended. This can be similar to the German Standards, DIN 4093-4 [14], which specifies a 50 kg hammer falling over a 50 cm distance on a 43.7 mm diameter cone.

When advancing a cone into the ground, pressure is needed to create a hole by pushing the soil aside and exerting point load pressure. The pressure applied to advance the cone can be equated to the resistance to penetration, and the soil strength measured this way depends on several parameters that include but are not limited to the soil water content and the bulk density. These two parameters influence the resistance to penetration and need to be reported in all penetration profiles presented.

The cavity expansion theory is an analytical approach conducted to determine the pressure–expansion relationships. Assuming an incompressible solid with a plastic region surrounding the cavity and a linearly infinite deformable solid zone beyond that region, Vesic [15] studied the cavity expansion of cavities using the Mohr rupture condition to determine the spherical and cylindrical cavity expansion stresses. The application of cavity expansion theory may not correctly model the strain path followed by the clay under pressure. Baligh [16] suggested a strain path method that accounts for the complex history of the soil during penetration. Solutions for unsaturated soils are more complex than those for saturated soils.

Cater et al. [17] presented a solution for the pressure–expansion relationship in pressuremeters considering small strain deformations in cylindrical and spherical cavities in clayey sand soils. Schnaid et al. [18,19] analyzed cavities based on Carter et al. work for pressuremeters. Advanced models that have been presented by other researchers include those in [20,21,22].

Reference [23] presented a dynamic analysis of a smooth penetrometer free-falling into uniform clay. Collins and Miller [24] also studied SWCCs using cone penetrometers for two sites of low- to high-plasticity clayey material.

Reference [25] used a penetrometer to evaluate the compaction in levees. They claimed that simple penetrometers can be used successfully in the evaluation of the degree of compaction in compacted partially saturated fine-grained soils. Utilizing this test method can be a quick and reliable approach to assessing the quality of clay–sand liners. In order to use it, several measures need to be taken into account; these include the following:

The weight of the hammer;

The geometry of the cone;

The uniformity of the moisture content within the site;

The composition of the soil material.

Without considering the above conditions, the field shear-strength data may be erroneous and are not likely to reflect the true measurements. However, it can still be used for comparison purposes within the site and can enable the easy recognition of poorly compacted zones. As the material used in clay–sand liners is uniform and does not include gravel or boulders the assessment using the cone penetration method can be ideal.

This study was directed towards the possible use of a hand-held penetrometer to investigate and survey the density and shear strength of clay–sand liners and establish a contour profile that indicates loose zones and areas suggested for repair works. The process is fast and will save considerable assessment time. A modified version of the dynamic penetration probe suggested by ASTM Standards was adopted for this purpose. Two clay–sand mixtures were investigated and their relevant geotechnical parameters were compared.

2. Materials and Sample Preparations

2.1. Materials

2.1.1. Sand

Poorly graded sand was used in preparing the clay–sand mixture for the liner material. The sand had a maximum grain size of 1 mm, with less than 5% of the material passing sieve number 200. The specific gravity was 2.65. This type of sand is commonly used in sand–bentonite or clay–sand mixtures for research purposes in laboratories.

2.1.2. Bentonite

The bentonite material was processed from highly plastic clay. Bentonite is used in many applications, which include drilling operations and water barrier layers. When inundated, it becomes semi-impermeable due to the expansion of the clay particles. The quality and index properties can vary between one type of clay and another. In clay liners, natural clay with high plasticity can also be a good additive to sand for the purpose of reducing its permeability. However, bentonite is ideal due to the homogeneous size and components of the product.

OCMA-grade commercial bentonite supplied by the Arabian Minerals & Chemical Co., Ltd. in Saudi Arabia was selected for use in this study. The liquid limit of the bentonite was measured as 316 and the plastic limit as 62. A typical chemical analysis of OCMA-grade bentonite manufactured in the Arabian Gulf region is presented in

Table 1. Chemical composition of OCMA grade bentonite.

Oxides	Concentration (wt %)
SiO2	55.20
Al2O3	17.00
FeO3	2.90
Na2O	1.90
MgO	4.60
CaO	0.90
TiO2	<0.01
K2O	0.10
Loss of Ignition	16.70
Total	99.30

Source ore—Arabian Gulf Region, OCMA-grade (1998).

. Figure 2 presents the grain size distribution.

