1.Introduction

Organic soil has several engineering problems due to its low shear strength, high compressibility, swelling, and shrinkage characteristics. The construction process presents a challenge for geotechnical engineers to address these issues. Today, Portland cement is a widely utilized material for soil stabilization. Still, cement manufacturing generates environmental impacts because it involves the significant release of carbon dioxide (CO2) emissions. Organic matter in the soil can reduce the reaction between cement and soil particles [1]. Replacing the natural soil layer with suitable soil from another site represents another method of organic soil stabilization [2]. As a result of this process, sometimes the project's cost may rise, particularly if the proposed location for the replacement soil is relatively far away. This has prompted researchers to explore alternative solutions. One of these methods involves soil stabilization with geopolymers. According to ASTM D2974, soil is classified as organic soil if the organic content exceeds 20 % [3].

Previous studies have investigated the engineering properties of organic soil, showing its high compressibility and low shear strength. Khairina et al. demonstrated that adding 10 % gypsum to organic soil with concrete waste increases the unconfined compressive strength (UCS) and reduces compressibility [4]. Adel et al. demonstrated that adding an amount of lime to organic soil, including both decomposed and non-decomposed samples, significantly affects their shear strength. The results showed that the cohesiveness values (c) increased by 5 % in non-decomposed samples and 7 % in decomposed samples with higher organic content [5]. Nath et al. evaluated the behavior of organic soil stabilized with fly ash as a function of strength. They found an increase in UCS with increasing fly ash amount due to the pozzolanic reaction of fly ash [6]. Ibrahim et al. demonstrated that increasing crushed waste concrete (CWC) to 50 % can increase organic clay's UCS values by 25 % [7]. Mohd et al. showed that adding lime and salt to organic soil increased the cohesion value from 7.1 to 23.6 kPa [8]. Some studies found that adding fly ash to organic soil increased the UCS and resilient modulus [9]. Moayedi et al. illustrated that adding 5 mol/L of sodium silicate (SS) to stabilized organic soil can increase the UCS values [10]. Habbi studied the organic soil's compressibility. The results demonstrated that an increase in organic content for all samples increases the compressibility characteristics (compression index, rebound index, coefficient of consolidation, and coefficient of secondary compression). The use of cement dust and fly ash has improved organic soil's compressibility [11]. Ali and Tatt studied that adding additives such as cement, lime, and fly ash to a mixture of organic soil improved its strength, particularly after the 7-day curing period [12]. Abbas et al. studied that additive of cement dust and fly ash with percentages (3, 6, and 9 %) by weight of organic soil improved compressibility characteristics [13]. Boobathiraja et al. studied that adding cement and lime in the range of 10–30 % as a percentage of dry soil mass caused an increase in the UCS of organic soil [14]. Rafizul et al. investigated how organic soils behave when mixed with cement, bentonite, and lime with different ratios of (5, 10, 15, 20, and 25 %) of the dry weight of organic soil. They found that MDU increased while OMC decreased as the admixture percentage increased [15]. Talib and Noriyuki developed another alternative binder using sugarcane bagasse ash (SCBA) for organic soil stabilization. Research demonstrated that combining SCBA with Portland cement enhanced the amount of UCS-stabilized organic soil [16].

As for improving the consolidation of organic soil. Thiyyakkandi and Annex showed that the consolidation coefficient decreased with increased organic matter [17]. Wardwell and Nelson showed the increase of organic content accompanied by a rise in the coefficient of secondary compression (Cα), causing an increase in the void ratio of soil [18]. Mustapa et al. showed that adding gypsum content with known percentages led to an improvement in the strength of organic soil. They found that any increased addition in gypsum content led to increased UCS values while the compressibility was reduced simultaneously [19].

The above literature review shows no application for organic soil stabilization using RGP-based geopolymers. The main aim of this study is to assess the effectiveness of using recycled glass powder-based geopolymer in improving organic soil engineering properties by conducting UCS and durability tests. Also, this study investigated the effect of RGP contents on MDU and optimum moisture content (OMC), as well as the effect of an alkaline activator solution on UCS and durability.

