Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume Roads

Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume... applied sciences Article Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume Roads 1 1 , 2 , 3 1 , José Luis Díaz-López , Manuel Cabrera * , José Ramón Marcobal , Francisco Agrela * and Julia Rosales Construction Engineering Area, University of Córdoba, Campus of Rabanales, 14014 Córdoba, Spain; jl.diaz@uco.es (J.L.D.-L.); jrosales@uco.es (J.R.) Department of Pavements, Sacyr Construction, 28040 Madrid, Spain; jose.marcobal@upm.es Transport Engineering, Territory and Urban Planning Department, Campus Ciudad Universitaria, Universidad Politécnica de Madrid, 28040 Madrid, Spain * Correspondence: manuel.cabrera@uco.es (M.C.); fagrela@uco.es (F.A.) Abstract: The application of new materials for soil stabilisation is a growing field of study in recent years. In this work, the effect of two types of silica-based nanomaterials combined with binders (quicklime and cement) are studied to stabilise soils and form structural layers for rural and low volume roads. The physical and chemical properties of the materials have been determined, as well as the mechanical behaviour of the stabilised soil. Three hybrid stabilised soil sections have been designed using a multilayer elastic model, executed at full scale and measuring the evolution of their properties in the medium to short term. The results show that the application of silica-based nanomaterials and two types of binders on the tread layers provide high structural stability and good behaviour of the sections. Citation: Díaz-López, J.L.; Cabrera, Keywords: hybrid stabilisation; mechanical behaviour; real scale application; nanomaterials M.; Marcobal, J.R.; Agrela, F.; Rosales, J. Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume Roads. 1. Introduction Appl. Sci. 2021, 11, 9780. https:// doi.org/10.3390/app11219780 Over several decades, a large primary and secondary network of pavement roads and highways has been introduced by all developed countries to connect urban centres, Academic Editor: Angeles metropolis and industrial areas with each other. This connectivity, combined with marine Sanroman Braga and air modes of transport, has generated a great socioeconomic impact derived from the transport of goods and passengers [1]. Received: 9 September 2021 However, there are many rural areas, small cities and, especially, developing countries Accepted: 15 October 2021 where it is not always possible to build this great network of roads due to economic, Published: 20 October 2021 accessibility or constructability issues, as well as the fact that other modes of transport have a lower level of development and, therefore, cannot even be implemented. In this Publisher’s Note: MDPI stays neutral way, rural and low volume roads are a major opportunity to improve the connectivity and with regard to jurisdictional claims in socioeconomic development of these areas [2]. published maps and institutional affil- Rural roads have different functions according to the level of development of the iations. country where they are built. In developed countries, rural roads are usually designed to connect towns with low populations and agricultural and livestock areas with a low volume of vehicles. On the contrary, in developed countries, rural roads are designed to meet the socioeconomic needs of the rural population, connecting remote areas to basic Copyright: © 2021 by the authors. health services, education and markets [3]. Licensee MDPI, Basel, Switzerland. These types of road can be composed of a subgrade and a thin asphaltic layer or, This article is an open access article most commonly, by compaction of an unpaved unbound granular material [4] or even distributed under the terms and compaction of stabilised soils that are found in the location of the road [5]. conditions of the Creative Commons However, in most cases, these soils, especially clayey soils, present geotechnical Attribution (CC BY) license (https:// problems, such as a lack of bearing capacity, high plasticity or swelling potential, that creativecommons.org/licenses/by/ prevent their use. In order to improve the properties of soils and increase their range of 4.0/). Appl. Sci. 2021, 11, 9780. https://doi.org/10.3390/app11219780 https://www.mdpi.com/journal/applsci Appl. Sci. 2021, 11, 9780 2 of 20 use in civil engineering applications, soil stabilisation for application in road layers has become widespread in recent years [6,7]. Soil stabilisation is defined as the improvement of the shear strength, durability stiff- ness and reduction of the plasticity and swelling potential of soils achieved by mechanical means or the addition of stabilising products, such as a hydraulic binder, fly and rice husk ash, chemical stabilisation, recycled waste and by-products, etc. [8]. Among all the stabilising products, stabilisation with binders (commonly lime and cement) have been widely used by numerous authors in recent decades [9–12]. Never- theless, the production of these traditional materials generates a negative environmental impact due to the use of raw material resources and the high CO emissions involved in their production [13]. For the reason mentioned above, in recent years, several studies that analyse the possible stabilisation of soils with industrial by-products or recycled materials, such as biomass fly and bottom ash [14–17], phosphogypsum [18,19], steel slag [20,21] and magne- sia oxides [22,23], among many others, have been increased. In addition, in recent years, an alternative to conventional stabilising products and wastes and by-products has emerged: nanomaterials. Nanomaterials are particles with a typical size of between 1 and 100 nm, with a very high specific surface area, which implies a very high reactivity [24], achieving soil improvements with very low dosages. Nanomaterials commonly used in soil stabilisation are composed of simple oxides, such as SiO , TiO or CaCO , or carbon nanotubes [25,26]. 2 2 3 Although a large number of nanoparticles has been developed, the pozzolanic capacity of Nano-SiO2 and its reactivity with lime and cement to form calcium silicate hydrate (CSH) compounds have led most studies to focus on it [27–31]. Kulanthaivel et al. [27] studied the effect of using synthesised Nano-SiO by a sol– gel process together with cement to stabilise clay soil, concluding that a 7% addition of nano-SiO improves the unconfined compressive strength in a ratio of 5:24 and reduces the permeability in the range of 0.01976 cm/s–0.01198 cm/s of the soil, which is consistent with previous studies [28]. Ghasabkolaei et al. [30] and Bahmani et al. [31] studied soil stabilisation with cements and nanosilane with dosages lower than 1% by weight of Nano-SiO , observing an increase in the unconfined compressive strength in all the mixtures with nanoparticulate additives and a high formation rate of calcium silicate hydrate (CSH), which implies an improvement in the soil properties. The present study shows new structural solutions for rural and low-volume roads based on soil stabilised with organosilanes and small amounts of lime and cement. The design parameters of the new sections, which were obtained through laboratory tests, were tested in a real-scale application to verify the feasibility of using the developed solutions. In addition, a medium-to-short-term performance study of the road sections built was carried out to check their durability. 2. Research Purpose and Experimental Programme This work is a continuation of the one carried out by Rosales et al. [32] in which a conventional control section was made according to Spanish specifications [33], and two alternative experimental sections based on nanomaterials and quicklime were performed in order to reduce the total thickness of the treated layers. In the present work, three new trial sections are proposed with a hybrid stabilisation process in which nanomaterials and small amounts of quicklime and cement are used to reduce their treated thickness with respect to the control section by 40% and improve the mechanical properties and durability. To carry out the trial sections in a successful manner, the following phases were followed in the present work: Phase 1. Conceptualisation and predesign of alternative sections based on the study conducted by Rosales et al. [32]. Appl. Sci. 2021, 11, 9780 3 of 20 Phase 2. Physicochemical characterisation of the materials involved in the work— namely, expansive clayey soil, sandy soil, which is a rejection of the production of crushed gravel, lime, cement and nanomaterials. Phase 3. Laboratory study to obtain the design parameters of the sections. In this phase, the percentages of soil for the mixtures were defined, as well as the quantities of the binders and organosilanes. The California Bearing Ratio (CBR) index and unconfined compressive strength (UCS) were obtained. Phase 4. Trial section structural design using Everstress 5.0 software. From the CBR index and simple compressive strength, the elastic modulus of the layers composing the sections were determined. Once the elastic parameters of the materials had been defined, the different section options were evaluated by means of a multilayer elastic analysis in the software. Phase 5. Construction and survey of the experimental sections. In this phase, the road sections were built, and short- and long-term checks of the bearing capacity of the soil were performed by non-destructive tests, such as plate loading and a falling weight deflectometer (FWD). 3. Phase 1. Section Pre-Design According to the results obtained by Rosales et al. [32], three alternative sections composed of two layers were proposed: one of lime stabilised expansive clayed soil (CS) and another upper one of a mixture of CS and sandy soil (SS) stabilised with cement. In the first pre-design stage, the layer thicknesses and soil proportions were not Appl. Sci. 2021, 11, x FOR PEER REVIEW 4 of 21 defined; however, the sections and their construction process were defined in a qualitative manner, as shown in Figure 1. Figure 1. Manufacturing process scheme of the hybrid stabilised sections. (a) Step 1: Pre-compaction Figure 1. Manufacturing process scheme of the hybrid stabilised sections. (a) Step 1: Pre-compaction of CS. (b) Step 2: Spreading and mixing of SS and cement. (c) Step 3: Compaction of CS+SS. of CS. (b) Step 2: Spreading and mixing of SS and cement. (c) Step 3: Compaction of CS+SS. 4. Laboratory Tests 4.1. Phase 2: Materials and Physicochemical Characterisation In this section, the materials used during the research for the development of the tests and construction of the trial sections are shown. 4.1.1. Soils: Expansive Clayey Soil and Sandy Soil In this work, two types of soils have been analysed for the subsequent stabilisation and construction of the layers of the trial section: an expansive clayey soil (CS) and a quarry reject from the production of crushed gravel called sandy soil (SS). The clayey soil (CS) comes from the plot where the real-scale sections are built, in Villacarrillo (Jaén), Andalusia. The sandy soil comes as a reject from a crushed gravel production quarry located near Villacarrillo. The rejects are subjected to a sieving process to achieve a particle size within the parameters of Spanish regulations [34]. Table 1 shows the physicochemical properties of the two soils analysed, and Figure 2 shows the granulometric curves of the soils. Appl. Sci. 2021, 11, x FOR PEER REVIEW 5 of 21 Appl. Sci. 2021, 11, 9780 4 of 20 Table 1. Physical and chemical properties of CS and SS. Properties CS SS Standard Test Atterberg limits UNE-EN ISO 17892-12:2019 As shown in Figure 1, the process of construction of the hybrid stabilised sections was Liquid Limit (%) 53.0 16.1 carried out in three steps: Step 1. Spreading, mixing and pre-compaction of the quicklime (or quicklime and Plastic limit (%) 21.6 11.9 nanomaterial) stabilised the clayey soil layer on the original ground. The nanomaterials Plasticity index (%) 31.4 4.2 were added, according to the dosage shown in the following sections, into the mixing water. Grain size distribution UNE-EN ISO 17892-4:2019 The curing time of the layer was one day. Gravel (>4 mm) (%) 0.0 77.1 Step 2. After one day, the spreading and mixing of the sandy soil layer on top of the Sand (0.063–4 mm) (%) 5.4 17.4 compacted quicklime (or quicklime and nanomaterial) stabilised the clayey soil layer and Silt and clay (<0.063 mm) (%) 94.6 5.5 the spreading of the cement. Max Step imun d 3. Compaction ry density (kg/ of the layer m ) composed 1.55 of 2. a 28 per centage of UNE 10 compacted 350quicklime 1:1994 (or quicklime and nanomaterial) stabilised clayey soil plus sandy soil and cement. The result Organic matter content (%) 0.31 0.02 UNE 103204:2019 is called hybrid stabilised soil. Water-soluble sulphate (% SO3) 0.05 0.07 UNE-EN 1744-1:2010+A1:2013 The following sections analyse the physicochemical properties of the materials in- Main components XRF (%) volved, as well as the mechanical properties of the mixes to be defined to optimise both the P 0.08 0.03 thickness of the mixes and the percentages of the binders. Si 25.57 3.32 4. Laboratory Tests Ca 14.16 31.2 4.1. Phase 2: Materials and Physicochemical Characterisation Al 2.69 0.83 In this section,S the materials used during 0.03 the 0. resear 05 ch for the development of the tests and construction of the trial sections are shown. K 1.66 1.74 Mg 1.37 16.2 4.1.1. Soils: Expansive Clayey Soil and Sandy Soil Fe 2.42 1.06 In this work, two types of soils have been analysed for the subsequent stabilisation and construction of the layers of the trial section: an expansive clayey soil (CS) and a quarry As shown in Table 1, the clayey soil shows a very high plasticity, with a plasticity reject from the production of crushed gravel called sandy soil (SS). index greater than 30, a discontinuous granulometry in which 95% of the particles are fine The clayey soil (CS) comes from the plot where the real-scale sections are built, in clays and silts and a maximum particle size of 5 mm. Villacarrillo (Jaén), Andalusia. In contrast, the sandy soil has a continuous grain size, composed mainly of particles The sandy soil comes as a reject from a crushed gravel production quarry located near larger than 4 mm, with a maximum aggregate size of 32 mm and a percentage of fine Villacarrillo. The rejects are subjected to a sieving process to achieve a particle size within grains of less than 6%. The plasticity index of SS is 4.2, which indicates that it is not a very the parameters of Spanish regulations [34]. Table 1 shows the physicochemical properties pl of as the tic ma twoteria soilsl. analysed, and Figure 2 shows the granulometric curves of the soils. Figure 2. Particle size distribution of clayey soil and sandy soil. Figure 2. Particle size distribution of clayey soil and sandy soil. Both materials have low percentages of organic matter content and water-soluble sul- phates, which make them suitable for use in road layers in accordance with Spanish reg- ulations. Appl. Sci. 2021, 11, 9780 5 of 20 Table 1. Physical and chemical properties of CS and SS. Properties CS SS Standard Test Atterberg limits UNE-EN ISO 17892-12:2019 Liquid Limit (%) 53.0 16.1 Plastic limit (%) 21.6 11.9 Plasticity index (%) 31.4 4.2 Grain size distribution UNE-EN ISO 17892-4:2019 Gravel (>4 mm) (%) 0.0 77.1 Sand (0.063–4 mm) (%) 5.4 17.4 Silt and clay (<0.063 mm) (%) 94.6 5.5 Maximun dry density (kg/m ) 1.55 2.28 UNE 103501:1994 Organic matter content (%) 0.31 0.02 UNE 103204:2019 UNE-EN Water-soluble sulphate (% SO ) 0.05 0.07 1744-1:2010+A1:2013 Main components XRF (%) P 0.08 0.03 Si 