Stabilization of clayey subgrade with waste pumice for road infrastructure

1 Introduction

Road pavements are layered systems with better quality materials on top where the intensity of stress is high and inferior materials at the bottom where the intensity is low. Typical bituminous pavements are composed of wearing course, base course, subbase, and subgrade layers, as seen in Figure 1 [1]. The road base or the ground on which the road lies is called the subgrade. Performance of roads depends on many factors that include the subgrade soil type, the density of traffic, and even the weather conditions the road is subjected to. The subgrade must be able to support loads transmitted from the pavement structure. This load-bearing capacity is often affected by the degree of compaction, moisture content, and especially soil type. Some clay-type soils shrink and swell depending on their moisture content, whereas soils with excessive fines may be susceptible to frost heave in freezing areas [2]. Types of pavement subgrade soil affect the selection of road project routes in view of problematic characteristics. In such cases, either the highway route is changed or the layer that has problematic soil is replaced with a new layer that has the proper soil type. Sometimes, these kinds of problems necessitate increasing the route length of a road construction project and hence project costs. Selection of an appropriate subgrade stabilization process to improve subgrade resistance to permanent deformation is very important for overall pavement performance [3].

Figure 1

Typical cross-section of a conventional flexible pavement [1].

The method of stabilization is well known and has been used throughout the world for many decades in improving some soil properties [4]. Some of the materials used for soil stabilization are sea peel, pumice, froth concrete, thermic power station ash, fly ash, volcanic ash, lime, cement, auto tire pieces, industrial churn, marble powder, gravel, and rubber [4–19].

Stabilization and solidification refer to treatment processes that are designed to accomplish one or more of the following objectives [20]: a) improve the handling and physical characteristics of the waste, b) limit the mobility of toxic constituents of the waste, and c) render the waste as a usable material in construction.

Pumice is an amorphous foam produced during volcanic eruptions. It is composed mostly of silica and alumina in relative amounts according to the geological area of origin, and it also includes other chemical elements, such as different oxides and water [21]. Pumice is a volcanic rock of which the porous structure is formed by dissolved gases being precipitated during cooling as the lava hurtles through the air. Connectivity of the pore structure may range from completely closed to completely open [22].

There is plenty of pumice formed during explosive volcanic eruptions in Turkey, especially in Isparta. Turkey has seven billion m³ total pumice reserve [23]. Pumice is used in the fields of construction, chemical industry, agriculture, and dentistry. Particles larger than 3 mm are used in these sectors but smaller pumice particles are stocked in the field.

The objectives of soil stabilization with additives are to improve volume stability, strength and stress-strain properties, permeability, and durability. The development of high strength and stiffness is achieved by reduction of the void space, by bonding particles and aggregates together, by maintenance of flocculent structures, and by prevention of swelling. Uniform mixing process is the most important factor affecting the quality of results [24].

In the western part of the Netherlands, the subgrade consists of clayey materials. In the construction of major roads in this area, the subgrade is replaced to a depth of at least 50 cm by sand material. The sand is either dredged from river estuaries or produced on land in sand pits [25]. Soil stabilization is an efficient technique that has been used for many years in solving soil problems in civil engineering projects [26]. When stabilized materials that meet all gradation, durability, and strength requirements are utilized in pavement structures, soils that have been mixed with a stabilizing material are considered modified. Stabilization of high plasticity clayey subgrade with pumice may prevent the pavement from swelling.

Tremblay et al. [27] reported the results of a laboratory study in which 13 different organic compounds were added separately into two different soils, which were then treated with 10% cement. They found that oils and hydrocarbons, which are insoluble in water, delayed cement hydration but did not affect the final strength. Cimen [28] stated that pumice is a suitable additive material for stabilization of high plasticity clay. Kavlak [29] suggested that the Isparta-Gelincik pumice can be used for mechanic stabilization of problematic highway subgrades because he observed that the strength of these subgrades increased with pumice stabilization. In another study, Cimen [30] blended clay with marble dust, pumice, and lime. After the optimum dosages for each additive were determined, the samples were prepared at maximum dry density and the optimum water contents were cured for between 7 and 28 days. Unconfined pressure tests, triaxial pressure tests, and California bearing ratio (CBR) tests were performed on the samples. By evaluating the results obtained in situ and from laboratory experiments, the feasibility of using waste materials in soil improvement and their applicability in the field was investigated.

