Abstract
Crushed stone aggregate materials are commonly used for basal embankment support on soft subgrades or for horizontal drainage in soft soils. When the drainage material is constrained by earth pressure and water pressure, it changes to an unsaturated water-permeable state. If the pore water pressure decreases as the compaction of the soft ground increases, the water does not drain through the drainage material. Therefore, it is necessary to measure the critical pressure value (air entry value) going from saturation to desaturation. That is, even if the pore water pressure is small, the water must be discharged through the drainage material. Therefore, this study presents experimental results for gravity drainage and soil-water retention curves (SWRCs) using mixed crushed stone aggregate samples at different particle size distributions and different compaction conditions. SWRC experiments indicate that porosity, hydraulic conductivity under saturated conditions, and air entrapment decrease with increasing sand content. Also, the samples retained more water at higher suction values, although the porosity decreased with increasing sand content. This phenomenon is due to the tendency of the uniformity coefficient to increase as more fine particles are added to the crushed stone aggregate. In addition, it is considered that the function as a drainage material is possible only when it shows a higher air entry value than that of the loose crushed stone aggregate drainage material. Therefore, for drainage materials using crushed stone aggregate, analysis of the soil-water retention curve for unsaturated soil should be preceded.
1. Introduction
One approach for facilitating the reclamation of soft soils is with a radial drain. As shown in Figure 1, the vertical drains are connected at the soil surface with a horizontal drain made from crushed gravel. Although these soils function well as drainage layers, the mechanical behavior of the gravel layer is often enhanced through groundwater drainage during consolidation of the soft sediments.
Horizontal drain layers permit water expelled from consolidated soft soils to be discharged laterally. Numerous studies have been conducted regarding this phenomenon. The unsaturated properties of recycled materials used in the ground are similar to those of natural aggregates [1], and the effect of the degree of saturation during compaction is important [1] for wet crushed stone. In road construction, the characteristics of crushed stone have been studied as lack of natural aggregates [2]. However, environmental risk assessment of recycled construction products is required [3] and can be used only when certain criteria are satisfied. Accordingly, studies such as statistical analysis of harmful pollutants for recycled aggregates and natural materials [4], and test methods to determine the suitability of recycled aggregates according to moisture absorption rates using recycled aggregate concrete were conducted [5]. However, the mixed design of the aggregate is very important, the coarse aggregate forms a skeleton in the mixture, and the fine aggregate mainly fills the voids. That means, the function varies according to the size composition of the particles [6]. Also, quality control criteria in Korea are also ambiguous. It was pointed out that the problem of the use of crushed stone aggregate is considerably larger than the particle size, and clogging may occur [7]. To solve this problem, the permeability of crushed stone aggregate suitable for soft ground and the drainage performance, according to the change of pore water pressure were evaluated [8]. Also, in order to present the quality control criteria of crushed stone aggregate, the quality control criteria of horizontal drainage material suitable for soft ground was suggested through model experiments on the clogging phenomenon [9]. The Korea Expressway Corporation [10] presented the required discharge capacity standards for horizontal drain layers. Lee et al. [11] suggested a nonlinear analysis method considering a horizontal drain. Lee et al. and Kim et al. [12, 13] developed a compaction analysis method that considered the discharge capacity of the horizontal drain layer and studied the required discharge capacity. Kim et al. [14] analyzed possible factors affecting discharge capacity under unsaturated conditions, including air bubbles in the horizontal discharge materials. Kim et al. [14] compared the discharge capacity and settlement speed of recycled aggregate and crushed stone for soft ground treatment with those of a sand mat to investigate their utility as horizontal drain materials. Considering copper slag as drain material, Jang et al. [15] evaluated the promising engineering characteristics of copper slag as a substitute material for sand and considered soft ground improvement materials. Chun [16] assessed discharge capacity degradation in horizontal drain layers using a model experiment with rainwater inflow, sediments, and recycled aggregate. Regarding the practical design of sand mats that are responsive to the settlement characteristics of soft ground, Kim et al. [17] investigated conventional sand mat design methods by evaluating settlement amounts and drainage characteristics through numerical analysis of an embankment constructed on soft ground. In a subsequent study, Lee et al. [18] considered the use of recycled aggregate as horizontal drain layer material and compared the engineering characteristics of recycled aggregate and sand. In addition to being more economical than sand, Lee et al. [19] determined that recycled aggregate effectively satisfied the criteria for horizontal drain layers and were a competitive drainage alternative to sand.