2.2. Sample Preparation Methods

This study was aimed at investigating clay–sand liners for some engineering parameters in the laboratory. Different methods of sample preparations were conducted. All sand and clay materials were oven-dried. To prepare the layers of clay–sand mixtures in the laboratory, 80 cm by 80 cm boxes 10 mm in height were constructed. Based on the compaction property, a predetermined amount of sand was mixed with a computed amount of bentonite (by dry weight) to form a mixture similar to that applied in the field. The moisture content, bulk, and dry density were computed to simulate the intended liner condition. Compaction was carried out using a vibrating dynamic machine for the square box samples or tamping using compaction hammers as in standard proctor molds. The layer thickness was selected to be from 8 to 10 cm according to the type of test required. The energy (compactive effort) supplied in the compaction test was 595 kJ/m³ Compaction energy was applied to place three equal portions of the mixture such that the required dry density was achieved. The square box samples were prepared to show a grid and distribution of the dynamic hammer points. Dynamic hammers of variable weights were considered, but this study was limited to the proposed hand-held dynamic probe described.

Figure 3 presents the compaction method on a laboratory layer prepared with the required density and a grid of testing points.

3. Test Methods and Procedures

3.1. Dynamic Cone Penetrometer

The dynamic cone penetration used is shown schematically in Figure 1. It consisted of a 16 mm diameter steel drive rod with a replaceable 60 degree cone tip with a 20 mm base diameter, a 4 kg hammer that can drop from a fixed height of 575 mm, a coupler assembly, and a handle. All components were as specified in ASTM D 6951/6951M-09 [1]. A lighter hammer was used to suit the strength of the clay–sand liner layers.

The test method was quite simple. The DCP device was inspected for damaged parts or excessive use of the replaceable cone tip. The operator held the device by the handle perfectly vertically, lifted the hammer to the specified height, and then let it free-fall. The recorder attached can record the total penetration for the number of blows used or alternatively direct measurements can be scaled off for the number of blows applied. The penetration rate can be worked out if needed.

In order to exclude the effect of loose contact between the cone and the soil, the DCP needed to be held vertically, and the tip seated such that the top of the widest part of the tip was flush with the surface of the layer to be tested. The initial reading was obtained from the graduated drive rod. The final reading was taken after applying a specified number of blows.

This study involved testing the layers of two clay contents of 5% and 10%. Each mixture was prepared with three moisture conditions: wet of optimum, optimum moisture content, and dry of optimum moisture content. A grid of 9 points was labeled in each box, as shown in Figure 3. For comparison purposes between the two mixtures, the results are presented in three rows. Row No. 1 includes three points for the bottom line, Row No. 2 includes three points in the center, and Row No. 3 includes three points at the top. The legend shows the locations of the points for each row.

3.2. Standard Proctor for Liner Material

The clay–sand mixture materials were tested for compaction using the standard proctor test. The procedure explained in ASTM D 698: Standard Test Methods for Laboratory Compaction Characteristics of Soil Using Standard Effort (12 400 ft-lbf/ft3 (600 kN-m/m³)) was followed [26]. The maximum dry densities for the mixtures of 5% and 10% bentonite were reported as 1.723 and 1.742 gm/cm³ kPa at optimum moisture contents of 13.85% and 13.61%. The results are presented in

Table 2. Compaction and shear strength properties of tested samples of clay–sand mixtures.

Mixture Bentonite Content	Moisture State	Actual wc (%)	Actual ϒwet (gm/cm3)	Actual ϒdry (gm/cm3)	Measured Cohesion C kPA	Measured Friction Angle φ
5% B	DOMC	9.88	1.820	1.656	2.2	38.6
	OMC	13.85	1.961	1.723	15.2	33.5
	WOMC	17.33	1.895	1.615	6.5	36.4
10% B	DOMC	10.63	1.869	1.689	22.3	33.4
	OMC	13.61	1.979	1.742	30	33.5
	WOMC	18.68	1.887	1.590	18	34

Note: wc = water content, ϒ_wet = wet unit weight, ϒ_dr = dry unit weight, φ = friction angle. C = cohesion.

, along with other density levels selected for the purposes of this research.

3.3. Direct Shear Test

The two bentonite sand mixtures were tested for direct shear in accordance with ASTM D 3080, Standard Test Method for Direct Shear Test of Soils Under Consolidated drained Conditions [27]. The optimum moisture content of the tested samples was selected at three different levels: wet of optimum moisture content, optimum moisture content, and dry of optimum moisture content. The compaction energy applied was the same as the standard proctor test. The resulting dry densities for the 5% bentonite mixtures were: 1.656, 1.723, and 1.615 gm/cm³ at moisture contents of 9.88, 13.85, and 17.33, respectively. The resulting dry densities for the 10% bentonite mixtures were: 1.689, 1.742, and 1.590 gm/cm³ at moisture contents of 10.63, 13.61, and 18.68, respectively.

The specimen size was prepared to fit a 10 cm by 10 cm shear box mold placed in a motorized apparatus running at a slow strain rate equal to 0.12 mm/min. The vertical stresses applied were 50 kPa, 100, kPa, and 200 kPa.

3.4. Consolidation and Swell Characteristics of the Clay–Sand Layers

The consolidation and swelling characteristics were determined for the different soils using compacted specimens on the one-dimensional oedometer device following ASTM D4546 and ASTM D2435-011 [28,29].