25.57 3.32 Ca 14.16 31.2 Al 2.69 0.83 S 0.03 0.05 K 1.66 1.74 Mg 1.37 16.2 Fe 2.42 1.06 Appl. Sci. 2021, 11, x FOR PEER REVIEW As shown in Table 1, the clayey soil shows a very high plasticity, with a plasticity 6 of 21 index greater than 30, a discontinuous granulometry in which 95% of the particles are fine clays and silts and a maximum particle size of 5 mm. In contrast, the sandy soil has a continuous grain size, composed mainly of particles Regarding the composition of the soils, an analysis of the major compounds by X-ray larger than 4 mm, with a maximum aggregate size of 32 mm and a percentage of fine fluorescence (XRF) and mineralogical analysis by X-ray diffraction (XRD) was carried out, grains of less than 6%. The plasticity index of SS is 4.2, which indicates that it is not a very as shown in Table 1 and Figures 3 and 4. plastic material. The clayey soil presents a main composition of silicon, calcium and aluminium, a Both materials have low percentages of organic matter content and water-soluble sul- typical composition of clays, which is observed in a mineralogy composed of quartz, cal- phates, which make them suitable for use in road layers in accordance with Spanish regulations. cite and montmorrollite-type clay minerals. Regarding the composition of the soils, an analysis of the major compounds by X-ray The sandy soil has a main composition of calcium and magnesium, a typical compo- fluorescence (XRF) and mineralogical analysis by X-ray diffraction (XRD) was carried out, sition of dolomite minerals, as shown in the XRD of Figure 4. as shown in Table 1 and Figures 3 and 4. Figure 3. XRD of clayey soil. Figure 3. XRD of clayey soil. Figure 4. XRD of sandy soil. 4.1.2. Binders: Quicklime and Cement Binders are materials that react chemically with water, forming cementitious com- pounds that can bind and improve the properties of soils, among other functions. In the present work, two binders have been applied: commercial quicklime (QL)-type CL 90-Q and a commercial ordinary Portland cement (OPC)-type CEM II/B-L 32.5R. Table 2 shows the composition in the form of oxides of both binders. Table 2. Compositions of the binders. Composition (%)/Binder SiO2 Al2O3 Fe2O3 CaO MgO SO3 K2O CEM II/B-L 32.5R 16.20 3.83 2.75 60.41 0.84 2.64 0.64 CL 90-Q 2.37 1.47 0.21 57.26 13.78 0.72 0.07 Appl. Sci. 2021, 11, x FOR PEER REVIEW 6 of 21 Regarding the composition of the soils, an analysis of the major compounds by X-ray fluorescence (XRF) and mineralogical analysis by X-ray diffraction (XRD) was carried out, as shown in Table 1 and Figures 3 and 4. The clayey soil presents a main composition of silicon, calcium and aluminium, a typical composition of clays, which is observed in a mineralogy composed of quartz, cal- cite and montmorrollite-type clay minerals. The sandy soil has a main composition of calcium and magnesium, a typical compo- sition of dolomite minerals, as shown in the XRD of Figure 4. Appl. Sci. 2021, 11, 9780 6 of 20 Figure 3. XRD of clayey soil. Figure Figure 4. 4. XRD XRD of of sandy sandy soi soil. l. The clayey soil presents a main composition of silicon, calcium and aluminium, a 4.1.2. Binders: Quicklime and Cement typical composition of clays, which is observed in a mineralogy composed of quartz, calcite Binders are materials that react chemically with water, forming cementitious com- and montmorrollite-type clay minerals. pounds that can bind and improve the properties of soils, among other functions. The sandy soil has a main composition of calcium and magnesium, a typical composi- In the present work, two binders have been applied: commercial quicklime (QL)-type tion of dolomite minerals, as shown in the XRD of Figure 4. CL 90-Q and a commercial ordinary Portland cement (OPC)-type CEM II/B-L 32.5R. Table 2 shows the composition in the form of oxides of both binders. 4.1.2. Binders: Quicklime and Cement Binders are materials that react chemically with water, forming cementitious com- Table 2. Compositions of the binders. pounds that can bind and improve the properties of soils, among other functions. Com In the pospr ition esent (%work, )/Bindtwo er binde SiO rs 2 have Al2O been 3 Fe applied: 2O3 CaO commer MgO cial quicklime SO3 K (QL)- 2O type CL 90-Q and a commercial ordinary Portland cement (OPC)-type CEM II/B-L 32.5R. CEM II/B-L 32.5R 16.20 3.83 2.75 60.41 0.84 2.64 0.64 Table 2 shows the composition in the form of oxides of both binders. CL 90-Q 2.37 1.47 0.21 57.26 13.78 0.72 0.07 Table 2. Compositions of the binders. Composition (%)/Binder SiO Al O Fe O CaO MgO SO K O 2 2 3 2 3 3 2 CEM II/B-L 32.5R 16.20 3.83 2.75 60.41 0.84 2.64 0.64 CL 90-Q 2.37 1.47 0.21 57.26 13.78 0.72 0.07 4.1.3. Nanomaterials Three types of silica-based nanomaterials were used in this study, named N1, N2 and N3. N1 is a concentrated liquid solution of sodium silicate for soil stabilisation, together with binders such as lime or cement. N2 is an organosilane, and N3 is a silica-based acrylic copolymeric in aqueous solution form, intended to be used together in a 1:1 ratio. For the three nanomaterials, its chemical composition was obtained by an X-ray fluorescence (XRF) analysis and a thermogravimetric analysis (TGA). Table 3 shows the results of the XRF analysis, and Figure 5 shows the results of the TGA. As can be observed in Table 3, the composition of N1 was mainly silicon and sodium, elements that form the nano-sized sodium silicates that make up this material. N2 and N3 were composed only of silicon as the main element and traces of the other elements. This sodium is dissolved in the form of organosilanes in N2 and in the form of an acrylic copolymer in N3. The results of the TGA analysis are shown in Figure 5. In N1, a progressive weight loss was observed of up to 200 C, which was due to the loss of water in the nanosilane. This nanomaterial showed a high silica content of 32%. N2 showed a higher weight loss; the silica content at 400 C was 17%. N3 completely loses its mass at a temperature of 400 C. Appl. Sci. 2021, 11, x FOR PEER REVIEW 7 of 21 4.1.3. Nanomaterials Three types of silica-based nanomaterials were used in this study, named N1, N2 and N3. N1 is a concentrated liquid solution of sodium silicate for soil stabilisation, together Appl. Sci. 2021, 11, 9780 7 of 20 with binders such as lime or cement. N2 is an organosilane, and N3 is a silica-based acrylic copolymeric in aqueous solution form, intended to be used together in a 1:1 ratio. For the three nanomaterials, its chemical composition was obtained by an X-ray flu- The higher stability of silica in the form of sodium silicate compared to other solutions, orescence (XRF) analysis and a thermogravimetric analysis (TGA). Table 3 shows the re- such as organosilanes or acrylic copolymers were observed, organosilane compounds lost sults of the XRF analysis, and Figure 5 shows the results of the TGA. 83% of their mass and acrylic compounds did not retain their mass after the test. Table 3. Main composition XRF (%) of the nanomaterials. 4.2. Phase 3: Mechanical Behaviour of Soil Mixtures Main Composition XRF (%) N1 N2 N3 In this section, the requirements for the thickness layer designs are determined. The Si 24.200 40.600 38.400 Modified Proctor compaction test, California Bearing Ratio (CBR), unconfined compressive Ca 0.279 0.014 0.097 strength and shear test are performed and shown. Figure 6 shows photographs of the Al 0.180 0.086 0.531 experimental methods and materials applied in this work. S 0.020 0.014 0.051 Table 3. Main composition XRF (%) of the nanomaterials. K 0.026 - 0.053 Mg 0.150 - 0.093 Main Composition XRF (%) N1 N2 N3 Fe 0.057 0.050 0.488 Si 24.200 40.600 38.400 Na 12.100 - 0.207 Ca 0.279 0.014 0.097 Al 0.180 0.086 0.531 As can be observed in Table 3, the composition of N1 was mainly silicon and sodium, S 0.020 0.014 0.051 elements that form the n K ano-sized sodium silic 0.026 ates that make - up this material. N2 0.053 and N3 Mg 0.150 - 0.093 were composed only of silicon as the main element and traces of the other elements. This Fe 0.057 0.050 0.488 sodium is dissolved in the form of organosilanes in N2 and in the form of an acrylic co- Na 12.100 - 0.207 polymer in N3. Appl. Sci. 2021, 11, x FOR PEER REVIEW 8 of 21 4.2. Phase 3: Mechanical Behaviour of Soil Mixtures In this section, the requirements for the thickness layer designs are determined. The Modified Proctor compaction test, California Bearing Ratio (CBR), unconfined compres- sive strength and shear test are performed and shown. Figure 6 shows photographs of the Figure 5. TGA curves of the nanomaterials. Figure 5. experiment TGA al met curves of hods and the n mat anomaterials. erials applied in this work. The results of the TGA analysis are shown in Figure 5. In N1, a progressive weight loss was observed of up to 200 °C, which was due to the loss of water in the nanosilane. This nanomaterial showed a high silica content of 32%. N2 showed a higher weight loss; the silica content at 400 °C was 17%. N3 completely loses its mass at a temperature of 400 °C. The higher stability of silica in the form of sodium silicate compared to other solu- tions, such as organosilanes or acrylic copolymers were observed, organosilane com- pounds lost 83% of their mass and acrylic compounds did not retain their mass after the test. Figure Figure 6. 6. E Experimental xperimental meth methods ods and and mater materials ials photographs photographs. . (a) Soils an (a) Soils d binders. ( and binders. b) Modified (b) Modified Proctor test. (c) CBR index test. (d) Unconfined compressive strength test. Proctor test. (c) CBR index test. (d) Unconfined compressive strength test. 4.2.1. Mix Design Six mixes of soil with binders and/or nanomaterials were defined to design the three alternative innovative sections studied in this article. Table 4 shows the mixtures analysed in the laboratory, the dosage of each material and its designation. Table 4. Dosages of the mixtures. Materials (kg/m ) Designation CS SS Quicklime OPC N1 N2 N3 CS + 1.5%QL 1590.00 - 23.85 - - - - CS + 1.5%QL + N1 1590.00 - 23.85 - 0.910 - - CS + 1.5%QL + N2&N3 1590.00 - 23.85 - - 1.000 1.000 AM-1 660.00 1380.00 9.90 20.40 - - - AM-2 660.00 1380.00 9.90 20.40 0.910 - - AM-3 660.00 1380.00 9.90 20.40 - 1.000 1.000 The dosages of the mixtures shown in the table above are expressed in weigh (kg) per cubic metre of material. Each material was added emulating the real process of road construction: Quicklime was added to the dry mass of CS in all mixtures. OPC was added to the total dry mass of soil (CS and SS), and nanomaterials were added to the dry volume of CS. 3 3 Finally, AM-1, AM-2 and AM-3 were composed of 0.40 m of CS and 0.60 m of SS per one cubic metre of the total mix. The AM mixes were made with a special manufacturing process, which is explained below: Step 1. Forty percent in volume of CS was mixed with an addition of 1.5 w% of quick- lime (AM-1), 1.5 w% of quicklime plus N1 (AM-2) or 1.5 w% of quicklime and N2 plus N3 Appl. Sci. 2021, 11, 9780 8 of 20 4.2.1. Mix Design Six mixes of soil with binders and/or nanomaterials were defined to design the three alternative innovative sections studied in this article. Table 4 shows the mixtures analysed in the laboratory, the dosage of each material and its designation. Table 4. Dosages of the mixtures. Materials (kg/m ) Designation CS SS Quicklime OPC N1 N2 N3 CS + 1.5%QL 1590.00 - 23.85 - - - - CS + 1.5%QL + N1 1590.00 - 23.85 - 0.910 - - CS + 1.5%QL + N2&N3 1590.00 - 23.85 - - 1.000 1.000 AM-1 660.00 1380.00 9.90 20.40 - - - AM-2 660.00 1380.00 9.90 20.40 0.910 - - AM-3 660.00 1380.00 9.90 20.40 - 1.000 1.000 The dosages of the mixtures shown in the table above are expressed in weigh (kg) per cubic metre of material. Each material was added emulating the real process of road construction: Quicklime was added to the dry mass of CS in all mixtures. OPC was added to the total dry mass of soil (CS and SS), and nanomaterials were added to the dry volume of CS. 3 3 Finally, AM-1, AM-2 and AM-3 were composed of 0.40 m of CS and 0.60 m of SS per one cubic metre of the total mix. The AM mixes were made with a special manufacturing process, which is explained below: Step 1. Forty percent in volume of CS was mixed with an addition of 1.5 w% of quicklime (AM-1), 1.5 w% of quicklime plus N1 (AM-2) or 1.5 w% of quicklime and N2 plus N3 (AM-3). The mixture was compacted according to the Modified Proctor compaction test and stored in a wet chamber with a minimum moisture of 90%. Step 2. One day after compaction, the mixture was decompressed into particles with a size smaller than 25 mm. Step 3. The stabilised CS was mixed with 60 v% of SS and 1 w% of OPC compared to the total dry mass of soil. The mixture was compacted according to the Modified Proctor test and stored in a wet chamber (compressive strength and shear tests) or curing tank (CBR test) before the samples were tested. 4.2.2. Compaction Test: Modified Proctor The level of compaction is a key parameter in road layer construction. A proper compaction ensures the strength and durability of the road; conversely, if the road has not been adequately compacted, it becomes unstable, which can result in a possible differential settlement. Settlement in a road implies pavement deformation and cracking that, added to storm water infiltration, causes a serious impact on traffic and road safety [35]. Compaction is studied through the moisture–dry density relationship, which is deter- mined according to the Modified Proctor Test, UNE EN 103501: 1994 standard. The moisture–dry density relationship, which is shown in Figure 7, is a very useful tool for understanding compacted soil behaviour. The maximum dry density to the optimum moisture content is obtained, which allowed the samples made to achieve the optimal mechanical behaviour. Additionally, the curve shape indicates the sensitivity of the soil to the water addition. If the curve is flatter, the maximum dry density is less affected by moisture changes; however, if the curve is sharper, small changes in the moisture content greatly affect the maximum dry density [36], as can be observed in Figure 7. Appl. Sci. 2021, 11, x FOR PEER REVIEW 9 of 21 (AM-3). The mixture was compacted according to the Modified Proctor compaction test and stored in a wet chamber with a minimum moisture of 90%. Step 2. One day after compaction, the mixture was decompressed into particles with a size smaller than 25 mm. Step 3. The stabilised CS was mixed with 60 v% of SS and 1 w% of OPC compared to the total dry mass of soil. The mixture was compacted according to the Modified Proctor test and stored in a wet chamber (compressive strength and shear tests) or curing tank (CBR test) before the samples were tested. 