The drainage of groundwater from the consolidating soil can be important. The rising groundwater caused by consolidation and other means must be quickly expelled to prevent degradation in the mechanical performance of any roads or structures. In South Korea, sand mats are the primary material used for drainage layers. However, the high cost and insufficient supply of sand are encouraging use of alternative materials such as crushed stone. Existing research and performance verification of crushed stone is insufficient. The combined drainage system shown in Figure 1 promotes consolidation and ground stabilization through the discharge of excessive soft ground pore water through vertical and horizontal drain materials. The drainage characteristics of this system can be effectively represented using conventional saturated soil mechanics theories prior to settlement. Once settlement of the horizontal drain layer occurs due to the soft ground consolidation, partially unsaturated areas may be generated and affect the water flow, making it difficult to analyze drain characteristics. Therefore, the characteristics of any unsaturated areas must be analyzed using the unsaturated soil mechanical approach. Requiring further investigation is the engineering characteristics of the mixed crushed stone and sand soil, the saturated and unsaturated soil mechanics approaches following soft ground settlement, and the drain state based on pore water and negative pressures.
This study used sand mixed with crushed stone aggregate as horizontal drain material for soft ground. The grain size distribution of the crushed stone mat was determined through physical tests; the coefficient of permeability was determined through drain capacity tests. In addition, unsaturated soil tests were conducted to analyze the soil-water characteristics based on the grain size distribution and degree of compaction. These collective tests supported the analysis of the mechanical properties of saturated soil in the horizontal drainage layer of the soft ground and the mechanical properties of the unsaturated soil after subsidence of the horizontal drainage layer.
2. Experimental
2.1. Material Characteristics
The materials used in this study were representative of the subbase materials used in South Korea and included 25 mm diameter crushed stone aggregate and sand. The crushed stone aggregate included screenings—a <5 mm diameter secondary product that results when the aggregate is produced and that contains substantial amounts of lime.
The crushed stone aggregate material was physically characterized as poorly graded gravel (GP) based on the Unified Soil Classification System (USCS). The grain size distribution of the aggregate was poor; the 4.76 mm sieve passing quantity was <50%, and the 0.074 mm sieve passing quantity was <5%. The sand was characterized as poorly graded sand (SP) based on the USCS. The grain size distribution of the sand was also poor; the 4.76 mm sieve passing quantity was <50% and the 0.074 mm sieve passing quantity was <5%.
Tables 1 and 2 list the physical properties of the two materials including the specific gravity (Gs), the percent grain size distribution, the USCS designation, and the maximum material unit weights based on different compaction conditions (loose, normal, and dense). Figure 2 graphically depicts the grain size distribution of the two materials used in this study.
2.2. Sample Preparation
A series of different samples was prepared using different proportions of crushed stone aggregate and sand. Table 3 summarizes the various mix proportions as well as the grain sizes based on different compaction conditions. The samples were compacted using a consistent energy level. To minimize separation during compaction due to the different specific gravities of the two materials, the crushed stone aggregate was added before the sand and the mixed layer was densified by tapping the side of the soil container.
Table 4 summarizes the void ratios for select samples. The void ratio generally decreases as the sand content and level of compaction (from loose to dense) increases. The void ratio was highest (1.027) for the loosely compacted, 100% crushed stone aggregate sample (CS-1), and was lowest (0.647) for the densely compacted, 100% sand sample (CS-08).
3. Tests and Experimentation
3.1. Maximum and Minimum Unit Weights
Several different experimental methods have been proposed for determining maximum and minimum material unit weights; the results vary greatly depending on the method used. The method specified in the Korean Standard [21] is similar to the method specified in the American Society for Testing and Materials [22]. The maximum unit weight can typically be determined by applying vibration to the sample or dropping the sample from a specified height. In this study, however, the Bowles method of applying energy using a horizontal blow was used because it has been previously shown to produce higher unit weight values than the ASTM method [17].
To determine the maximum unit weights, dry samples were inserted into molds with a diameter of 150 mm and a height of 350 mm. Only a subset of samples was tested. Specifically, crushed stone aggregate-sand mix proportions of 100-0% (CS-1), 90-10% (CS-2), 80-20% (CS-3), 70-30% (CS-4), 50-50% (CS-6), and 30-70% (CS-8) were considered under various compaction conditions. To simulate loose, normal, and dense compaction conditions, 50, 250, and 500 horizontal blows were applied, respectively, using two rubber hammers (0.62 kg). The unit weights were subsequently measured.