The compressibility properties for the six sets of samples were investigated to cover two clay–sand mixtures; namely 5% bentonite and 10% bentonite.

4. Test Results

All test results are presented in graphical forms. Figure 4, Figure 5 and Figure 6 present the DCP profiles for the tested mixtures at variable moisture test conditions. Figure 7 and Figure 8 present the one-dimensional consolidation results for the 5% and 10% bentonite content mixtures. Figure 9, Figure 10, Figure 11, Figure 12, Figure 13 and Figure 14 present the direct shear test results for the 5 and 10% bentonite mixtures. Table 2 provides a summary of the shear strength tests.

5. Discussion

5.1. Dynamic Cone Penetrometer

The dynamic cone penetrometer profiles obtained for two layers with different bentonite content (5% and 10%) indicate variations in three different measurements. These include the linearity, slope, and magnitude. In Figure 4, presenting the bottom row results, we can clearly observe that the material compacted at the optimum moisture content had a linear trend. This was validated in both the 5% and 10% bentonite content. The shape observed for the dry of optimum results indicates a bilinear trend in which a greater number of blows (beyond five) was required to advance the DCP for the 5% clay content. Further advancement of the cone was resisted by a pocket of sand grains closely packed after applying five blows of the hammer. A bilinear shape did not develop for the compacted material including 10% bentonite for the dry of optimum moisture content and the other two states of moisture.

The slope of penetration for the optimum moisture mix was intermediate between the dry of optimum and wet of optimum compacted layers of 5% bentonite content while it was flatter and less steep than the dry of optimum and wet of optimum conditions in the 10% bentonite mixture.

The number of blows required to advance the cone penetrometer 5 cm into the 5% bentonite mixture was in the range of 4 to 7 blows for the mixture compacted at optimum moisture conditions. For the dry of optimum and wet of optimum conditions, the ranges of blows were from 2 to 4 blows and from 7 to 10 blows, respectively. For the 10% bentonite mixture, the number of blows required to advance the cone penetrometer for 5 cm was in the range of 10 to 12 blows under the optimum moisture conditions. For the dry of optimum and wet of optimum conditions, the ranges of blows were from 7 to 8 blows and from 5 to 6 blows, respectively.

In Figure 5, which presents the central row, the results appear to be similar to those presented in Figure 4, except for an outlier point (at or >10 blows) observed for the 5% bentonite that deviated from the linear optimum state line.

Figure 6 presents the top row points. The results appear to be similar to those presented in Figure 4 except for a point of deviation from linearity (at or >10 blows) observed for the 5% bentonite. In addition, the slope of the optimum state is not necessarily flatter than the other moisture states in the 10% bentonite mixture layer.

Based on the above observations, we can see that the comparison between the DCP points data is useful and valid and had good accuracy when the material was homogeneous. When the clay content and the grain size distribution were different, the linearity and the shape and magnitude of penetration were variable. To rely on the DCP data, it is recommended to conduct grain size distribution tests to confirm the uniformity of the site material and to enable adjustments if necessary.

The state of moisture during compaction influences the outcome of the DCP with regard to the shape, slope, and number of blows. It is recommended to avoid compaction under a wet of optimum state.

Compactness and density also influence the number of blows. Stress portioning is a state in which clay supports part of the cone pressure and coarse grains support the other part. A high clay content may reduce stresses exerted on the coarse grains, and then a lower number of blows will be reported.

5.2. Consolidation and Swell Tests

Figure 7 and Figure 8 present the compressibility plots for six samples subjected to one-dimensional consolidation tests, which included the 5% and 10% bentonite mixtures. The samples were prepared under three different moisture states. Swell was hardly observed for the 5% bentonite mixtures, while a 1% vertical strain increase was observed for the 10% bentonite mixture. The dry of optimum water content reported the highest swell.

The slopes of the curves in the e-log plots of the consolidation tests determine the compression index and the swell index. The compression index and the swell index are parameters used in computing volume changes in clays when the stresses applied to the soil are known. Figure 7 indicates nearly parallel lines for the loading and off-loading paths, implying the same compression indices and swell indices for all moisture state conditions This could be due to the clay content being low and not having a significant influence on the compressibility or the swelling phase. In the 10% bentonite mixtures, the compression index was variable for low stresses until the stress level reached 200 kPa.

Table 3. Compression and swell index properties of clay–sand mixtures tested.