4.2.2. Compaction Test: Modified Proctor The level of compaction is a key parameter in road layer construction. A proper com- paction ensures the strength and durability of the road; conversely, if the road has not been adequately compacted, it becomes unstable, which can result in a possible differen- tial settlement. Settlement in a road implies pavement deformation and cracking that, added to storm water infiltration, causes a serious impact on traffic and road safety [35]. Compaction is studied through the moisture–dry density relationship, which is de- termined according to the Modified Proctor Test, UNE EN 103501: 1994 standard. The moisture–dry density relationship, which is shown in Figure 7, is a very useful tool for understanding compacted soil behaviour. The maximum dry density to the opti- mum moisture content is obtained, which allowed the samples made to achieve the opti- mal mechanical behaviour. Additionally, the curve shape indicates the sensitivity of the soil to the water addition. If the curve is flatter, the maximum dry density is less affected Appl. Sci. 2021, 11, 9780 9 of 20 by moisture changes; however, if the curve is sharper, small changes in the moisture con- tent greatly affect the maximum dry density [36], as can be observed in Figure 7. Figure 7. Moisture–dry density curves. Figure 7. Moisture–dry density curves. Analysing Figure 7, SS shows the highest maximum dry density, 2.28 kg/m , and Analysing Figure 7, SS shows the highest maximum dry density, 2.28 kg/m , and minimum optimum moisture content, 6.2%, likely due to its physical properties, such as a minimum optimum moisture content, 6.2%, likely due to its physical properties, such as higher quantity of coarse particles and its higher density. However, CS stabilised with lime a higher quantity of coarse particles and its higher density. However, CS stabilised with and quicklime plus nanomaterials show values contrary to SS due to the clay gradation, lime and quicklime plus nanomaterials show values contrary to SS due to the clay grada- which is composed mainly of fine particles and its great absorption of water, which is a tion, which is composed mainly of fine particles and its great absorption of water, which typical characteristic of expansive soil. is a typical characteristic of expansive soil. As can be observed, the addition of nanomaterials in the amount used in this article (0.056 w% N1 and 0.12% N1&N2) shows a low effect on the compaction parameters, especially for the dry density, which is consistent with the results obtained in previous studies of soil stabilisation with nanoparticles [29]. However, Alireza et al. [37] showed a decrease in the maximum dry density and an increase in the optimum compaction humidity in a soil stabilised with 5% lime from nano-SiO additions greater than 1%. Finally, the AM-1, AM-2 and AM-3 mixtures have an intermediate behaviour among the materials that compose them. A slight increase in the maximum dry density due to the addition of nanomaterials was observed in the AM-2 and AM-3 mixtures, likely due to the interaction of the stabilised soils with nanomaterials and the OPC. 4.2.3. Design Parameters: California Bearing Ratio (CBR) and Compressive Strength The structural behaviour of the road layers is determined by the load bearing capacity. According to the type of road layer executed, this bearing capacity is measured according to the CBR (California Bearing ratio) or unconfined compressive strength. The CBR index measures the bearing capacity of soils and compacted aggregates used in the construction of road bases or subbases. The CBR index depends on the density and moisture conditions of the samples. In this study, the samples were compacted according to the optimum moisture obtained in the Modified Proctor Test to reach the maximum dry density and the highest possible bearing capacity. The CBR value is carried out in accordance with UNE 103-502. The unconfined compressive strength (UCS) was performed according to the NLT- 305/90 standard in specimens 177.8 mm high and 152.4 mm in diameter. The UCS measured the resistance in cohesive soil or cement treatment soil or granular materials. Like the CBR samples, these samples were manufactured under the optimal compaction conditions obtained in the Modified Proctor Test. Spanish specifications [34] establish that lime-stabilised soil must comply with the minimum CBR index; however, cement-treated soil must exceed a minimum value of Appl. Sci. 2021, 11, x FOR PEER REVIEW 10 of 21 As can be observed, the addition of nanomaterials in the amount used in this article (0.056 w% N1 and 0.12% N1&N2) shows a low effect on the compaction parameters, es- pecially for the dry density, which is consistent with the results obtained in previous stud- ies of soil stabilisation with nanoparticles [29]. However, Alireza et al. [37] showed a decrease in the maximum dry density and an increase in the optimum compaction humidity in a soil stabilised with 5% lime from nano- SiO2 additions greater than 1%. Finally, the AM-1, AM-2 and AM-3 mixtures have an intermediate behaviour among the materials that compose them. A slight increase in the maximum dry density due to the addition of nanomaterials was observed in the AM-2 and AM-3 mixtures, likely due to the interaction of the stabilised soils with nanomaterials and the OPC. 4.2.3. Design Parameters: California Bearing Ratio (CBR) and Compressive Strength The structural behaviour of the road layers is determined by the load bearing capac- ity. According to the type of road layer executed, this bearing capacity is measured ac- cording to the CBR (California Bearing ratio) or unconfined compressive strength. The CBR index measures the bearing capacity of soils and compacted aggregates used in the construction of road bases or subbases. The CBR index depends on the density and moisture conditions of the samples. In this study, the samples were compacted ac- cording to the optimum moisture obtained in the Modified Proctor Test to reach the max- imum dry density and the highest possible bearing capacity. The CBR value is carried out in accordance with UNE 103-502. The unconfined compressive strength (UCS) was performed according to the NLT- 305/90 standard in specimens 177.8 mm high and 152.4 mm in diameter. The UCS meas- ured the resistance in cohesive soil or cement treatment soil or granular materials. Like the CBR samples, these samples were manufactured under the optimal compaction con- Appl. Sci. 2021, 11, 9780 10 of 20 ditions obtained in the Modified Proctor Test. Spanish specifications [34] establish that lime-stabilised soil must comply with the minimum CBR index; however, cement-treated soil must exceed a minimum value of un- confined compressive strength. The results obtained in the laboratory test are shown in unconfined compressive strength. The results obtained in the laboratory test are shown in Figure 8. Figure 8. Figure Figure 8. 8. CBR i CBR index ndex and com and compr pressive essive streng strength. th. Analysing the results shown in Figure 8, the use of different nanosilica improve the Analysing the results shown in Figure 8, the use of different nanosilica improve the bearing capacity of soils. bearing capacity of soils. The addition of 0.056 w% of N1 and 0.112 w% of N2&N3 increase the CBR index by The addition of 0.056 w% of N1 and 0.112 w% of N2&N3 increase the CBR index by 31.5% and 33.2%, respectively. 31.5% and 33.2%, respectively. Similar increases in the CBR index have been observed in previous studies [38] with similar dosages, probably due to the reactions of these nanomaterials with cement and the soil minerals themselves, leading to pozzolanic cementitious reactions. This behaviour has been observed in previous studies with higher increases in the bearing capacity or simple compressive strength, due to the higher addition percentages of the nanomaterials, between 1 and 7% [37,39]. The AM-1, AM-2 and AM-3 mixtures showed similar CBR index values, in the order of 100% of the CBR index with slightly higher values in the mixtures with nanomaterials, which shows an excellent bearing capacity. At the same time, unconfined compressive strength values of between 1.08 and 1.23 MPa were obtained, relatively high values for mixes with a total of 1 w% cement, normal values for stabilisation being a minimum of 3%, according to Spanish specifications [34]. 5. Phase 4: Trial Sections Structural Design In this section, the design of three trial sections based on the results obtained in laboratory tests are shown. Three alternative sections (AS) were designed to reduce the thickness of the control section analysed by Rosales et al. [32], while the mechanical and durability properties of the current section were maintained or improved. To guarantee the adequate structural performance of the alternative sections, a calcula- tion process was carried out using Everstress 5.0 software. The maximum load capacity of the control section, measured in the maximum number of standard axles of 13-tonne-heavy vehicles, was determined and compared with the alternative sections, which must present a value equal to or greater than the control section. To determine the maximum number of equivalent single axle loads of 13 tonnes, the following methodology was followed: (1) Determination of the elastic modulus (E) of the layers from the CBR index or com- pressive strength according to the type of material and its mechanical behaviour. (2) Calculation of the vertical deformations in the subgrade using Everstress 5.0 software. Appl. Sci. 2021, 11, 9780 11 of 20 (3) Calculation of the number of equivalent axles according to the fatigue law described by the Spanish specifications [40] according to Expression (1). 0.28 # = 2.16E 2  N (1) # = unit vertical deformation in the subgrade. N = number of equivalent axles of 13 tonnes. In the first place, the calculation of materials with a reduced CBR, such as soil from the construction site or stabilised or granular materials that do not present significant compressive strength, was accomplished by the Transport and Road Research Laboratory method [41], which is in accordance with the study of different bibliographies that indicate that the application of this method is appropriate for materials with CBR of less than 10% and without unconfined compressive strength. The formula of Powell et al. [41] for stabilised materials and unbound granular mate- rials was applied according to Expression (2). 0.64 E(MPa) = 17.6 CBR (2) Furthermore, with soils in which the CBR results were increased, a compressive strength test was carried out beforehand. The elastic modulus of these soils that present CBR values greater than 20% and a compressive strength greater than 0.2 MPa was cal- culated by the Molenaar equation. This equation considers the unconfined compressive strength as the main modulus calculation parameter. The formula of Molenaar [42] for materials treated with binders was applied according to Expression (3). E (MPa) = 1435 [UCS]ˆ0.885 (3) Table 5 shows the elastic modulus of the analysed mixtures according to the expression shown above. Table 5. Elastic modulus, E (MPa), of the soils or mixtures. Design Parameter Elastic Modulus (MPa) by Soil or Mixture Calculation CBR (%) UCS (MPa) CG * 60.05 - 242 CS + 1.5%QL 12.3 - 88 CS + 1.5%QL + N1 17.95 - 112 CS + 1.5%QL + N2&N3 18.4 114 AM-1 - 1.08 1536 AM-2 - 1.26 1761 AM-3 - 1.13 1599 * Obtained from Rosales et al. [32]. Once the elastic modulus of the materials that make up each layer were determined, they were entered into the software, together with the thicknesses of each layer and with a stress of 800 kPa, in accordance with Spanish regulations. First, the conventional section was analysed, which presented a bearing capacity of approximately 75,000 equivalent axels. Subsequently, three series of alternative sections (AS): AS-1, AS-2 and AS-3 of 45 cm, 50 cm and 55 cm, maintaining the proportion of 40% CS and 60% SS of the design mixes, were analysed. Table 6 shows the results obtained for the vertical deformation of the subgrade and the number of equivalent axels for each solution. Appl. Sci. 2021, 11, 9780 12 of 20 Table 6. Results of the vertical deformations in the subgrade obtained in the multilayers analysis and the number of axles obtained by calculus. Vertical Deformations in the Number of Equivalent Axles Section Subgrade (" ) (10 ) of 13 Tonnes Conventional Solution 932.96 74,737.32 AS-1-45 cm 1138.72 36,678.91 AS-2-45 cm 1049.18 49,140.69 AS-3-45 cm 1095.34 42,136.58 AS-1-50 cm 880.26 91,986.36 AS-2-50 cm 810.96 123,284,21 AS-3-50 cm 850.40 104,052.13 AS-1-55 cm 750.50 162,585.39 AS-2-55 cm 645.06 279,193.29 AS-3-55 cm 678.72 232,814.61 As is observed in Table 6, the 45-cm section series structural capacity was insufficient and was therefore rejected. The 55-cm series exceeded by three times the required capacity of 75,000 equivalent axels and was therefore rejected in order not to oversize the section. Appl. Sci. 2021, 11, x FOR PEER REVIEW 13 of 21 Therefore, the thickness of the section was 50 cm, due to a higher structural capacity than the control solution and a reduction in the section thickness of 30 cm. 6. 6. Phase Phase 5:5: Construc Construction tion and and S Section ection In In Situ Situ TT eests sts 6.1. Trials Sections Execution 6.1. Trials Sections Execution After the design phase, the three trial sections of 50-cm total thicknesses were con- After the design phase, the three trial sections of 50-cm total thicknesses were con- structed, with a length of 100 m for alternative sections 1 and 2 (AS-1 and AS-2) and a structed, with a length of 100 m for alternative sections 1 and 2 (AS-1 and AS-2) and a length of 50 m for alternative section 3 (AS-3). length of 50 m for alternative section 3 (AS-3). The three trial sections were built near the newly built road in Villacarrillo, Jaén, The three trial sections were built near the newly built road in Villacarrillo, Jaén, Spain, [32] and the performance of the AS was compared with the control section, which Spain, [32] and the performance of the AS was compared with the control section, which was used alongside the layout of the road. Figure 9 shows the control section and the was used alongside the layout of the road. Figure 9 shows the control section and the location of the trial section. location of the trial section. Figure 9. General location of the trial sections. Figure 9. General location of the trial sections. The final solutions were as follows: - Alternative section 1 (AS-1). Thirty centimetres of CS (40%) stabilised with 1.5% lime, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out accord- ing to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% quicklime. - Alternative section 2 (AS-2). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.056% N1, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out according to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% lime and 0.056% N1. - Alternative section 3 (AS-3). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.12% N2&N3, and SS (60%) stabilised with 1% CEM-II on the total soil (CS+SS) made according to the process described in Figure 1 on a 20-cm layer of CS stabilised with 1.5% lime and 0.12% N2&N3. Figure 10 shows a scheme of the developed sections. Appl. Sci. 2021, 11, 9780 13 of 20 The final solutions were as follows: - Alternative section 1 (AS-1). Thirty centimetres of CS (40%) stabilised with 1.5% lime, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out according to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% quicklime. - Alternative section 2 (AS-2). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.056% N1, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out according to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% lime and 0.056% N1. - Alternative section 3 (AS-3). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.12% N2&N3, and SS (60%) stabilised with 1% CEM-II on the total soil (CS+SS) made according to the process described in Figure 1 on a 20-cm layer of CS stabilised Appl. Sci. 2021, 11, x FOR PEER REVIEW 14 of 21 with 1.5% lime and 0.12% N2&N3. Figure 10 shows a scheme of the developed sections. Figure 10. Trial sections scheme. Figure 10. Trial sections scheme. 