Bin-Shafique et al. [31] investigated the long-term performance of fly ash-stabilized low-plasticity clay and high-plasticity expansive clay soil subbases. Plasticity index tests, unconfined compression tests, and vertical swell tests were performed on the specimens for 7-day curing, wet-dry cycles (with tap water and saline water), and freeze-thaw cycles. These implied that stabilized soils were at least three times higher than unstabilized soils. Seco et al. [32] used six different waste additives, lime, and Portland cement for expansive soils. Of the waste materials, the most notable was the behavior of rice-husk fly ash, which was found to be highly effective in stabilizing expansive soils. Cetin et al. [33] conducted CBR and resilient modulus tests on sample mixtures of fly ash and lime kiln dust. These sample mixtures showed lower base thicknesses. The authors suggested that construction costs can be reduced by stabilizing road surface materials with high carbon fly ash.

In the present study, stabilizing fine-grained subgrade soils with high clay content using the Isparta-Karakaya pumice waste was investigated. Pumice waste in different ratios was blended with problematic clay to improve the mechanical properties of the soil.

2 Materials and methods

During the Paleocene and Quaternary periods, pumice deposits in Isparta and its neighboring regions were formed after eruptions of the Gölcük volcano, which is located southwest of Isparta, Turkey. There are four sources of pumice in Isparta, namely, the Karakaya, Gölcük, Gelincik and Aliköy zones, and the pumice from each source exhibits different properties. In the present study, pumice was obtained from the Isparta-Karakaya region, which spans an area of 1940 hectares. Its apparent reserve is about 0.8 billion m³, with a possible reserve of 1.2 billion m³, and a total reserve of about 2 billion m³ [34].

Isparta-Karakaya pumice is generally used for lightweight briquette production. Particles that can pass through a No. 40 sieve are not used in the production process. This group of pumice is called pumice waste. The amount of the Isparta-Karakaya pumice waste is estimated to be about 20,000 tonnes.

The pore structure of the Isparta-Karakaya pumice is shown in Figure 2. The pore perimeter is surrounded by ferrous material. In Figure 3, there are black opaque minerals, such as pyrite, magnetite, chalcopyrite, and hematite, among others. Amorphous materials surround the opaque minerals. Pyroxene and feldspar crystals within the pumice are shown in Figures 4 and 5, respectively.

Figure 2

The pore structure of the Isparta-Karakaya pumice.

Figure 3

Opaque minerals within the Isparta-Karakaya pumice.

Figure 4

Pyroxene feno crystals within the Isparta-Karakaya pumice.

Figure 5

Zoned feldspar crystal within the Isparta-Karakaya pumice.

There are very few studies on the use of pumice in the construction of highway pavements. To date, some lightweight aggregates, with the exception of pumice, have been used for various purposes. Bottom ashes were applied in wearing courses, surface treatments, base courses, road bases, and subbase layers after being crushed as granular material [35]. Coal products were integrated in subbase layers as a stabilization material [36]. These products were utilized as fill in embankments, in subbase and base courses, and as aggregate in concrete and hot mix asphalt mixtures [36, 37]. Lime-pulverized fly ash slurry that is injected under pressure was employed to stabilize embankments and levees. Increases in strength by 15%–30% were reported [38].