Mixture Bentonite Content	Moisture State	Compression Index (Cc)	Swelling Index (Cr)
5% B	DOMC	0.034	0.012
	OMC	0.033	0.011
	WOMC	0.031	0.010
10% B	DOMC	0.038	0.015
	OMC	0.032	0.011
	WOMC	0.026	0.011

presents the compression indices and swell indices for all tested samples. All tested samples appear to converge and assume an identical compression index. A slight variation in the swell index is reflected by the off-loading plots. Higher bentonite content is known to achieve lower hydraulic conductivity, but excessive bentonite can be a source of swell–shrink hazards. Dafalla [30] studied the compressibility and swelling of clay–sand mixtures in more detail. References [31,32,33] provide more information on the hydraulic conductivity of clay–sand mixtures similar to the materials presented in this study.

5.3. Shear Strength Tests

The shear strength tested in the laboratory for the 5% bentonite mixture, compacted at the optimum moisture content and subjected to 50, 100, and 200 kPa of vertical normal stresses, yielded an angle of shearing resistance of 33.5 degrees and a cohesion of 15.2 kPa. The friction angle increased to 38.6 and 36.4 degrees for the dry of optimum and the wet of optimum conditions, respectively. The cohesion was reported as 2.2 kPa and 6.5 kPa for the dry of optimum and wet of optimum conditions, respectively.

For the 10% bentonite content mixture compacted at the optimum moisture content and the maximum dry density, the angle of shearing resistance was measured as 33.5 degrees and the cohesion as 30 kPa. The friction angle did not change significantly and was measured at 33.4° and 34.0° for the dry of optimum and the wet of optimum conditions, respectively. The cohesion was reported as 22.5 kPA and 18 kPa for the dry of optimum and wet of optimum conditions, respectively.

The above parameters are useful when computing the allowable bearing capacity of the clay–sand layer. This may be needed if the layer is to support light structures.

Higher shear strength is associated with higher bentonite content. This is also reflected in the penetration of the dynamic cone in this particular study. The shear strength may be reduced if the bentonite content is very high, as the clay may dominate the behavior of the mixture [34]. Stress partitioning may be the governing factor, and closer contact between particles can be achieved by applying more compaction energy. The shape of the sand, whether rounded or angular, and the interlocking between the particles can have a significant influence on the shearing strength.

While shearing off the samples, vertical and horizontal displacements were observed. In examining the horizontal displacement, we can see that increasing the shear stress beyond the shear stress corresponding to 2 mm of horizontal displacement will result in excessive horizontal displacement for the two mixtures tested (5% and 10% bentonite content). The vertical displacement was high for the 200 kPa stress. Dilation did not occur at this stress for the 5% bentonite.

5.4. Contours of Equal DCP Penetration

The evaluation of homogeneous clay–sand layers can be performed using dynamic cone penetration soundings, which provide useful and quick information about the status of the layer quality with regard to the density and shear strength. For efficient assessment, a grid of points with about 10 m spacings is created, and the penetration obtained is recorded for a fixed number of blows. Alternatively, a suggested penetration level is defined with a mark on the penetration tool, and then the number of blows needed to achieve this penetration is counted. This method can enable plotting a contour drawing showing the penetration or the number of blows. Very dense, dense, and loose zones can be identified and labeled. Fine-tuning can be made by closer grids of 5 m or smaller for areas showing large variations or unusual measurements.

It is known in practice that the control of compaction works is based on the deviation from the optimum moisture content. The compaction effort needed to achieve the maximum dry density is low when the material is compacted at the optimum moisture content. However, after compaction, weather conditions can affect the moisture content within the top soils, but the dry density remains unchanged. The use of the dynamic cone penetrometer as a predictive and measurement tool is widely accepted in geotechnical engineering but needs to be enhanced by additional testing in order to correct and refine the data.

The required testing includes moisture profiling and grain size distribution. Grouping the data of equal moisture content can help with comparisons. The grain size distribution can introduce the clay content, which has an impact on the number of blows. The correction factor can be based on a chart of different clay contents prepared for each site. For example, for an adjustment of a zone showing 10% clay content compared to a zone with 5% clay content, an average factor of 2.1 can be applied based on the outcome of this study. Figure 15 shows the contour lines for an equal number of blows for 5% bentonite mixture and 10% bentonite mixture.

6. Conclusions

Geotechnical and geo-environmental engineers require a cheap and quick approach for the quality assessment of the topsoil conditions for landfill liners and/or clay–sand layers. The hand-held dynamic cone penetrometer was found to be very effective for this task.

The dry density and shear strength evaluations can be based on the collected data, the penetrometer tool used, and correlation factors. These factors can be determined for a site material in a laboratory. This paper suggests performing moisture profiling and grain size distribution tests to supplement and refine the outcome of the cone penetrometer mapping. Grouping the data of equal moisture contents can help with comparisons. The grain size distribution can determine the clay content, which has an impact on the number of blows. The correction factor can be obtained based on a chart of different clay contents prepared for each site.

The state of moisture during compaction influences the DCP with regard to the shape, slope, and number of blows. It is recommended to avoid compaction under a wet of optimum state.