3.2. Soil-Water Characteristics
To determine the soil-water retention curves (SWRCs) for the samples considered in this study, an experimental apparatus was fabricated by modifying a pressure plate extractor. The pressure plate extractor uses a ceramic disc with a large air entry suction value. When unsaturated soil is placed upon the disc and air pressure is applied, the negative pore water pressure can be determined by measuring the volume of water extracted.
The attractive force of the soil is identical to the air pressure on the disc because the soil maintains the same conditions as the air. The air pressure thus becomes the negative pore water pressure; the volume-water ratio is determined using the water volume extracted by this pressure. Figure 3 shows a schematic diagram of a typical pressure plate extractor.
The pressure plate extractor is designed to adjust the attractive force by allowing changes to the air pressure from the top of the apparatus. One shortcoming of this device is that it may allow water from the ceramic disc to flow into the soil when the top plate is removed (at the conclusion of a test). As a result, considerable error may be introduced when deriving the SWRCs. To reduce the potential for error, Figures 4 and 5 show a proposed modification to the apparatus in schematic and constructed formats, respectively.
The ceramic disc was attached to a brass frame and a burette was added to measure the water volume flowing to the bottom. The greatest advantage of this modified apparatus is that it measures changes in water volume movement based on changes in attractive force without opening the top plate. Further, it minimizes the water flowing from the disc into the soil because of decompression. Even at high pressures and with only small changes in attractive force, water movement and outflow can be easily measured.
In this study, the pressure plate extractor was used at low pressures while a filter paper method was used at high pressures. When using the pressure plate extractor, samples were fabricated and carefully placed such that they made good contact with the ceramic disc on top of the apparatus. The measurement cell at the bottom of the apparatus was filled with water to remove any air. After confirming the adequacy of contact, the disc-sample interface was saturated with water supplied by a tube from the bottom of the apparatus. The volume of water supply was measured in advance when considering changes in the soil volume to water content ratio. When the burette scale indicated a sufficient water volume for saturation, the test was initiated. Changes in the burette scale measurements were monitored while adjustments were made to the air pressure from the top of the apparatus to obtain the specified attractive force. The range of pore air pressure used in this experiment was 0.1–1000 kPa.
3.3. Unsaturated Soil Characteristics
For a subset of samples, unsaturated soil tests were performed. Specifically, samples included crushed stone aggregate-sand mix proportions of 100-0% (CS-1), 90-10% (CS-2), 80-20% (CS-3), 70-30% (CS-4), 50-50% (CS-6), and 30-70% (CS-8). These samples were selected to investigate the effects of increased or nonexistent sand content on unsaturated soil characteristics. The level of compaction was also considered as an influencing factor.
4. Results and Discussion
4.1. Grain Size Distribution Based on Mix Proportion
Figure 6 graphically depicts the grain size distribution for the varied mix proportion samples considered in this study under different levels of compaction. Figure 7 presents the coefficients of uniformity and curvature. In general, higher sand contents had a greater effect on the coefficients of uniformity and curvature. Lower sand contents combined with lower compaction levels had a lesser effect on the coefficients of uniformity and curvature. Comparatively, normal and dense compaction conditions had a greater effect on the coefficients of uniformity and curvature, even when sand contents were low. None of the sand mix proportions considered in this study simultaneously satisfied the coefficients of uniformity and curvature. Two homogeneous materials that are mixed may appear uniform because of voids caused by varying grain diameters.
(a)
(b)
(c)
4.2. Maximum and Minimum Unit Weights
Figure 8 shows the maximum and minimum unit weights for select samples considered in this study. Specifically, crushed stone aggregate-sand mix proportions of 100-0% (CS-1), 90-10% (CS-2), 80-20% (CS-3), 70-30% (CS-4), 50-50% (CS-6), and 30-70% (CS-8) were considered under various compaction conditions. Unit weights consistently increased as sand contents and compaction levels increased. As the sand content increased, the unit weight increased and the void ratio decreased to maintain the same relative density, which is the same result as in the study conducted by Moon et al. [8]. The fine particles are included in the void between the sand particles, and the total void ratio decreases as the content of fine particles increases. A change in the total void ratio according to the content means a change in the unit weight Lade et al. [10].
4.3. Change of Coefficient of Permeability According to Discharge Pressure and Compaction
The coefficient of permeability was derived after the discharge capacity test was performed according to the change of discharge pressure shown in Figure 9. When the compaction condition was normal, the coefficient of permeability generally increased together with the discharge pressure. The increase of the coefficient of permeability when the discharge pressure changed from zero to 150 kPa was greater than the increase when it changed from 150 kPa to 300 kPa. The reason for this seems to be that the void ratio tends to be greater or smaller than expected due to the nonuniformity of crushed stone and sand grains. This factor influenced the coefficient of permeability.