2.2 Experiments conducted on clay and pumice

Classification and standard compaction tests were conducted on the clay samples. The hydrometer results show that the pumice is composed of 58% clay and 42% silt. The liquid limit is 126%, plastic limit is 47%, and plasticity index is 79%. Therefore, the soil sample is high-plasticity clay (CH) based on the Unified Soil Classification System. X-ray analysis was performed to determine the mineralogical factors of the soil sample at the Laboratory of General Directorate of Mineral Research and Exploration. A sample was subjected to normal inflection at 2.5–70°C and to inflection with saturated ethylene glycol at 2.5–40°C. Excessive clay minerals were determined. The results show that the dominant clay minerals in order of high presence ratio are montmorillonite, kaolin, a few illit, quartz, feldspar, and a few mixed-layer clay minerals.

Standard compaction tests were done on the clay and pumice samples according to the ASTM standards. The dry unit weight and water content relations are shown in Figure 6. The maximum dry density and optimum water content of the clay samples are 1.19 g/cm³ and 38%, respectively, whereas the maximum dry density and optimum water content of the pumice samples are 1.07 g/cm³ and 16%, respectively. The optimum water content of pumice is lower than that of clay. These measurements suggest that the possible optimum water content value of clay-pumice mixtures is less than that of natural clay.

Figure 6

Results of standard compaction tests on the clay and pumice samples.

2.3 Experiments conducted on the pumice-clay mixture

Coarse-grained soil has greater pore structure than does fine-grained soil. For this reason, vibroflotation, dynamic compaction, and compaction piles are methods suitable for stabilization. The compaction characteristics of fine-grained soil are largely governed by its moisture content. Compared with those of coarse-grained fills, the properties of cohesive fills depend to a greater extent on the placement conditions [42].

Standard compaction, unconfined compression, and swelling pressure tests were conducted to determine the effects of pumice. In the standard compaction tests, clay samples that could pass through a No. 4 sieve and pumice samples that could pass through a No. 40 sieve were used. In the unconfined compression and swelling pressure tests, clay and pumice samples that could pass through a No. 40 sieve were used. Pumice was mixed into the clay samples in ratios of 10%, 15%, 20%, 30%, 40%, and 50% by weight in the tests. Atterberg limits tests were performed on the pumice-clay mixture samples. From these tests, the shrinkage limit, liquid limit, and plastic limit of the mixture samples were determined [43]. The results are given in Table 4.

Table 4

Standard compaction conditions and results of unconfined compressive strength, swelling pressure, and Atterberg limits tests [44].

Tests Samples	LL (%)	PL (%)	SL	γdmax (g/cm3)	Wopt (%)	qu (kg/cm2)	Ps (kg/cm2)
Clay	126	47	37	1.19	38	2.06	3.10
Pumice	35	–	7	1.07	16	0.54	–
%10 pumice+%90 clay	102	38	25	1.23	34	3.00	2.85
%15 pumice+%85 clay	96	35	21	1.24	32	3.60	2.75
%20 pumice+%80 clay	82	34	17	1.26	31	3.70	2.50
%30 pumice+%70 clay	72	28	13	1.36	28	4.44	2.25
%40 pumice+%60 clay	72	28	12	1.42	25	4.00	2.25
%50 pumice+%50 clay	66	27	8	1.46	27	3.50	2.00

LL, liquid limit; PL, plastic limit; SL, shrinkage limit; γ_dmax, maximum dry density; w_opt, optimum water content; q_u, unconfined compressive strength; P_s, swelling pressure.

Each sample for the unconfined compression and swelling pressure tests was prepared according to its own maximum unit weight and maximum water content values. For the unconfined compression test, the clay sample and the pumice-clay mixture samples were compacted in a standard compaction mold. The clay-pumice mixture samples were prepared in a ring 5 cm in diameter and 10 cm in height. Vertical load was applied on the soil samples until they could no longer support any additional load. The highest vertical stress is defined as the unconfined compressive strength of soil (q_u). For the swelling pressure test, the clay sample and the pumice-clay mixture samples were compacted in a standard compaction mold and were taken using an oedometer ring 71.4 mm in diameter and 16.3 mm in height. Swelling pressure was determined by constant volume tests according to the ASTM standards on the use of one-dimensional oedometer apparatus. The procedure for the swelling pressure test involved inundating the sample in the oedometer while preventing the sample from swelling. The swell pressure is reported as the maximum applied stress required in maintaining constant volume. Table 4 shows the maximum dry unit weight and the optimum water content values for which the samples were prepared and the values for the unconfined compressive strength.