6.2. Trial Sections Survey 6.2. Trial Sections Survey According to Spanish specifications [33], three categories of subgrades, according to According to Spanish specifications [33], three categories of subgrades, according to the equivalent elastic modulus, are stabilised: low quality, 60–120 MPa; medium quality, the equivalent elastic modulus, are stabilised: low quality, 60–120 MPa; medium quality, 120–300 MPa and high quality, greater than 300 MPa. 120–300 MPa and high quality, greater than 300 MPa. For rural and low-volume roads, a low-quality subgrade with a modulus between 60 For rural and low-volume roads, a low-quality subgrade with a modulus between 60 and 120 MPa is structurally valid; however, due to the control section being designed as and 120 MPa is structurally valid; however, due to the control section being designed as a medium-quality subgrade, this criterion was maintained for the test sections, setting a a medium-quality subgrade, this criterion was maintained for the test sections, setting a minimum of a 120-MPa equivalent modulus. minimum of a 120-MPa equivalent modulus. To determine and compare the equivalent modulus value, the following methods To determine and compare the equivalent modulus value, the following methods were analysed. were analysed. Analysis of the theoretical deflection produced by a 500-kPa applied on a 300-mm- Analysis of the theoretical deflection produced by a 500-kPa applied on a 300-mm- diameter plate in a multilayer elastic model, the method specified in Spanish regula- diameter plate in a multilayer elastic model, the method specified in Spanish regulations tions [40]. The theoretical equivalent modulus value per section was obtained. [40]. The theoretical equivalent modulus value per section was obtained. Analysis of the deflections measured in the falling weight deflectometer (FWD) test. Analysis of the deflections measured in the falling weight deflectometer (FWD) test. The average equivalent modulus value of the section was obtained. The average equivalent modulus value of the section was obtained. Analysis of the second load cycle in the plate bearing test. A point value of equivalent Analysis of the second load cycle in the plate bearing test. A point value of equivalent modulus per section was obtained. modulus per section was obtained. Table 7 shows the results obtained in the Everstress 5.0 program, analysing the mul- tilayer elastic model with a stress of 500 kPa applied on a 300-mm-diameter plate. Apply- ing Formula (4), the equivalent modulus of compressibility of each section, Ev, was deter- mined. (MPa) = 13.150/ (mm/100)[40] (4) Table 7. Theorical deflection and modulus of compressibility obtained in Everstress 5.0. Everstress Results AS-1 AS-2 AS-3 Theorical deflection (mm/100) 78.02 73.51 75.63 Theorical modulus of compressibility, Ev (MPa) 168.55 178.89 173.87 As can be observed in Table 7, the three alternative sections present a sufficient the- oretical equivalent modulus to be considered a medium-quality subgrade. Appl. Sci. 2021, 11, 9780 14 of 20 Table 7 shows the results obtained in the Everstress 5.0 program, analysing the mul- tilayer elastic model with a stress of 500 kPa applied on a 300-mm-diameter plate. Ap- plying Formula (4) [40], the equivalent modulus of compressibility of each section, E , was determined. E (MPa) = 13.150/d (mm/100) (4) v 0 Table 7. Theorical deflection and modulus of compressibility obtained in Everstress 5.0. Everstress Results AS-1 AS-2 AS-3 Theorical deflection (mm/100) 78.02 73.51 75.63 Theorical modulus of compressibility, E (MPa) 168.55 178.89 173.87 Appl. Sci. 2021, 11, x FOR PEER REVIEW 15 of 21 As can be observed in Table 7, the three alternative sections present a sufficient theoretical equivalent modulus to be considered a medium-quality subgrade. Once the theoretical equivalent modulus values for each section have been obtained, Once the theoretical equivalent modulus values for each section have been obtained, they are verified with the values obtained in the in situ tests, FDW and plate bearing test. they are verified with the values obtained in the in situ tests, FDW and plate bearing test. As Figure 11 shows, the climatology of the area shows two distinct annual seasons, As Figure 11 shows, the climatology of the area shows two distinct annual seasons, with 6 months with abundant rainfall and 6 dry months [43]. Therefore, two in situ test with 6 months with abundant rainfall and 6 dry months [43]. Therefore, two in situ test campaigns were carried out: the first in July 2020 in the dry season and the second in campaigns were carried out: the first in July 2020 in the dry season and the second in March 2021 in the wet season. March 2021 in the wet season. Figure 11. Annual rainfall distribution in Villacarrillo, Spain. Figure 11. Annual rainfall distribution in Villacarrillo, Spain. Once the data for both periods was obtained, the average annual value of equivalent Once the data for both periods was obtained, the average annual value of equivalent modulus was obtained for each of the methods. modulus was obtained for each of the methods. 6.2.1. Deflection Measurements by FWD 6.2.1. Deflection Measurements by FWD The deflection measurement enables characterising the structural capacity of the The deflection measurement enables characterising the structural capacity of the formed subgrade, as well as its layers along the road layout. A Dynatest HDW 8081 falling formed subgrade, as well as its layers along the road layout. A Dynatest HDW 8081 falling weight deflectometer (FWD) was used. weight deflectometer (FWD) was used. A pressure of 850 kPa was applied through a 300-mm-diameter plate. The surface A pressure of 850 kPa was applied through a 300-mm-diameter plate. The surface deformation due to the application of this load was measured by seven geophones located deformation due to the application of this load was measured by seven geophones located at 0, 300, 450, 600, 900, 1200 and 1500 mm from the application of the load. at 0, 300, 450, 600, 900, 1200 and 1500 mm from the application of the load. Surface deformation measurements were taken every 10 m in sections AS-1 and AS-2 Surface deformation measurements were taken every 10 m in sections AS-1 and AS- and every 5 m in section AS-3 in both lanes, obtaining the average per section. 2 and every 5 m in section AS-3 in both lanes, obtaining the average per section. Figures 12–14 show the mean deflection values measured in the dry and wet seasons Figures 12–14 show the mean deflection values measured in the dry and wet seasons for each section. for each section. Figure 12. Deflection measurement in AS-1. Appl. Sci. 2021, 11, x FOR PEER REVIEW 15 of 21 Once the theoretical equivalent modulus values for each section have been obtained, they are verified with the values obtained in the in situ tests, FDW and plate bearing test. As Figure 11 shows, the climatology of the area shows two distinct annual seasons, with 6 months with abundant rainfall and 6 dry months [43]. Therefore, two in situ test campaigns were carried out: the first in July 2020 in the dry season and the second in March 2021 in the wet season. Figure 11. Annual rainfall distribution in Villacarrillo, Spain. Once the data for both periods was obtained, the average annual value of equivalent modulus was obtained for each of the methods. 6.2.1. Deflection Measurements by FWD The deflection measurement enables characterising the structural capacity of the formed subgrade, as well as its layers along the road layout. A Dynatest HDW 8081 falling weight deflectometer (FWD) was used. A pressure of 850 kPa was applied through a 300-mm-diameter plate. The surface deformation due to the application of this load was measured by seven geophones located at 0, 300, 450, 600, 900, 1200 and 1500 mm from the application of the load. Surface deformation measurements were taken every 10 m in sections AS-1 and AS- 2 and every 5 m in section AS-3 in both lanes, obtaining the average per section. Appl. Sci. 2021, 11, 9780 15 of 20 Figures 12–14 show the mean deflection values measured in the dry and wet seasons for each section. Appl. Sci. 2021, 11, x FOR PEER REVIEW 16 of 21 Appl. Sci. 2021, 11, x FOR PEER REVIEW 16 of 21 Figure 12. Deflection measurement in AS-1. Figure 12. Deflection measurement in AS-1. Figure 13. Deflection measurement in AS-2. Figure 13. Deflection measurement in AS-2. Figure 13. Deflection measurement in AS-2. Figure 14. Deflection measurement in AS-3. Figure 14. Deflection measurement in AS-3. Figure 14. Deflection measurement in AS-3. Analysing Figures 12–14, two patterns of behaviour were observed: firstly, the addi- Analysing Analysing Figures Figures 1 12 2–1 –14 4, t , two wo pa patterns tterns of of be behaviour haviour we wer re e obs observed: erved: firfirstly stly, the a , thed ad- di- tion of nanomaterials, both sodium silicates and organosilanes, together with acrylic co- dition tion of na of nanomaterials, nomaterials, both sodi both sodium um silica silicates tes and organosilanes, to and organosilanes,ge together ther with with acryli acrylic c co- polymers, improved the bearing capacity of the soil, reducing the deflections measured copolymers, polymers, im impr proved t ovedh the e bear bearing ing ca capacity pacity of ofthe soil, reduc the soil, reducing ing the the deflections deflections measur measured ed during the test. during the test. during the test. Secondly, the sections with nanomaterials showed a drop in the bearing capacity due Secondly, the sections with nanomaterials showed a drop in the bearing capacity due to rainfall, in contrast to the AS-1 section, which remained practically constant after a to rainfall, in contrast to the AS-1 section, which remained practically constant after a year’s weather. year’s weather. However, the AS-1 section showed greater dispersion in the data, which could be However, the AS-1 section showed greater dispersion in the data, which could be attributed to problems during the execution of the work, presenting over-compacted areas attributed to problems during the execution of the work, presenting over-compacted areas that reduce the average deflection of the section. that reduce the average deflection of the section. Table 8 shows the average annual deflection of each section, as well as the weighted Table 8 shows the average annual deflection of each section, as well as the weighted deflection for a load of 500 kPa, which enables calculating the elastic modulus through deflection for a load of 500 kPa, which enables calculating the elastic modulus through Equation (4). Equation (4). Table 8. Modulus of the compressibility mean in each section. Table 8. Modulus of the compressibility mean in each section. Anual Average Deflection Weighted Deflection Anual Average Deflection Weighted Deflection Trial Section Ev (MPa) Trial Section Ev (MPa) (0.01 mm) (0.01 mm) (0.01 mm) (0.01 mm) AS-1 183 108 122 AS-1 183 108 122 AS-2 168 99 133 AS-2 168 99 133 AS-3 159 93 142 AS-3 159 93 142 Appl. Sci. 2021, 11, 9780 16 of 20 Secondly, the sections with nanomaterials showed a drop in the bearing capacity due to rainfall, in contrast to the AS-1 section, which remained practically constant after a year ’s weather. However, the AS-1 section showed greater dispersion in the data, which could be attributed to problems during the execution of the work, presenting over-compacted areas that reduce the average deflection of the section. Table 8 shows the average annual deflection of each section, as well as the weighted deflection for a load of 500 kPa, which enables calculating the elastic modulus through Equation (4). Table 8. Modulus of the compressibility mean in each section. Anual Average Weighted Deflection Trial Section E (MPa) Deflection (0.01 mm) (0.01 mm) AS-1 183 108 122 AS-2 168 99 133 AS-3 159 93 142 As is shown in Table 8, the average values of the sections show a reduction in the average annual deflections and an increase in the equivalent modulus of the sections made with nanomaterials. Comparing the deflection results obtained for the new hybrid stabilised sections, a bet- ter behaviour was observed than the alternative section analysed by Rosales et al. [32], and a slight drop was observed when compared to the 80-cm-thickness control section. However, a section reduction of approximately 40% and a similar mechanical behaviour confirmed the suitability of the proposed method of hybrid stabilised solutions for road layers. All the sections can be classified as in the medium category, although with lower values than those obtained theoretically, due to the irregularities of the terrain and construction peculiarities of the linear works. 6.2.2. Plate-Bearing Test The plate loading test is an in situ test used to measure the final bearing capacity of the subgrade built from the sections designed through its compressibility modulus. The plate bearing test was performed according to UNE 103808:2006, and it consists of measuring the settlement of a rigid circular plate resting on the ground, subjected to different loads in a staggered manner, called the load cycle. This circular-shaped plate has a surface area of 700 cm (diameter 298.5 mm), and the measurements enable determining the compressibility modulus in the first load cycle (E ) and in the second load cycle (E ). v 1 v 2 The plate bearing test enables the calculation of the punctual behaviour of the section in the place where the load was applied, so its result was not as representative as those obtained in FDW; however, it enables verifying the behaviour of the subgrade and analysing its evolution over time. The results of the plate loading test are shown in Figures 15–17. As is shown in Figure 15, all the sections presented a medium quality, which verified the adequate structural behaviour in relation to the volume of the loads the sections will support during their useful life. An improvement in the quality of subgrades AS-2 and AS-3 was observed due to the addition of nanomaterials, similar to that observed in the FDW results. Analysing Figure 16, a drop in the equivalent modulus was observed in all three sections due to the rainy period; however, this drop was greater in sections AS-2 and AS-3 than AS-1. This sharp drop may be due to a lack of bearing capacity in some of the layers that make up the section because of the rain or perhaps a specific problem associated with the uncertainty of the plate loading test. Appl. Sci. 2021, 11, x FOR PEER REVIEW 17 of 21 As is shown in Table 8, the average values of the sections show a reduction in the average annual deflections and an increase in the equivalent modulus of the sections made with nanomaterials. Comparing the deflection results obtained for the new hybrid stabilised sections, a better behaviour was observed than the alternative section analysed by Rosales et al. [32], and a slight drop was observed when compared to the 80-cm-thickness control section. However, a section reduction of approximately 40% and a similar mechanical behaviour confirmed the suitability of the proposed method of hybrid stabilised solutions for road layers. All the sections can be classified as in the medium category, although with lower values than those obtained theoretically, due to the irregularities of the terrain and con- struction peculiarities of the linear works. Appl. Sci. 2021, 11, 9780 17 of 20 6.2.2. Plate-Bearing Test The plate loading test is an in situ test used to measure the final bearing capacity of Finally, Figure 17 shows the average annual behaviour of the sections. Despite the large the subgrade built from the sections designed through its compressibility modulus. drop observed in the rainy weather for sections AS-2 and AS-3, an adequate mechanical The plate bearing test was performed according to UNE 103808:2006, and it consists behaviour was observed, being practically identical to that obtained in the theoretical of measuring the settlement of a rigid circular plate resting on the ground, subjected to analysis using the multilayer elastic model, which confirms the validity of the multilayer different loads in a staggered manner, called the load cycle. This circular-shaped plate has elastic method used and the validity of the sections constructed. a surface area of 700 cm (diameter 298.5 mm), and the measurements enable determining Comparing the results with those obtained by Rosales et al. [32], an increase in the the compressibility modulus in the first load cycle (Ev1) and in the second load cycle (Ev2). modulus of the compressibility of the current hybrid sections was observed. This increase The plate bearing test enables the calculation of the punctual behaviour of the section confirmed the adequate behaviour of the new solutions constructed with a reduction of the in the place where the load was applied, so its result was not as representative as those input materials. obtained in FDW; however, it enables verifying the behaviour of the subgrade and ana- Additionally, the evolution of the E /E ratio was shown, limited by Spanish v2 v1 lysing its evolution over time. specifications to 2.2, a limitation that was met by all sections in both the dry and wet seasons. The results of the plate loading test are shown in Figures 15–17. Appl. Sci. 2021, 11, x FOR PEER REVIEW 18 of 21 Figure Figure 15. 