2.Methods and Materials

2.1.Soil

It was collected from the agricultural area in Anbar Governorate, west of Iraq. Soil samples were collected from a depth (0.3) m. According to the Unified Soil Classification System (USCS), the soil was classified as low-plasticity organic silty clay (OL). Tables 1 and 2 show the physical properties and chemical composition properties of soil.

Table 1. The physical and classification properties of soil.

Property

Soil

Specification

Liquid Limit (LL) (%)

44

According to ASTM D4318

Plastic Limit (PL) (%)

27

According to ASTM D4318

Plasticity index (PI) (%)

17

According to ASTM D4318

Specific gravity (Gs)

2.45

According to ASTM D854

Organic content (OC) (%)

20.8

According to ASTM D2974

Passing sieve #200 (%)

86

According to ASTM D422

Unified Soil Classification System (USCzS)

OL

According to ASTM D2487

Optimum moisture content) (%)

20

According to ASTM D698

Maximum dry unit) (KN/m3)

15

According to ASTM D698

Table 2. The chemical composition of soil by XRF.

Chemical composition (%)

SiO2

Al2O3

CaO

Fe2O3

MgO

Na2O

K2O

SO3

Mn

Organic soil

39

20.91

20.4

5.74

9.28

1.24

1.31

1.10

0.08

2.2.Recycled Glass Powder

The waste glass powder was collected from a waste landfill in Ramadi city. The glass was crushed using the traditional (metal mortar and pestle) tools. Then, it was sieved through sieve #200, and the particles finer than 75 μm were used as the base of the geopolymer. The glass powder specific gravity was determined to be 2.52, according to ASTM D854. The XRF test of RGP sample shows that its generally composed of SiO2, Al2O3, CaO, Fe2O3, MgO, etc. and it is generally composed of a high percentage of silica (SiO2), which is about 72.1 %. The XRF is conducted at the laboratories of Ministry of Industry and Minerals of Iraq. Table 3. Shows the chemical composition of RGP.

Table 3. The chemical composition of RGP by XRF.

Chemical composition (%)

SiO2

Al2O3

CaO

Fe2O3

MgO

Na2O

K2O

SO3

Mn

Glass powder

72.1

1.30

6.06

0.30

4.60

15.20

0

0.30

0

2.3.Alkaline solution

This study used sodium hydroxide (SH) (NaOH) and sodium silicate (Na2SiO3) as alkaline activators with a constant concentration of 4M and their properties are listed in Tables 4 and 5, respectively.

Table 4. The properties of (NaOH).

Chemical formula

Mass (g/mol)

Purity (%)

Density (g/cm3)

pH

NaOH

40

98.4

2.13 @ 20 °C

14

Table 5. The composition and properties of Na2SiO3.

Na2O (%)

SiO2 (%)

Viscosity (N.s/m2)

Specific gravity

pH

13.7

33

800

1.534

12

2.4.Geopolymer

It is an alternative binder instead of cement, made from reacted silica and alumina with an alkaline solution. It is used in many engineering applications, such as soil stabilization for instance. Geopolymer was developed as an eco-friendly alternative to Portland cement, as it is produced with less consumption of energy and reduces the emission of CO2. Geopolymers have many properties, such as flexibility, corrosion resistance, and being more resistant to environmental and climatic conditions, making them important materials in a variety of engineering applications.

Many laboratory tests were achieved on untreated organic soil. On the other hand, the testing was conducted on organic soil mixed with 5, 10, 15, 20, and 25 % RGP to study the impact of RGP-based geopolymer on the UCS of the stabilized soil. Many specimens of natural soil mixed with 5, 10, 15, 20, and 25 % RGP were performed according to ASTM D698 by using the standard proctor test to obtain the MDU and OMC. To find the value of UCS for each mixture with different ratios of RGP, specimens were prepared at similar MDU and OMC values found from the standard proctor test. Also, a small plastic spilt mold with a hammer was used to achieve the processing of the specimen's compaction. As shown in Fig. 1, the plastic mold dimensions are 50 mm in inner diameter and 100 mm in height, while the hammer's weight is 2.5 kg, falling from 30.5 cm in height. Compaction processing was conducted in 3 layers, and every layer was compressed with 8 blows. Where layers and blow numbers were obtained to get a similar compaction effort (596 kN.m/m3) for the proctor test. At first, the different ratios of RGP+ soil (5, 10, 15, 20, and 25 %) by dry mass of soil were mixed to prepare testing specimens. The solution of the alkaline activator was prepared with a constant concentration of NaOH (4M), and after this, it was added to the mixture of soil and RGP. The specimens of untreated soil were tested directly, and other treated specimens were cured at 65 °C for 48 hours and then kept at room temperature of 25 °C for 7, 14, and 28 days. For each mixture and curing period, three specimens were prepared [20, 21].