The unit weight of the compacted sample was calculated using Equation 1:

(1)γcomp=γd max(1+wopt) (1)

where γ_{d max} and w_opt are maximum dry density and optimum water content respectively.

In addition, CBR tests were performed on the clay-pumice mixture samples. The results are shown in Table 5. The CBR degree increased when the pumice ratio was increased to 40%. However, the CBR degree decreased when the pumice ratio was increased further to 50%.

Table 5

CBR tests results.

Tests samples	CBR (%)
Clay	0.95
%10 pumice+%90 clay	1.10
%20 pumice+%80 clay	2.50
%30 pumice+%70 clay	3.80
%40 pumice+%60 clay	4.00
%50 pumice+%50 clay	3.50

The relationship among the amount of pumice in the mixture, compacted unit weight, unconfined compressive strength, swelling pressure, and CBR is shown in Figure 7. With the increasing amount of pumice in the mixture, a higher dry unit weight was obtained for lower water contents, whereas decreased liquid and plastic limits were observed. As q_u increased, low unconfined compressive strength was observed only on pure pumice sample, whereas increased unconfined compressive strength was recorded for the pumice-clay mixture samples. However, the unconfined compressive strength decreased for the pumice-clay mixtures for pumice ratios 40% and above. The swelling pressure values decreased with increasing pumice ratios. The CBR values increased with increasing pumice ratios above 40%.

Figure 7

Effect of pumice ratio at the mixtures.

(A) q_u, unconfined compressive strength. (B) γ_comp, compacted unit weight. (C) CBR. (D) P_s, swelling pressure.

The stabilization process is due to the mechanical performance of the clay-pumice mixture. Puzzolanic reaction was not observed for these mixtures. The changes in the plasticity index are due to the addition of a non-cohesive fraction to the material. A chemical change in the clay was not observed. In this study, the pumice from the Isparta-Karakaya region was used. It was found that the Karakaya pumice has no puzzolanic effect, whereas the volcanic slag aggregates have puzzolanic effects.

3 Conclusions

In this study, stabilization of high-plasticity clayey subgrade of road pavements using the Isparta-Karakaya pumice waste was investigated. Once the index properties were determined, standard compaction, constant volume swelling, unconfined compression, shrinkage, and liquid and plastic limit tests were conducted to examine the effects of pumice on the compaction properties and the unconfined compressive strength of compacted clay. The Isparta-Karakaya pumice is composed of silica and aluminum. The natural unit weight of the pumice is extremely low. However, its permeability coefficient is high. The durability value against freezing of the pumice is 26. The Los Angeles abrasion and CBR values prove the limit values of Turkish Standards.

The shrinkage, liquid limit, and plastic limit values decreased with increasing pumice ratio because pumice is a non-plastic material. Standard compaction tests were performed by adding varying amounts of pumice to clay. With increasing pumice ratio in the mixture, it was observed that the maximum unit weight increased, whereas the optimum water content decreased. Moreover, the swelling pressure decreased with increasing amounts of pumice in the mixture. The unconfined compressive strength increased with pumice additive ratios of up to 30%. In contrast, the unconfined compressive strength decreased with pumice additive ratios of 40%–50%. The CBR degree increased as the pumice ratio was increased up to 40%, but it decreased as the pumice ratio reached 50%. Therefore, the ideal amount of the Isparta-Karakaya pumice additive to use for clayey subgrades is 30% by weight.

The use of pumice for stabilization of clayey subgrades has economic, environmental, and engineering benefits. Locally available pumice has little commercial value. Because it is readily available, the transportation costs involved in using this resource is minimal.