15. Plat Plate e bearing bearing test test rr eesults sults in the in thedr dry y season. season. Figure 16. Plate bearing test results in the wet season. Figure 16. Plate bearing test results in the wet season. Figure 17. Annual average plate bearing test results. As is shown in Figure 15, all the sections presented a medium quality, which verified the adequate structural behaviour in relation to the volume of the loads the sections will support during their useful life. An improvement in the quality of subgrades AS-2 and AS-3 was observed due to the addition of nanomaterials, similar to that observed in the FDW results. Analysing Figure 16, a drop in the equivalent modulus was observed in all three sec- tions due to the rainy period; however, this drop was greater in sections AS-2 and AS-3 than AS-1. This sharp drop may be due to a lack of bearing capacity in some of the layers that make up the section because of the rain or perhaps a specific problem associated with the uncertainty of the plate loading test. Finally, Figure 17 shows the average annual behaviour of the sections. Despite the large drop observed in the rainy weather for sections AS-2 and AS-3, an adequate me- chanical behaviour was observed, being practically identical to that obtained in the theo- retical analysis using the multilayer elastic model, which confirms the validity of the mul- tilayer elastic method used and the validity of the sections constructed. Appl. Sci. 2021, 11, x FOR PEER REVIEW 18 of 21 Appl. Sci. 2021, 11, 9780 18 of 20 Figure 16. Plate bearing test results in the wet season. Figure 17. Annual average plate bearing test results. Figure 17. Annual average plate bearing test results. 7. Conclusions As is shown in Figure 15, all the sections presented a medium quality, which verified In this work, the construction of trial sections based on hybrid stabilised soils with the adequate structural behaviour in relation to the volume of the loads the sections will small amounts of lime, cement and nanomaterials was studied. For this purpose, an in- support during their useful life. depth laboratory study was carried out to determine the physicochemical properties of An improvement in the quality of subgrades AS-2 and AS-3 was observed due to the the materials and the mechanical behaviours of the mixtures of stabilised soil. A structural addition of nanomaterials, similar to that observed in the FDW results. design of the trial sections was carried out through a multilayer elastic model. Finally, the Analysing Figure 16, a drop in the equivalent modulus was observed in all three sec- sections were built in a real-scale application, and their short and medium-term in situ tions due to the rainy period; however, this drop was greater in sections AS-2 and AS-3 properties were monitored. than AS-1. The following conclusions have been drawn from this study: This sharp drop may be due to a lack of bearing capacity in some of the layers that The stabilisation of soils with nanosilica slightly increases the maximum dry density of make up the section because of the rain or perhaps a specific problem associated with the the samples and slightly reduces their optimum water content. A reduction in the optimum uncertainty of the plate loading test. water content implies a reduction in the consumption of natural resources (water, fuel, less Finally, Figure 17 shows the average annual behaviour of the sections. Despite the CO emissions, etc.) to achieve the same or higher degree of compaction. large drop observed in the rainy weather for sections AS-2 and AS-3, an adequate me- The addition of nanosilica improves the bearing capacity of the stabilised soils as mea- chanical behaviour was observed, being practically identical to that obtained in the theo- sured by the CBR index. The nanomaterial N1, composed of sodium silicate, shows a higher retical analysis using the multilayer elastic model, which confirms the validity of the mul- reactivity in combination with cement, increasing the unconfined compressive strength. tilayer elastic method used and the validity of the sections constructed. The proposed hybrid stabilisation solution reduces the thickness of the control section by 37.5% and increases the bearing capacity measured in the number of equivalent axles by up to 25%. The sections built with N1 and N2&N3 show an improvement in the annual mean equivalent modulus of compressibility, with these sections showing a greater drop in bearing capacity during the rainy season. The results support the application of the developed alternative solutions with a hybrid stabilisation process due to their increased bearing capacity for rural and low- volume roads. However, due to the typical climatic conditions in Southern Spain, there is no evidence of effectiveness in the other conditions. In countries with high humidity and large temperature differences, the solution should be verified. As a general conclusion, the use of nanomaterials in percentages between 0.06 and 0.12% improves the mechanical behaviour of stabilised soils and allows a reduction in the thickness of the road layer, improving its general structural capacity, although there is a drop in capacity during rainy periods, which will be the subject of future studies. Author Contributions: Conceptualization, F.A. and J.R.M.; methodology, J.R. and F.A.; investigation, J.L.D.-L. and M.C.; resources, F.A. and J.R.M.; data curation, J.L.D.-L. and J.R.; writing—original draft preparation, J.L.D.-L. and M.C.; writing—review and editing, M.C., J.R. and J.L.D.-L. and supervision, F.A. All authors have read and agreed to the published version of the manuscript. Appl. Sci. 2021, 11, 9780 19 of 20 Funding: This work was financed by the project “SLAKED-LIME REDUCTION ON EXPANSIVE SOILS BY MEANS OF NANOMATERIALS AND THE REUTILIZATION OF WASTED MATERIAL AND STABILIZING BY-PRODUCTS—ECARYSE. Ref. RTC-2017-62025”, granted by the call for “FEDER/Ministerio de Educacion Ciencia y Universidades-Agencia Estatal de Investigación”. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Acknowledgments: Call for Sub-modality 2.2. Pre-doctoral contracts UCO of the University of Cordoba’s Research Plan 2020. Conflicts of Interest: The authors declare no conflict of interest. References 1. Ali, Y.; Socci, C.; Pretaroli, R.; Severini, F. Economic and environmental impact of transport sector on Europe economy. Asia-Pac. J. Reg. Sci. 2018, 2, 361–397. [CrossRef] 2. Asomani-Boateng, R.; Fricano, R.J.; Adarkwa, F. Assessing the socio-economic impacts of rural road improvements in Ghana: A case study of transport sector program support (II). Case Stud. Transp. Policy 2015, 3, 355–366. [CrossRef] 3. Chamorro, A.; Tighe, S. Development and Application of a Sustainable Management System for Unpaved Rural Road Networks. Transp. Res. Rec. 2019, 2673, 891–901. [CrossRef] 4. Jiménez, J.; Ayuso, J.; Galvín, A.; López, M.; Agrela, F. Use of mixed recycled aggregates with a low embodied energy from non-selected CDW in unpaved rural roads. Constr. Build. Mater. 2012, 34, 34–43. [CrossRef] 5. Sirivitmaitrie, C.; Puppala, A.J.; Saride, S.; Hoyos, L. Combined lime–cement stabilization for longer life of low-volume roads. Transp. Res. Rec. 2011, 2204, 140–147. [CrossRef] 6. Lim, S.; Wijeyesekera, D.; Lim, A.; Bakar, I. Critical review of innovative soil road stabilization techniques. Int. J. Eng. Adv. Technol. 2014, 3, 2249–8958. 7. Firoozi, A.A.; Olgun, C.G.; Firoozi, A.A.; Baghini, M.S. Fundamentals of soil stabilization. Int. J. Geo-Eng. 2017, 8, 26. [CrossRef] 8. Afrin, H. A review on different types soil stabilization techniques. Int. J. Transp. Eng. Technol. 2017, 3, 19–24. [CrossRef] 9. Horpibulsuk, S.; Rachan, R.; Chinkulkijniwat, A.; Raksachon, Y.; Suddeepong, A. Analysis of strength development in cement- stabilized silty clay from microstructural considerations. Constr. Build. Mater. 2010, 24, 2011–2021. [CrossRef] 10. Khemissa, M.; Mahamedi, A. Cement and lime mixture stabilization of an expansive overconsolidated clay. Appl. Clay Sci. 2014, 95, 104–110. [CrossRef] 11. Olinic, T.; Olinic, E. The effect of quicklime stabilization on soil properties. Agric. Agric. Sci. Procedia 2016, 10, 444–451. [CrossRef] 12. Prusinski, J.R.; Bhattacharja, S. Effectiveness of Portland cement and lime in stabilizing clay soils. Transp. Res. Rec. 1999, 1652, 215–227. [CrossRef] 13. Costa, C. Hydraulic binders. In Materials for Construction and Civil Engineering; Springer: Cham, Switzerland, 2015; pp. 1–52. 14. Barišic, ´ I.; Netinger Grubeša, I.; Dokšanovic, ´ T.; Markovic, ´ B. Feasibility of agricultural biomass fly ash usage for soil stabilisation of road works. Materials 2019, 12, 1375. [CrossRef] [PubMed] 15. Cabrera, M.; Rosales, J.; Ayuso, J.; Estaire, J.; Agrela, F. Feasibility of using olive biomass bottom ash in the sub-bases of roads and rural paths. Constr. Build. Mater. 2018, 181, 266–275. [CrossRef] 16. Galvín, A.P.; López-Uceda, A.; Cabrera, M.; Rosales, J.; Ayuso, J. Stabilization of expansive soils with biomass bottom ashes for an eco-efficient construction. Environ. Sci. Pollut. Res. 2021, 28, 24441–24454. [CrossRef] 17. Vichan, S.; Rachan, R. Chemical stabilization of soft Bangkok clay using the blend of calcium carbide residue and biomass ash. Soils Found. 2013, 53, 272–281. [CrossRef] 18. Amrani, M.; Taha, Y.; Kchikach, A.; Benzaazoua, M.; Hakkou, R. Phosphogypsum recycling: New horizons for a more sustainable road material application. J. Build. Eng. 2020, 30, 101267. [CrossRef] 19. Zeng, L.-L.; Bian, X.; Zhao, L.; Wang, Y.-J.; Hong, Z.-S. Effect of phosphogypsum on physiochemical and mechanical behaviour of cement stabilized dredged soil from Fuzhou, China. Geomech. Energy Environ. 2021, 25, 100195. [CrossRef] 20. Manso, J.M.; Ortega-López, V.; Polanco, J.A.; Setién, J. The use of ladle furnace slag in soil stabilization. Constr. Build. Mater. 2013, 40, 126–134. [CrossRef] 21. Thomas, A.; Tripathi, R.; Yadu, L. A laboratory investigation of soil stabilization using enzyme and alkali-activated ground granulated blast-furnace slag. Arab. J. Sci. Eng. 2018, 43, 5193–5202. [CrossRef] 22. Wang, D.; Du, Y.; Xiao, J. Shear properties of stabilized loess using novel reactive magnesia-bearing binders. J. Mater. Civ. Eng. 2019, 31, 04019039. [CrossRef] 23. Gu, K.; Jin, F.; Al-Tabbaa, A.; Shi, B.; Liu, C.; Gao, L. Incorporation of reactive magnesia and quicklime in sustainable binders for soil stabilisation. Eng. Geol. 2015, 195, 53–62. [CrossRef] 24. Krishnan, J.; Shukla, S. The behaviour of soil stabilised with nanoparticles: An extensive review of the present status and its applications. Arab. J. Geosci. 2019, 12, 436. [CrossRef] 25. Ghasabkolaei, N.; Choobbasti, A.J.; Roshan, N.; Ghasemi, S.E. Geotechnical properties of the soils modified with nanomaterials: A comprehensive review. Arch. Civ. Mech. Eng. 2017, 17, 639–650. [CrossRef] Appl. Sci. 2021, 11, 9780 20 of 20 26. Kawashima, S.; Hou, P.; Corr, D.J.; Shah, S.P. Modification of cement-based materials with nanoparticles. Cem. Concr. Compos. 2013, 36, 8–15. [CrossRef] 27. Kulanthaivel, P.; Soundara, B.; Velmurugan, S.; Naveenraj, V. Experimental investigation on stabilization of clay soil using nano-materials and white cement. Mater. Today Proc. 2021, 45, 507–511. [CrossRef] 28. Choobbasti, A.J.; Kutanaei, S.S. Microstructure characteristics of cement-stabilized sandy soil using nanosilica. J. Rock Mech. Geotech. Eng. 2017, 9, 981–988. [CrossRef] 29. Meeravali, K.; Ruben, N.; Rangaswamy, K. Stabilization of soft-clay using nanomaterial: Terrasil. Mater. Today Proc. 2020, 27, 1030–1037. [CrossRef] 30. Ghasabkolaei, N.; Janalizadeh, A.; Jahanshahi, M.; Roshan, N.; Ghasemi, S.E. Physical and geotechnical properties of cement- treated clayey soil using silica nanoparticles: An experimental study. Eur. Phys. J. Plus 2016, 131, 134. [CrossRef] 31. Bahmani, S.H.; Farzadnia, N.; Asadi, A.; Huat, B.B. The effect of size and replacement content of nanosilica on strength development of cement treated residual soil. Constr. Build. Mater. 2016, 118, 294–306. [CrossRef] 32. Rosales, J.; Agrela, F.; Marcobal, J.R.; Diaz-López, J.L.; Cuenca-Moyano, G.M.; Caballero, Á.; Cabrera, M. Use of Nanomaterials in the Stabilization of Expansive Soils into a Road Real-Scale Application. Materials 2020, 13, 3058. [CrossRef] [PubMed] 33. de Fomento, E. Instrucción de carreteras. Norma 6.1 IC: Secciones de firme. Norm. Instr. Construcción 2003, 41. 34. de Fomento, M. PG-3: Pliego de Prescripciones Técnicas Generales Para Obras de Carreteras y Puentes; Ediciones Liteam SL: Madrid, Spain, 2002. 35. Zhu, X.B.; Wang, H.; Zhang, Y.C. Evaluation of subgrade compactness. In Applied Mechanics and Materials; Trans Tech Publications Ltd.: Bäch, Switzerland, 2013; pp. 1663–1667. 36. Poon, C.S.; Chan, D. Feasible use of recycled concrete aggregates and crushed clay brick as unbound road sub-base. Constr. Build. Mater. 2006, 20, 578–585. [CrossRef] 37. Alireza, S.G.S.; Mohammad, M.S.; Hasan, B.M. Application of nanomaterial to stabilize a weak soil. In Proceedings of the 7th International Conference on Case Histories in Geotechnical Engineering, Chicago, IL, USA, 29 April–4 May 2013. 38. Mousavi, F.; Abdi, E.; Rahimi, H. Effect of polymer stabilizer on swelling potential and CBR of forest road material. KSCE J. Civ. Eng. 2014, 18, 2064–2071. [CrossRef] 39. Haeri, S.M.; Hosseini, A.M.; Shahrabi, M.M.; Soleymani, S. Comparison of strength characteristics of Gorgan loessial soil improved by nano-silica, lime and Portland cement. In From Fundamentals to Applications in Geotechnics; IOS Press: Amsterdam, The Netherlands, 2015; pp. 1820–1827. 40. Dutor, A.-B.; Castilla-Molina, J.; Gómez-Casado, J.-A. Instrucción para el diseño de firmes de la red de carreteras de Andalucía; ICAFIR: Sevilla, Spain, 2007. 41. Powell, W.; Potter, J.; Mayhew, H.; Nunn, M. The Structural Design of Bituminous Roads; Transport and Road Research Laboratory (TRRL): Crowthorne, UK, 1984. 42. Xuan, D.; Houben, L.; Molenaar, A.; Shui, Z. Mechanical properties of cement-treated aggregate material—A review. Mater. Des. 2012, 33, 496–502. [CrossRef] 43. de Andalucía, J. Instituto De Investigación Y Formación Agraria Y Pesquera (IFAPA). Available online: https://www. juntadeandalucia.es/agriculturaypesca/ifapa/riaweb/web/estacion/23/102 (accessed on 21 May 2021). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Sciences Multidisciplinary Digital Publishing Institute

Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume Roads

Loading next page...