For the durability test, alkaline activator solution in three different ratios of solution SH:SS represented by 100:0, 72:28, and 50:50 were used to prepare soil specimens. All these specimens were exposed to wetting and drying cycles; similarly, treated soil specimens were exposed to wetting, drying, and brushing cycles. Each cycle started with submerging the specimens for 5 hours in the distilled water, 42 hours in the dried oven (71±2 °C), and brushing with a wire scratch brush. At the end of each cycle, the specimens are weighed and measured in volume. As expected, all specimens lost weight with time. After completing 12 cycles, the specimen was put in the oven (110 °C) to dry it until the mass became constant, and the UCS was conducted on all specimens.

Figure 1. Plastic spilt mold with dimension (50×100 mm).

3.Results and Discussion

Compaction Tests

The compaction test showed a relationship between dry density and moisture content for unstabilized (natural soil) and stabilized soil with different percentages (5, 10, 15, 20, and 25 %) of RGP, as shown in Fig. 2. The results show that as the RGP content ratio increases, MDU values increase and OMC values decrease. This may be attributed to the fact that finer particles of RGP fill the voids existing between soil particles [22]. This is similar to what was observed by the previous studies [23–25].

(a)

       (b)                                                                     (c)

Figure 2. a) The compaction curves of RGP-soil mixture, b) MDU versus RGP content, c) OMC versus RGP content.

Unconfined Compressive Strength

4.2.1. The impact of RGP content and curing period

The UCS values of soil with geopolymer of many percentages of RGP content (5, 10, 15, 20, and, 25 %) and curing time 7, 14, and 28 days with fixed alkaline solution SH:SS = 100:0, molarity 4M at a curing temperature 65 °C are shown in Fig. 3. As expected, the values of UCS for all soil specimens stabilized with RGP geopolymer were more than that of the untreated soil for all three curing times. Also, this figure shows that the value of UCS increases with increasing RGP content up to 20 %, where this increase represented by 29, 38, and 46 times compared to untreated soil for 7, 14, and 28 days respectively.

Figure 3. The values of UCS versus RGP content for different curing times 7,14 and 28 days at curing temperature 65 °C.

The maximum value of UCS was obtained at 20 % RGP content due to the stronger bond between the soil particles and RGP geopolymer gel [26]. This is similar to what was observed in previous studies [27–29]. A slight decrease in UCS values above 20 % RGP content was observed. The additional amount of RGP may have adverse effects on the UCS. These effects might be caused by the substantial amount of RGP particles contacting the soil structure, which would reduce soil cohesiveness. RGP serves as a filler material rather than contributing to the geopolymer binder [29].

4.2.2. Stress-strain behavior of RGP geopolymer-treated soil

Fig. 4 illustrates the impact of RGP Geopolymer on the stress-strain behavior of treated soil for three curing times 7, 14, and 28 days. All treated specimens had a lower failure strain compared to untreated soil. Where the axial strain values of specimens treated with 20 % RGP decreased by 31.4, 31.5, and 39.37 % of curing times 7, 14, and 28 days. As seen in the behavior of treated soil closer to ductility with increasing RGP content, this may be returned to a delay in developing geopolymer bonds due to the low curing temperature. Similar studies demonstrated that treated soil with metakalin-based geopolymer developed UCS and ductility [27], and another explained that RGP-based geopolymer can improve soil ductility [29].

Figure 4. Axial strain (%) values versus RGP content for different curing times 7,14 and 28 days at curing temperature 65 °C.