(a)
(b)
(c)
(d)
The discharge pressure was adjusted to derive the coefficient of permeability according to compaction condition shown in Figure 9(d). The coefficient of permeability generally tends to decrease as the density and sand content increased. This suggests that the nonuniformity and the decrease of void ratio due to the mixing of samples affected the coefficient of permeability. Furthermore, the coefficient of permeability increased as the discharge pressure increased when the density was constant, while the crushed stone (100%) showed the greatest coefficient of permeability at the loose, normal, and dense conditions. The coefficient of permeability tended to decrease as the sand content increased at the constant pressures of 150 kPa and 300 kPa; however, no regularity was observed in the coefficients of permeability at different compaction conditions. The reason for this seems to be that even though the compaction condition was adjusted by horizontal blows, the pressures were greater than the forces of blows during the sample preparation that changed the arrangement of grains.
5. Maximum and Minimum Unit Weights
5.1. Soil-Water Characteristics
Soil-water retention characteristics considering the attractive force and volume-water ratio were determined for select varied mix proportion samples used in this study. The concept of the volume-water ratio was introduced to accurately account for the water content in the sample by factoring in both weight and volume parameters. Figure 10 presents the sample SWRCs for crushed stone aggregate-sand mix proportions of 100-0% (CS-1), 90-10% (CS-2), 80-20% (CS-3), 70-30% (CS-4), 50-50% (CS-6), and 30-70% (CS-8) under various compaction conditions. Initially, the volume-water ratio decreased as the sand content increased. The change in the content of mixed soil is affected by the surface force of soil particles that adsorb water molecules in the void [8]. Therefore, with the increase of the sand content, the volume-water ratio also decreased due to the decrease in the ratio of fine-grained soil. However, over time, the residual soil-water ratio increased. This phenomenon may be explained by the limited water volume movement in the pores, particularly in the densely compacted samples. This is because when the matric suction increases, the amount of air in the void increases, which obstructs the flow of water [23].
(a)
(b)
(c)
5.2. Soil-Water Retention Curves Based on Compaction Condition
The attractive forces for the same set of samples above (CS-1–4, CS-6, and CS-8) were measured under various compaction conditions and graphically correlated with the volume-water ratio to determine the effects of compaction on the soil-water characteristics. Figure 11 shows the sample SWRCs. When the sample contained only crushed stone aggregate, the residual soil-water ratio clearly varied with the degree of compaction. Changes in water movement resulting from increased attractive forces were similar for normal and dense compaction conditions but were significantly different under loose compaction conditions. Shin [24] reported that there was almost no difference in the air entry value regardless of the fine grain content, and the residual water content was larger in the sand without fine grains. When sand was mixed with the crushed stone aggregate, variation in the residual soil-water ratio decreased compared to the sample with crushed stone only. Also, for the same volumetric water content, the soil with higher unit weight has a greater matric suction value [25]. In this study, the residual volumetric water content increased as the sand content increased. Therefore, when selecting a horizontal drain material, the effect of material pore sizes and compaction on the soil-water characteristics must be considered.
(a)
(b)
(c)
(d)
(e)
(f)
6. Air Entry Suction
6.1. Air Entry Suction Based on Mix Proportion
Table 5 summarizes and Figure 12 graphically depicts the air entry suction values for the same set of samples above (CS-1∼4, CS-6, and CS-8) that off varying mix proportions. In general, the air entry suction value decreased as the sand content increased. Moreover, the air entry suction value was greater in samples that contained sand in the mix than samples that did not. This phenomenon is not surprising because sand is associated with water movement.
For sample CS-1 (100% crushed stone aggregate), the air entry suction value decreased as the level of compaction increased. For samples CS-2, CS-3, and CS-4, the varying mix proportions, subsequently varying pore sizes, and degree of compaction significantly affected the attractive force’s water exclusion capacity. The highest air entry suction value was observed for samples containing only 10% sand. As noted from Table 5, air entry suction values ranged from 53.38 (CS-2) to 39.19 kPa (CS-8) under loose compaction conditions. These air entry suction values exceeded those for normal and dense compaction conditions. In general, the air entry suction value decreased as the level of compaction and the sand content increased.
It shows the highest air entry value in the loose state. The reason may be because the numerous micropores in the surface, which tend to absorb more water [26], and the high percentage of fines in recycled concrete aggregates may also contribute to high air entry value.