 
/lp/multidisciplinary-digital-publishing-institute/feasibility-of-using-nanosilanes-in-a-new-hybrid-stabilised-soil-LJ9YeojFsp
Publisher
Multidisciplinary Digital Publishing Institute
Copyright
© 1996-2021 MDPI (Basel, Switzerland) unless otherwise stated Disclaimer The statements, opinions and data contained in the journals are solely those of the individual authors and contributors and not of the publisher and the editor(s). MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Terms and Conditions Privacy Policy
ISSN
2076-3417
DOI
10.3390/app11219780
Publisher site
See Article on Publisher Site

Abstract

applied sciences Article Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume Roads 1 1 , 2 , 3 1 , José Luis Díaz-López , Manuel Cabrera * , José Ramón Marcobal , Francisco Agrela * and Julia Rosales Construction Engineering Area, University of Córdoba, Campus of Rabanales, 14014 Córdoba, Spain; jl.diaz@uco.es (J.L.D.-L.); jrosales@uco.es (J.R.) Department of Pavements, Sacyr Construction, 28040 Madrid, Spain; jose.marcobal@upm.es Transport Engineering, Territory and Urban Planning Department, Campus Ciudad Universitaria, Universidad Politécnica de Madrid, 28040 Madrid, Spain * Correspondence: manuel.cabrera@uco.es (M.C.); fagrela@uco.es (F.A.) Abstract: The application of new materials for soil stabilisation is a growing field of study in recent years. In this work, the effect of two types of silica-based nanomaterials combined with binders (quicklime and cement) are studied to stabilise soils and form structural layers for rural and low volume roads. The physical and chemical properties of the materials have been determined, as well as the mechanical behaviour of the stabilised soil. Three hybrid stabilised soil sections have been designed using a multilayer elastic model, executed at full scale and measuring the evolution of their properties in the medium to short term. The results show that the application of silica-based nanomaterials and two types of binders on the tread layers provide high structural stability and good behaviour of the sections. Citation: Díaz-López, J.L.; Cabrera, Keywords: hybrid stabilisation; mechanical behaviour; real scale application; nanomaterials M.; Marcobal, J.R.; Agrela, F.; Rosales, J. Feasibility of Using Nanosilanes in a New Hybrid Stabilised Soil Solution in Rural and Low-Volume Roads. 1. Introduction Appl. Sci. 2021, 11, 9780. https:// doi.org/10.3390/app11219780 Over several decades, a large primary and secondary network of pavement roads and highways has been introduced by all developed countries to connect urban centres, Academic Editor: Angeles metropolis and industrial areas with each other. This connectivity, combined with marine Sanroman Braga and air modes of transport, has generated a great socioeconomic impact derived from the transport of goods and passengers [1]. Received: 9 September 2021 However, there are many rural areas, small cities and, especially, developing countries Accepted: 15 October 2021 where it is not always possible to build this great network of roads due to economic, Published: 20 October 2021 accessibility or constructability issues, as well as the fact that other modes of transport have a lower level of development and, therefore, cannot even be implemented. In this Publisher’s Note: MDPI stays neutral way, rural and low volume roads are a major opportunity to improve the connectivity and with regard to jurisdictional claims in socioeconomic development of these areas [2]. published maps and institutional affil- Rural roads have different functions according to the level of development of the iations. country where they are built. In developed countries, rural roads are usually designed to connect towns with low populations and agricultural and livestock areas with a low volume of vehicles. On the contrary, in developed countries, rural roads are designed to meet the socioeconomic needs of the rural population, connecting remote areas to basic Copyright: © 2021 by the authors. health services, education and markets [3]. Licensee MDPI, Basel, Switzerland. These types of road can be composed of a subgrade and a thin asphaltic layer or, This article is an open access article most commonly, by compaction of an unpaved unbound granular material [4] or even distributed under the terms and compaction of stabilised soils that are found in the location of the road [5]. conditions of the Creative Commons However, in most cases, these soils, especially clayey soils, present geotechnical Attribution (CC BY) license (https:// problems, such as a lack of bearing capacity, high plasticity or swelling potential, that creativecommons.org/licenses/by/ prevent their use. In order to improve the properties of soils and increase their range of 4.0/). Appl. Sci. 2021, 11, 9780. https://doi.org/10.3390/app11219780 https://www.mdpi.com/journal/applsci Appl. Sci. 2021, 11, 9780 2 of 20 use in civil engineering applications, soil stabilisation for application in road layers has become widespread in recent years [6,7]. Soil stabilisation is defined as the improvement of the shear strength, durability stiff- ness and reduction of the plasticity and swelling potential of soils achieved by mechanical means or the addition of stabilising products, such as a hydraulic binder, fly and rice husk ash, chemical stabilisation, recycled waste and by-products, etc. [8]. Among all the stabilising products, stabilisation with binders (commonly lime and cement) have been widely used by numerous authors in recent decades [9–12]. Never- theless, the production of these traditional materials generates a negative environmental impact due to the use of raw material resources and the high CO emissions involved in their production [13]. For the reason mentioned above, in recent years, several studies that analyse the possible stabilisation of soils with industrial by-products or recycled materials, such as biomass fly and bottom ash [14–17], phosphogypsum [18,19], steel slag [20,21] and magne- sia oxides [22,23], among many others, have been increased. In addition, in recent years, an alternative to conventional stabilising products and wastes and by-products has emerged: nanomaterials. Nanomaterials are particles with a typical size of between 1 and 100 nm, with a very high specific surface area, which implies a very high reactivity [24], achieving soil improvements with very low dosages. Nanomaterials commonly used in soil stabilisation are composed of simple oxides, such as SiO , TiO or CaCO , or carbon nanotubes [25,26]. 2 2 3 Although a large number of nanoparticles has been developed, the pozzolanic capacity of Nano-SiO2 and its reactivity with lime and cement to form calcium silicate hydrate (CSH) compounds have led most studies to focus on it [27–31]. Kulanthaivel et al. [27] studied the effect of using synthesised Nano-SiO by a sol– gel process together with cement to stabilise clay soil, concluding that a 7% addition of nano-SiO improves the unconfined compressive strength in a ratio of 5:24 and reduces the permeability in the range of 0.01976 cm/s–0.01198 cm/s of the soil, which is consistent with previous studies [28]. Ghasabkolaei et al. [30] and Bahmani et al. [31] studied soil stabilisation with cements and nanosilane with dosages lower than 1% by weight of Nano-SiO , observing an increase in the unconfined compressive strength in all the mixtures with nanoparticulate additives and a high formation rate of calcium silicate hydrate (CSH), which implies an improvement in the soil properties. The present study shows new structural solutions for rural and low-volume roads based on soil stabilised with organosilanes and small amounts of lime and cement. The design parameters of the new sections, which were obtained through laboratory tests, were tested in a real-scale application to verify the feasibility of using the developed solutions. In addition, a medium-to-short-term performance study of the road sections built was carried out to check their durability. 2. Research Purpose and Experimental Programme This work is a continuation of the one carried out by Rosales et al. [32] in which a conventional control section was made according to Spanish specifications [33], and two alternative experimental sections based on nanomaterials and quicklime were performed in order to reduce the total thickness of the treated layers. In the present work, three new trial sections are proposed with a hybrid stabilisation process in which nanomaterials and small amounts of quicklime and cement are used to reduce their treated thickness with respect to the control section by 40% and improve the mechanical properties and durability. To carry out the trial sections in a successful manner, the following phases were followed in the present work: Phase 1. Conceptualisation and predesign of alternative sections based on the study conducted by Rosales et al. [32]. Appl. Sci. 2021, 11, 9780 3 of 20 Phase 2. Physicochemical characterisation of the materials involved in the work— namely, expansive clayey soil, sandy soil, which is a rejection of the production of crushed gravel, lime, cement and nanomaterials. Phase 3. Laboratory study to obtain the design parameters of the sections. In this phase, the percentages of soil for the mixtures were defined, as well as the quantities of the binders and organosilanes. The California Bearing Ratio (CBR) index and unconfined compressive strength (UCS) were obtained. Phase 4. Trial section structural design using Everstress 5.0 software. From the CBR index and simple compressive strength, the elastic modulus of the layers composing the sections were determined. Once the elastic parameters of the materials had been defined, the different section options were evaluated by means of a multilayer elastic analysis in the software. Phase 5. Construction and survey of the experimental sections. In this phase, the road sections were built, and short- and long-term checks of the bearing capacity of the soil were performed by non-destructive tests, such as plate loading and a falling weight deflectometer (FWD). 3. Phase 1. Section Pre-Design According to the results obtained by Rosales et al. [32], three alternative sections composed of two layers were proposed: one of lime stabilised expansive clayed soil (CS) and another upper one of a mixture of CS and sandy soil (SS) stabilised with cement. In the first pre-design stage, the layer thicknesses and soil proportions were not Appl. Sci. 2021, 11, x FOR PEER REVIEW 4 of 21 defined; however, the sections and their construction process were defined in a qualitative manner, as shown in Figure 1. Figure 1. Manufacturing process scheme of the hybrid stabilised sections. (a) Step 1: Pre-compaction Figure 1. Manufacturing process scheme of the hybrid stabilised sections. (a) Step 1: Pre-compaction of CS. (b) Step 2: Spreading and mixing of SS and cement. (c) Step 3: Compaction of CS+SS. of CS. (b) Step 2: Spreading and mixing of SS and cement. (c) Step 3: Compaction of CS+SS. 4. Laboratory Tests 4.1. Phase 2: Materials and Physicochemical Characterisation In this section, the materials used during the research for the development of the tests and construction of the trial sections are shown. 4.1.1. Soils: Expansive Clayey Soil and Sandy Soil In this work, two types of soils have been analysed for the subsequent stabilisation and construction of the layers of the trial section: an expansive clayey soil (CS) and a quarry reject from the production of crushed gravel called sandy soil (SS). The clayey soil (CS) comes from the plot where the real-scale sections are built, in Villacarrillo (Jaén), Andalusia. The sandy soil comes as a reject from a crushed gravel production quarry located near Villacarrillo. The rejects are subjected to a sieving process to achieve a particle size within the parameters of Spanish regulations [34]. Table 1 shows the physicochemical properties of the two soils analysed, and Figure 2 shows the granulometric curves of the soils. Appl. Sci. 2021, 11, x FOR PEER REVIEW 5 of 21 Appl. Sci. 2021, 11, 9780 4 of 20 Table 1. Physical and chemical properties of CS and SS. Properties CS SS Standard Test Atterberg limits UNE-EN ISO 17892-12:2019 As shown in Figure 1, the process of construction of the hybrid stabilised sections was Liquid Limit (%) 53.0 16.1 carried out in three steps: Step 1. Spreading, mixing and pre-compaction of the quicklime (or quicklime and Plastic limit (%) 21.6 11.9 nanomaterial) stabilised the clayey soil layer on the original ground. The nanomaterials Plasticity index (%) 31.4 4.2 were added, according to the dosage shown in the following sections, into the mixing water. Grain size distribution UNE-EN ISO 17892-4:2019 The curing time of the layer was one day. Gravel (>4 mm) (%) 0.0 77.1 Step 2. After one day, the spreading and mixing of the sandy soil layer on top of the Sand (0.063–4 mm) (%) 5.4 17.4 compacted quicklime (or quicklime and nanomaterial) stabilised the clayey soil layer and Silt and clay (<0.063 mm) (%) 94.6 5.5 the spreading of the cement. Max Step imun d 3. Compaction ry density (kg/ of the layer m ) composed 1.55 of 2. a 28 per centage of UNE 10 compacted 350quicklime 1:1994 (or quicklime and nanomaterial) stabilised clayey soil plus sandy soil and cement. The result Organic matter content (%) 0.31 0.02 UNE 103204:2019 is called hybrid stabilised soil. Water-soluble sulphate (% SO3) 0.05 0.07 UNE-EN 1744-1:2010+A1:2013 The following sections analyse the physicochemical properties of the materials in- Main components XRF (%) volved, as well as the mechanical properties of the mixes to be defined to optimise both the P 0.08 0.03 thickness of the mixes and the percentages of the binders. Si 25.57 3.32 4. Laboratory Tests Ca 14.16 31.2 4.1. Phase 2: Materials and Physicochemical Characterisation Al 2.69 0.83 In this section,S the materials used during 0.03 the 0. resear 05 ch for the development of the tests and construction of the trial sections are shown. K 1.66 1.74 Mg 1.37 16.2 4.1.1. Soils: Expansive Clayey Soil and Sandy Soil Fe 2.42 1.06 In this work, two types of soils have been analysed for the subsequent stabilisation and construction of the layers of the trial section: an expansive clayey soil (CS) and a quarry As shown in Table 1, the clayey soil shows a very high plasticity, with a plasticity reject from the production of crushed gravel called sandy soil (SS). index greater than 30, a discontinuous granulometry in which 95% of the particles are fine The clayey soil (CS) comes from the plot where the real-scale sections are built, in clays and silts and a maximum particle size of 5 mm. Villacarrillo (Jaén), Andalusia. In contrast, the sandy soil has a continuous grain size, composed mainly of particles The sandy soil comes as a reject from a crushed gravel production quarry located near larger than 4 mm, with a maximum aggregate size of 32 mm and a percentage of fine Villacarrillo. The rejects are subjected to a sieving process to achieve a particle size within grains of less than 6%. The