Durability

According to ASTM D559, the durability test was conducted on all treated soil specimens prepared for this purpose. For the durability test, three different ratios of alkaline activator solution SH:SS represented by 100:0, 72:28, and 50:50 were used separately with 20 % RGP+80 % soil)to prepare two geopolymer specimens for each mixture. The durability test for all prepared specimens was tested by wetting and drying (w/d) cycles. After the end of 12 cycles, the loss in mass and change in volume were recorded and discussed, as shown below.

4.3.1. Volume change

All specimens stabilized with RGP geopolymer have approximately changes (±1 %) in volume and survive all 12 w/d cycles. Whereas untreated soil specimens are destroyed within the first w/d cycle. Fig. 5 illustrates the natural soil and treated soil specimens before and after the w/d cycles.

4.3.2. Mass loss (%)

The mass loss (%) for all treated soil specimens with the RGP geopolymer tested to the durability cycles is shown in Fig. 5. As shown in Fig. 6, it can be observed that a gradual increase in mass loss was observed until the fifth cycle for mixes treated with 20 % RGP+ (100:0) and 20 % RGP+ (50:50). After this, the mass loss approximately remained constant up to the twelfth cycle. Whereas a slight gradual increase after the fifth cycle for mix treated with 20 % RGP+ (72:28) up to the ninth cycle remained constant up to the twelfth cycle. The losses in mass for treated soil specimens for SH:SS = 100:0, 72:28, and 50:50 for 65 °C were 4.9, 6.65, and 8 %, respectively. In any case, if they were compared to untreated soil, the treated soil specimens behaved better durability in terms of mass loss. All the mass loss (%) for all treated soil specimens was limited according to American standards [25, 26].

(a)                                                                                      (b)

(c)                                                                                      (d)

Figure 5. Durability test (w/d cycles): a) natural soil during cycle No. 1, (b) natural soil after (10min) during cycle No. 1, (c) treated soil specimens during cycle No. 1, (d) treated soil specimens after cycle No. 12.

Figure 6. Mass loss (%) versus No. of w/d cycles at 65 °C.

4.3.3. Strength loss

The UCS of the stabilized soil specimens exposed to w/d cycles was tested after 12 cycles. Table 4 displays the UCS values before and after the durability test. This result indicates that a sudden reduction in strength values for all specimens stabilized with 20 % was 20, 28.5, and 35.1 %, respectively, for curing temperature 65 °C recorded at the end of 12 cycles. A similar result has been observed with [29, 30]. Additionally, several studies showed that the UCS values of the treated soil were negatively impacted by the durability test [31, 32]. In comparison to soil that had not been treated, the results were deemed to be highly significant.

Table 4. Shows the results of UCS before and after durability (w/d cycles) for different soil mixtures.

Mixture

UCS before durability (MPa)

UCS at the end of durability (MPa)

Soil+20% RGP+(SH:SS=100:0)

9.41

7.5

Soil+20% RGP+(SH:SS=72:28)

8.12

5.8

Soil+20% RGP+(SH:SS=50:50)

8.01

5.2

4.Conclusion

This paper aimed to study the behavior, shear strength, and durability of organic stabilized with RGP-based geopolymer. According to the results from the testing, the conclusions below can be found.

When the RGP was added to the natural soil (organic soil), the MDU values increased and the values of OMC decreased. The results obtained from UCS tests for all specimens treated with RGP geopolymer demonstrated a significant increase in the shear strength of stabilized organic soil when the RGP content increased by more than 20 %. The strain values at failure decreased by 31.4, 31.5, and 39.37 % of the curing times 7,14 and 28 days, respectively, less than that of untreated soil, indicating a substantial effect of the RGP geopolymer additive on this parameter. The durability of all treated soil specimens was tested with wetting and drying cycles; all specimens survived 12 cycles. The volume change was approximately equal (±1), and the mass loss was 4.9 %, 6.65 %, and 8 % for treated soil specimens for SH:SS = 100:0, 72:28,and 50:50 at 65 °C, respectively, and it was found to be within the required limits. There was a reduction in strength values for all specimens stabilized with 20 % RGP represented by 20, 28.5, and 35.1 %, respectively, for SH:SS = 100:0, 72:28, and 50:50 at 65 °C. The strength values obtained after the wetting and drying cycles prove that the RGP geopolymer is a promising alternative solution of chemical stabilizers and a potential stabilizer and can be accepted for organic soil stabilizing.