Saturated soil gradually changes to an unsaturated state as the soil-water ratio decreases because of external pressure or dead weight. The critical pressure value at which saturated soil changes to unsaturated soil is called the air entry suction. Therefore, air entry suction is the most important parameter when analyzing unsaturated soil. As consolidation of the soft ground progresses, excessive pore water pressure builds and, as a result, water discharges through the horizontal drain material. However, as the degree of consolidation levels off and the pore water pressure decreases, water no longer discharges through the horizontal drain material. Kim and Shin [23] stated that the permeability coefficient of a horizontal drainage material is not infinite, but a limited value, and that it causes a delay in consolidation and deteriorates the drainage function. Therefore, it is necessary to understand the flow of unsaturated state in the soil-water characteristics curve according to soil mixing. This is because the soil-water characteristics curve is greatly affected by the soil structure according to the particle size [27]. Under these low pore water pressure conditions, steps must be taken to ensure continued water discharge through the horizontal drain material. Under these low pore water pressure conditions, steps must be taken to ensure continued water discharge through the horizontal drain material.
A high air entry suction value suggests that a high pressure is required to change the soil from saturated to unsaturated conditions. In an unsaturated condition, the crushed stone mat acting as horizontal drain material cannot effectively discharge water.
This is because, when unsaturated, residual air bubbles inside the drainage material interfere with the flow of water. Park [28] evaluated that the air dissolved in the pore water was reduced to air bubbles during the drainage process, affecting the water drain ability. Miura [29] evaluated the water permeability of residual air bubbles using an air bubble generator and said that a decrease in drainage of about 20% occurred. Jang et al. [15] stated that the drainage material becomes unsaturated due to the inflow of air bubbles contained in the pore water of the ground after the installation of the drainage material, and that the state of unsaturated affects the drainage ability. Therefore, in the case of using crushed stone mixed aggregate (uniform terminology) as a drainage material, it is considered that a certain amount of sand is mixed to increase the air entry value, and smooth drainage can be induced by pore saturation.
7. Conclusions
In this study, various samples were prepared using different mix proportions of crushed stone aggregate and sand with an intended application as horizontal drain material for soft ground. The grain size distribution was determined through physical tests. The coefficient of permeability was determined through drain capacity tests. In addition, soil-water characteristics and air entry suction values of a crushed stone mat were analyzed based on mix proportion and level of compaction through unsaturated soil tests. The saturated and unsaturated soil mechanics characteristics that can appear in a horizontal drain layer and during settlement of a horizontal drain layer, respectively, were analyzed. The findings of this study are summarized below.(1)As the sand content increased, the coefficients of uniformity and curvature initially increased and then subsequently decreased. The coefficient of uniformity was highest when the sand content was 60–70%.(2)Maximum unit weights of 16.07 kN/m3, 15.71 kN/m3, and 13.75 kN/m3 were observed for the sample comprising 100% sand under loose, normal, and dense compaction conditions, respectively. Comparatively, minimum unit weights of 13.01 kN/m3, 14.65 kN/m3, and 15.38 kN/m3 were observed for the sample comprising 90% crushed stone aggregate and 10% sand under loose, normal, and dense compaction conditions.(3)In general, the coefficient of permeability increased as the discharge pressure increased but decreased as the compaction and sand content increased. When the discharge pressure was constant, the coefficient of permeability again decreased as the compaction and sand content increased. When the level of compaction was constant, the coefficient of permeability increased as the discharge pressure increased.(4)The volumetric water content increased as the level of compaction decreased. Furthermore, as the sand content of sample increased, the volumetric water content decreased, but the residual water content increased.(5)The air entry suction value decreased as the level of compaction and sand content increased. Decreased void ratios attributable to increased compaction and sand content explain this observation.(6)When only the coefficient of permeability is considered, the sample comprising 100% crushed stone aggregate is the superior horizontal drain material. However, even with a high coefficient of permeability, crushed stone aggregate will ineffectively function as horizontal drain material because its air entry suction value is low. Therefore, additional soil-water characteristics must be considered when selecting a horizontal drain material. . In other words, it is thought that it is necessary to identify the minimum air entry suction value from the soil-water characteristics curve of the material used and to design the load for each step that can increase the pore water pressure.
Data Availability
The data used to support the findings of this study are included within the article.
Conflicts of Interest
The authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript.
Authors’ Contributions
All authors have participated in the conception and design of the study, analysis and interpretation of the data, drafting the article, revising it critically for important intellectual content, and approval of the final version.