plasticity index of SS is 4.2, which indicates that it is not a very the parameters of Spanish regulations [34]. Table 1 shows the physicochemical properties pl of as the tic ma twoteria soilsl. analysed, and Figure 2 shows the granulometric curves of the soils. Figure 2. Particle size distribution of clayey soil and sandy soil. Figure 2. Particle size distribution of clayey soil and sandy soil. Both materials have low percentages of organic matter content and water-soluble sul- phates, which make them suitable for use in road layers in accordance with Spanish reg- ulations. Appl. Sci. 2021, 11, 9780 5 of 20 Table 1. Physical and chemical properties of CS and SS. Properties CS SS Standard Test Atterberg limits UNE-EN ISO 17892-12:2019 Liquid Limit (%) 53.0 16.1 Plastic limit (%) 21.6 11.9 Plasticity index (%) 31.4 4.2 Grain size distribution UNE-EN ISO 17892-4:2019 Gravel (>4 mm) (%) 0.0 77.1 Sand (0.063–4 mm) (%) 5.4 17.4 Silt and clay (<0.063 mm) (%) 94.6 5.5 Maximun dry density (kg/m ) 1.55 2.28 UNE 103501:1994 Organic matter content (%) 0.31 0.02 UNE 103204:2019 UNE-EN Water-soluble sulphate (% SO ) 0.05 0.07 1744-1:2010+A1:2013 Main components XRF (%) P 0.08 0.03 Si 25.57 3.32 Ca 14.16 31.2 Al 2.69 0.83 S 0.03 0.05 K 1.66 1.74 Mg 1.37 16.2 Fe 2.42 1.06 Appl. Sci. 2021, 11, x FOR PEER REVIEW As shown in Table 1, the clayey soil shows a very high plasticity, with a plasticity 6 of 21 index greater than 30, a discontinuous granulometry in which 95% of the particles are fine clays and silts and a maximum particle size of 5 mm. In contrast, the sandy soil has a continuous grain size, composed mainly of particles Regarding the composition of the soils, an analysis of the major compounds by X-ray larger than 4 mm, with a maximum aggregate size of 32 mm and a percentage of fine fluorescence (XRF) and mineralogical analysis by X-ray diffraction (XRD) was carried out, grains of less than 6%. The plasticity index of SS is 4.2, which indicates that it is not a very as shown in Table 1 and Figures 3 and 4. plastic material. The clayey soil presents a main composition of silicon, calcium and aluminium, a Both materials have low percentages of organic matter content and water-soluble sul- typical composition of clays, which is observed in a mineralogy composed of quartz, cal- phates, which make them suitable for use in road layers in accordance with Spanish regulations. cite and montmorrollite-type clay minerals. Regarding the composition of the soils, an analysis of the major compounds by X-ray The sandy soil has a main composition of calcium and magnesium, a typical compo- fluorescence (XRF) and mineralogical analysis by X-ray diffraction (XRD) was carried out, sition of dolomite minerals, as shown in the XRD of Figure 4. as shown in Table 1 and Figures 3 and 4. Figure 3. XRD of clayey soil. Figure 3. XRD of clayey soil. Figure 4. XRD of sandy soil. 4.1.2. Binders: Quicklime and Cement Binders are materials that react chemically with water, forming cementitious com- pounds that can bind and improve the properties of soils, among other functions. In the present work, two binders have been applied: commercial quicklime (QL)-type CL 90-Q and a commercial ordinary Portland cement (OPC)-type CEM II/B-L 32.5R. Table 2 shows the composition in the form of oxides of both binders. Table 2. Compositions of the binders. Composition (%)/Binder SiO2 Al2O3 Fe2O3 CaO MgO SO3 K2O CEM II/B-L 32.5R 16.20 3.83 2.75 60.41 0.84 2.64 0.64 CL 90-Q 2.37 1.47 0.21 57.26 13.78 0.72 0.07 Appl. Sci. 2021, 11, x FOR PEER REVIEW 6 of 21 Regarding the composition of the soils, an analysis of the major compounds by X-ray fluorescence (XRF) and mineralogical analysis by X-ray diffraction (XRD) was carried out, as shown in Table 1 and Figures 3 and 4. The clayey soil presents a main composition of silicon, calcium and aluminium, a typical composition of clays, which is observed in a mineralogy composed of quartz, cal- cite and montmorrollite-type clay minerals. The sandy soil has a main composition of calcium and magnesium, a typical compo- sition of dolomite minerals, as shown in the XRD of Figure 4. Appl. Sci. 2021, 11, 9780 6 of 20 Figure 3. XRD of clayey soil. Figure Figure 4. 4. XRD XRD of of sandy sandy soi soil. l. The clayey soil presents a main composition of silicon, calcium and aluminium, a 4.1.2. Binders: Quicklime and Cement typical composition of clays, which is observed in a mineralogy composed of quartz, calcite Binders are materials that react chemically with water, forming cementitious com- and montmorrollite-type clay minerals. pounds that can bind and improve the properties of soils, among other functions. The sandy soil has a main composition of calcium and magnesium, a typical composi- In the present work, two binders have been applied: commercial quicklime (QL)-type tion of dolomite minerals, as shown in the XRD of Figure 4. CL 90-Q and a commercial ordinary Portland cement (OPC)-type CEM II/B-L 32.5R. Table 2 shows the composition in the form of oxides of both binders. 4.1.2. Binders: Quicklime and Cement Binders are materials that react chemically with water, forming cementitious com- Table 2. Compositions of the binders. pounds that can bind and improve the properties of soils, among other functions. Com In the pospr ition esent (%work, )/Bindtwo er binde SiO rs 2 have Al2O been 3 Fe applied: 2O3 CaO commer MgO cial quicklime SO3 K (QL)- 2O type CL 90-Q and a commercial ordinary Portland cement (OPC)-type CEM II/B-L 32.5R. CEM II/B-L 32.5R 16.20 3.83 2.75 60.41 0.84 2.64 0.64 Table 2 shows the composition in the form of oxides of both binders. CL 90-Q 2.37 1.47 0.21 57.26 13.78 0.72 0.07 Table 2. Compositions of the binders. Composition (%)/Binder SiO Al O Fe O CaO MgO SO K O 2 2 3 2 3 3 2 CEM II/B-L 32.5R 16.20 3.83 2.75 60.41 0.84 2.64 0.64 CL 90-Q 2.37 1.47 0.21 57.26 13.78 0.72 0.07 4.1.3. Nanomaterials Three types of silica-based nanomaterials were used in this study, named N1, N2 and N3. N1 is a concentrated liquid solution of sodium silicate for soil stabilisation, together with binders such as lime or cement. N2 is an organosilane, and N3 is a silica-based acrylic copolymeric in aqueous solution form, intended to be used together in a 1:1 ratio. For the three nanomaterials, its chemical composition was obtained by an X-ray fluorescence (XRF) analysis and a thermogravimetric analysis (TGA). Table 3 shows the results of the XRF analysis, and Figure 5 shows the results of the TGA. As can be observed in Table 3, the composition of N1 was mainly silicon and sodium, elements that form the nano-sized sodium silicates that make up this material. N2 and N3 were composed only of silicon as the main element and traces of the other elements. This sodium is dissolved in the form of organosilanes in N2 and in the form of an acrylic copolymer in N3. The results of the TGA analysis are shown in Figure 5. In N1, a progressive weight loss was observed of up to 200 C, which was due to the loss of water in the nanosilane. This nanomaterial showed a high silica content of 32%. N2 showed a higher weight loss; the silica content at 400 C was 17%. N3 completely loses its mass at a temperature of 400 C. Appl. Sci. 2021, 11, x FOR PEER REVIEW 7 of 21 4.1.3. Nanomaterials Three types of silica-based nanomaterials were used in this study, named N1, N2 and N3. N1 is a concentrated liquid solution of sodium silicate for soil stabilisation, together Appl. Sci. 2021, 11, 9780 7 of 20 with binders such as lime or cement. N2 is an organosilane, and N3 is a silica-based acrylic copolymeric in aqueous solution form, intended to be used together in a 1:1 ratio. For the three nanomaterials, its chemical composition was obtained by an X-ray flu- The higher stability of silica in the form of sodium silicate compared to other solutions, orescence (XRF) analysis and a thermogravimetric analysis (TGA). Table 3 shows the re- such as organosilanes or acrylic copolymers were observed, organosilane compounds lost sults of the XRF analysis, and Figure 5 shows the results of the TGA. 83% of their mass and acrylic compounds did not retain their mass after the test. Table 3. Main composition XRF (%) of the nanomaterials. 4.2. Phase 3: Mechanical Behaviour of Soil Mixtures Main Composition XRF (%) N1 N2 N3 In this section, the requirements for the thickness layer designs are determined. The Si 24.200 40.600 38.400 Modified Proctor compaction test, California Bearing Ratio (CBR), unconfined compressive Ca 0.279 0.014 0.097 strength and shear test are performed and shown. Figure 6 shows photographs of the Al 0.180 0.086 0.531 experimental methods and materials applied in this work. S 0.020 0.014 0.051 Table 3. Main composition XRF (%) of the nanomaterials. K 0.026 - 0.053 Mg 0.150 - 0.093 Main Composition XRF (%) N1 N2 N3 Fe 0.057 0.050 0.488 Si 24.200 40.600 38.400 Na 12.100 - 0.207 Ca 0.279 0.014 0.097 Al 0.180 0.086 0.531 As can be observed in Table 3, the composition of N1 was mainly silicon and sodium, S 0.020 0.014 0.051 elements that form the n K ano-sized sodium silic 0.026 ates that make - up this material. N2 0.053 and N3 Mg 0.150 - 0.093 were composed only of silicon as the main element and traces of the other elements. This Fe 0.057 0.050 0.488 sodium is dissolved in the form of organosilanes in N2 and in the form of an acrylic co- Na 12.100 - 0.207 polymer in N3. Appl. Sci. 2021, 11, x FOR PEER REVIEW 8 of 21 4.2. Phase 3: Mechanical Behaviour of Soil Mixtures In this section, the requirements for the thickness layer designs are determined. The Modified Proctor compaction test, California Bearing Ratio (CBR), unconfined compres- sive strength and shear test are performed and shown. Figure 6 shows photographs of the Figure 5. TGA curves of the nanomaterials. Figure 5. experiment TGA al met curves of hods and the n mat anomaterials. erials applied in this work. The results of the TGA analysis are shown in Figure 5. In N1, a progressive weight loss was observed of up to 200 °C, which was due to the loss of water in the nanosilane. This nanomaterial showed a high silica content of 32%. N2 showed a higher weight loss; the silica content at 400 °C was 17%. N3 completely loses its mass at a temperature of 400 °C. The higher stability of silica in the form of sodium silicate compared to other solu- tions, such as organosilanes or acrylic copolymers were observed, organosilane com- pounds lost 83% of their mass and acrylic compounds did not retain their mass after the test. Figure Figure 6. 6. E Experimental xperimental meth methods ods and and mater materials ials photographs photographs. . (a) Soils an (a) Soils d binders. ( and binders. b) Modified (b) Modified Proctor test. (c) CBR index test. (d) Unconfined compressive strength test. Proctor test. (c) CBR index test. (d) Unconfined compressive strength test. 4.2.1. Mix Design Six mixes of soil with binders and/or nanomaterials were defined to design the three alternative innovative sections studied in this article. Table 4 shows the mixtures analysed in the laboratory, the dosage of each material and its designation. Table 4. Dosages of the mixtures. Materials (kg/m ) Designation CS SS Quicklime OPC N1 N2 N3 CS + 1.5%QL 1590.00 - 23.85 - - - - CS + 1.5%QL + N1 1590.00 - 23.85 - 0.910 - - CS + 1.5%QL + N2&N3 1590.00 - 23.85 - - 1.000 1.000 AM-1 660.00 1380.00 9.90 20.40 - - - AM-2 660.00 1380.00 9.90 20.40 0.910 - - AM-3 660.00 1380.00 9.90 20.40 - 1.000 1.000 The dosages of the mixtures shown in the table above are expressed in weigh (kg) per cubic metre of material. Each material was added emulating the real process of road construction: Quicklime was added to the dry mass of CS in all mixtures. OPC was added to the total dry mass of soil (CS and SS), and nanomaterials were added to the dry volume of CS. 3 3 Finally, AM-1, AM-2 and AM-3 were composed of 0.40 m of CS and 0.60 m of SS per one cubic metre of the total mix. The AM mixes were made with a special manufacturing process, which is explained below: Step 1. Forty percent in volume of CS was mixed with an addition of 1.5 w% of quick- lime (AM-1), 1.5 w% of quicklime plus N1 (AM-2) or 1.5 w% of quicklime and N2 plus N3 Appl. Sci. 2021, 11, 9780 8 of 20 4.2.1. Mix Design Six mixes of soil with binders and/or nanomaterials were defined to design the three alternative innovative sections studied in this article. Table 4 shows the mixtures analysed in the laboratory, the dosage of each material and its designation. Table 4. Dosages of the mixtures. Materials (kg/m ) Designation CS SS Quicklime OPC N1 N2 N3 CS + 1.5%QL 1590.00 - 23.85 - - - - CS + 1.5%QL + N1 1590.00 - 23.85 - 0.910 - - CS + 1.5%QL + N2&N3 1590.00 - 23.85 - - 1.000 1.000 AM-1 660.00 1380.00 9.90 20.40 - - - AM-2 660.00 1380.00 9.90 20.40 0.910 - - AM-3 660.00 1380.00 9.90 20.40 - 1.000 1.000 The dosages of the mixtures shown in the table above are expressed in weigh (kg) per cubic metre of material. Each material was added emulating the real process of road construction: Quicklime was added to the dry mass of CS in all mixtures. OPC was added to the total dry mass of soil (CS and SS), and nanomaterials were added to the dry volume of CS. 3 3 Finally, AM-1, AM-2 and AM-3 were composed of 0.40 m of CS and 0.60 m of SS per one cubic metre of the total mix. The AM mixes were made with a special manufacturing process, which is explained below: Step 1. Forty percent in volume of CS was mixed with an addition of 1.5 w% of quicklime (AM-1), 1.5 w% of quicklime plus N1 (AM-2) or 1.5 w% of quicklime and N2 plus N3 (AM-3). The mixture was compacted according to the Modified Proctor compaction test and stored in a wet chamber with a minimum moisture of 90%. Step 2. One day after compaction, the mixture was decompressed into particles with a size smaller than 25 mm. Step 3. The stabilised CS was mixed with 60 v% of SS and 1 w% of OPC compared to the total dry mass of soil. The mixture was compacted according to the Modified Proctor test and stored in a wet chamber (compressive strength and shear tests) or curing tank (CBR test) before the samples were tested. 4.2.2. Compaction Test: Modified Proctor The level of compaction is a key parameter in road layer construction. A proper compaction ensures the strength and durability of the road; conversely, if the road has not been adequately compacted, it becomes unstable, which can result in a possible differential settlement. Settlement in a road implies pavement deformation and cracking that, added to storm water infiltration, causes a serious impact on traffic and road safety [35]. Compaction is studied through the moisture–dry density relationship, which is deter- mined according to the Modified Proctor Test, UNE EN 103501: 1994 standard. The moisture–dry density relationship, which is shown in Figure 7, is a very useful tool for understanding compacted soil behaviour. The maximum dry density to the optimum moisture content is obtained, which allowed the samples made to achieve the optimal mechanical behaviour. Additionally, the curve shape indicates the sensitivity of the soil to the water addition. If the curve is flatter, the maximum dry density is less affected by moisture changes; however, if the curve is sharper, small changes in the moisture content greatly affect the maximum dry density [36], as can be observed in Figure 7. Appl. Sci. 2021, 11, x FOR PEER REVIEW 9 of 21 (AM-3). The mixture was compacted according to the Modified Proctor compaction test and stored in a wet chamber with a minimum moisture of 90%. Step 2. One day after compaction, the mixture was decompressed into particles with a size smaller than 25 mm. Step 3. The stabilised CS was mixed with 60 v% of SS and 1 w% of OPC compared to the total dry mass of soil. The mixture was compacted according to the Modified Proctor test and stored in a wet chamber (compressive strength and shear tests) or curing tank (CBR test) before the samples were tested. 4.2.2. Compaction Test: Modified Proctor The level of compaction is a key parameter in road layer construction. A proper com- paction ensures the strength and durability of the road; conversely, if the road has not been adequately compacted, it becomes unstable, which can result in a possible differen- tial settlement. Settlement in a road implies pavement deformation and cracking that, added to storm water infiltration, causes a serious impact on traffic and road safety [35]. Compaction is studied through the moisture–dry density relationship, which is de- termined according to the Modified Proctor Test, UNE EN 103501: 1994 standard. The moisture–dry density relationship, which is shown in Figure 7, is a very useful tool for understanding compacted soil behaviour. The maximum dry density to the opti- mum moisture content is obtained, which allowed the samples made to achieve the opti- mal mechanical behaviour. Additionally, the curve shape indicates the sensitivity of the soil to the water addition. If the curve is flatter, the maximum dry density is less affected Appl. Sci. 2021, 11, 9780 9 of 20 by moisture changes; however, if the curve is sharper, small changes in the moisture con- tent greatly affect the maximum dry density [36], as can be observed in Figure 7. Figure 7. Moisture–dry density curves. Figure 7. Moisture–dry density curves. Analysing Figure 7, SS shows the highest maximum dry density, 2.28 kg/m , and Analysing Figure 7, SS shows the highest maximum dry density, 2.28 kg/m , and minimum optimum moisture content, 6.2%, likely due to its physical properties, such as a minimum optimum moisture content, 6.2%, likely due to its physical properties, such as higher quantity of coarse particles and its higher density. However, CS stabilised with lime a higher quantity of coarse particles and its higher density. However, CS stabilised with and quicklime plus nanomaterials show values contrary to SS due to the clay gradation, lime and quicklime plus nanomaterials show values contrary to SS due to the clay grada- which is composed mainly of fine particles and its great absorption of water, which is a tion, which is composed mainly of fine particles and its great absorption of water, which typical characteristic of expansive soil. is a typical characteristic of expansive soil. As can be observed, the addition of nanomaterials in the amount used in this article (0.056 w% N1 and 0.12% N1&N2) shows a low effect on the compaction parameters, especially for the dry density, which is consistent with the results obtained in previous studies of soil stabilisation with nanoparticles [29]. However, Alireza et al. [37] showed a decrease in the maximum dry density and an increase in the optimum compaction humidity in a soil stabilised with 5% lime from nano-SiO additions greater than 1%. Finally, the AM-1, AM-2 and AM-3 mixtures have an intermediate behaviour among the materials that compose them. A slight increase in the maximum dry density due to the addition of nanomaterials was observed in the AM-2 and AM-3 mixtures, likely due to the interaction of the stabilised soils with nanomaterials and the OPC. 4.2.3. Design Parameters: California Bearing Ratio (CBR) and Compressive Strength The structural behaviour of the road layers is determined by the load bearing capacity. According to the type of road layer executed, this bearing capacity is measured according to the CBR (California Bearing ratio) or unconfined compressive strength. The CBR index measures the bearing capacity of soils and compacted aggregates used in the construction of road bases or subbases. The CBR index depends on the density and moisture conditions of the samples. In this study, the samples were compacted according to the optimum moisture obtained in the Modified Proctor Test to reach the maximum dry density and the highest possible bearing capacity. The CBR value is carried out in accordance with UNE 103-502. The unconfined compressive strength (UCS) was performed according to the NLT- 305/90 standard in specimens 177.8 mm high and 152.4 mm in diameter. The UCS measured the resistance in cohesive soil or cement treatment soil or granular materials. Like the CBR samples, these samples were manufactured under the optimal compaction conditions obtained in the Modified Proctor Test. Spanish specifications [34] establish that lime-stabilised soil must comply with the minimum CBR index; however, cement-treated soil must exceed a minimum value of Appl. Sci. 2021, 11, x FOR PEER REVIEW 10 of 21 As can be observed, the addition of nanomaterials in the amount used in this article (0.056 w% N1 and 0.12% N1&N2) shows a low effect on the compaction parameters, es- pecially for the dry density, which is consistent with the results obtained in previous stud- ies of soil stabilisation with nanoparticles [29]. However, Alireza et al. [37] showed a decrease in the maximum dry density and an increase in the optimum compaction humidity in a soil stabilised with 5% lime from nano- SiO2 additions greater than 1%. Finally, the AM-1, AM-2 and AM-3 mixtures have an intermediate behaviour among the materials that compose them. A slight increase in the maximum dry density due to the addition of nanomaterials was observed in the AM-2 and AM-3 mixtures, likely due to the interaction of the stabilised soils with nanomaterials and the OPC. 4.2.3. Design Parameters: California Bearing Ratio (CBR) and Compressive Strength The structural behaviour of the road layers is determined by the load bearing capac- ity. According to the type of road layer executed, this bearing capacity is measured ac- cording to the CBR (California Bearing ratio) or unconfined compressive strength. The CBR index measures the bearing capacity of soils and compacted aggregates used in the construction of road bases or subbases. The CBR index depends on the density and moisture conditions of the samples. In this study, the samples were compacted ac- cording to the optimum moisture obtained in the Modified Proctor Test to reach the max- imum dry density and the highest possible bearing capacity. The CBR value is carried out in accordance with UNE 103-502. The unconfined compressive strength (UCS) was performed according to the NLT- 305/90 standard in specimens 177.8 mm high and 152.4 mm in diameter. The UCS meas- ured the resistance in cohesive soil or cement treatment soil or granular materials. Like the CBR samples, these samples were manufactured under the optimal compaction con- Appl. Sci. 2021, 11, 9780 10 of 20 ditions obtained in the Modified Proctor Test. Spanish specifications [34] establish that lime-stabilised soil must comply with the minimum CBR index; however, cement-treated soil must exceed a minimum value of un- confined compressive strength. The results obtained in the laboratory test are shown in unconfined compressive strength. The results obtained in the laboratory test are shown in Figure 8. Figure 8. Figure Figure 8. 8. CBR i CBR index ndex and com and compr pressive essive streng strength. th. Analysing the results shown in Figure 8, the use of different nanosilica improve the Analysing the results shown in Figure 8, the use of different nanosilica improve the bearing capacity of soils. bearing capacity of soils. The addition of 0.056 w% of N1 and 0.112 w% of N2&N3 increase the CBR index by The addition of 0.056 w% of N1 and 0.112 w% of N2&N3 increase the CBR index by 31.5% and 33.2%, respectively. 31.5% and 33.2%, respectively. Similar increases in the CBR index have been observed in previous studies [38] with similar dosages, probably due to the reactions of these nanomaterials with cement and the soil minerals themselves, leading to pozzolanic cementitious reactions. This behaviour has been observed in previous studies with higher increases in the bearing capacity or simple compressive strength, due to the higher addition percentages of the nanomaterials, between 1 and 7% [37,39]. The AM-1, AM-2 and AM-3 mixtures showed similar CBR index values, in the order of 100% of the CBR index with slightly higher values in the mixtures with nanomaterials, which shows an excellent bearing capacity. At the same time, unconfined compressive strength values of between 1.08 and 1.23 MPa were obtained, relatively high values for mixes with a total of 1 w% cement, normal values for stabilisation being a minimum of 3%, according to Spanish specifications [34]. 5. Phase 4: Trial Sections Structural Design In this section, the design of three trial sections based on the results obtained in laboratory tests are shown. Three alternative sections (AS) were designed to reduce the thickness of the control section analysed by Rosales et al. [32], while the mechanical and durability properties of the current section were maintained or improved. To guarantee the adequate structural performance of the alternative sections, a calcula- tion process was carried out using Everstress 5.0 software. The maximum load capacity of the control section, measured in the maximum number of standard axles of 13-tonne-heavy vehicles, was determined and compared with the alternative sections, which must present a value equal to or greater than the control section. To determine the maximum number of equivalent single axle loads of 13 tonnes, the following methodology was followed: (1) Determination of the elastic modulus (E) of the layers from the CBR index or com- pressive strength according to the type of material and its mechanical behaviour. (2) Calculation of the vertical deformations in the subgrade using Everstress 5.0 software. Appl. Sci. 2021, 11, 9780 11 of 20 (3) Calculation of the number of equivalent axles according to the fatigue law described by the Spanish specifications [40] according to Expression (1). 0.28 # = 2.16E 2  N (1) # = unit vertical deformation in the subgrade. N = number of equivalent axles of 13 tonnes. In the first place, the calculation of materials with a reduced CBR, such as soil from the construction site or stabilised or granular materials that do not present significant compressive strength, was accomplished by the Transport and Road Research Laboratory method [41], which is in accordance with the study of different bibliographies that indicate that the application of this method is appropriate for materials with CBR of less than 10% and without unconfined compressive strength. The formula of Powell et al. [41] for stabilised materials and unbound granular mate- rials was applied according to Expression (2). 0.64 E(MPa) = 17.6 CBR (2) Furthermore, with soils in which the CBR results were increased, a compressive strength test was carried out beforehand. The elastic modulus of these soils that present CBR values greater than 20% and a compressive strength greater than 0.2 MPa was cal- culated by the Molenaar equation. This equation considers the unconfined compressive strength as the main modulus calculation parameter. The formula of Molenaar [42] for materials treated with binders was applied according to Expression (3). E (MPa) = 1435 [UCS]ˆ0.885 (3) Table 5 shows the elastic modulus of the analysed mixtures according to the expression shown above. Table 5. Elastic modulus, E (MPa), of the soils or mixtures. Design Parameter Elastic Modulus (MPa) by Soil or Mixture Calculation CBR (%) UCS (MPa) CG * 60.05 - 242 CS + 1.5%QL 12.3 - 88 CS + 1.5%QL + N1 17.95 - 112 CS + 1.5%QL + N2&N3 18.4 114 AM-1 - 1.08 1536 AM-2 - 1.26 1761 AM-3 - 1.13 1599 * Obtained from Rosales et al. [32]. Once the elastic modulus of the materials that make up each layer were determined, they were entered into the software, together with the thicknesses of each layer and with a stress of 800 kPa, in accordance with Spanish regulations. First, the conventional section was analysed, which presented a bearing capacity of approximately 75,000 equivalent axels. Subsequently, three series of alternative sections (AS): AS-1, AS-2 and AS-3 of 45 cm, 50 cm and 55 cm, maintaining the proportion of 40% CS and 60% SS of the design mixes, were analysed. Table 6 shows the results obtained for the vertical deformation of the subgrade and the number of equivalent axels for each solution. Appl. Sci. 2021, 11, 9780 12 of 20 Table 6. Results of the vertical deformations in the subgrade obtained in the multilayers analysis and the number of axles obtained by calculus. Vertical Deformations in the Number of Equivalent Axles Section Subgrade (" ) (10 ) of 13 Tonnes Conventional Solution 932.96 74,737.32 AS-1-45 cm 1138.72 36,678.91 AS-2-45 cm 1049.18 49,140.69 AS-3-45 cm 1095.34 42,136.58 AS-1-50 cm 880.26 91,986.36 AS-2-50 cm 810.96 123,284,21 AS-3-50 cm 850.40 104,052.13 AS-1-55 cm 750.50 162,585.39 AS-2-55 cm 645.06 279,193.29 AS-3-55 cm 678.72 232,814.61 As is observed in Table 6, the 45-cm section series structural capacity was insufficient and was therefore rejected. The 55-cm series exceeded by three times the required capacity of 75,000 equivalent axels and was therefore rejected in order not to oversize the section. Appl. Sci. 2021, 11, x FOR PEER REVIEW 13 of 21 Therefore, the thickness of the section was 50 cm, due to a higher structural capacity than the control solution and a reduction in the section thickness of 30 cm. 6. 6. Phase Phase 5:5: Construc Construction tion and and S Section ection In In Situ Situ TT eests sts 6.1. Trials Sections Execution 6.1. Trials Sections Execution After the design phase, the three trial sections of 50-cm total thicknesses were con- After the design phase, the three trial sections of 50-cm total thicknesses were con- structed, with a length of 100 m for alternative sections 1 and 2 (AS-1 and AS-2) and a structed, with a length of 100 m for alternative sections 1 and 2 (AS-1 and AS-2) and a length of 50 m for alternative section 3 (AS-3). length of 50 m for alternative section 3 (AS-3). The three trial sections were built near the newly built road in Villacarrillo, Jaén, The three trial sections were built near the newly built road in Villacarrillo, Jaén, Spain, [32] and the performance of the AS was compared with the control section, which Spain, [32] and the performance of the AS was compared with the control section, which was used alongside the layout of the road. Figure 9 shows the control section and the was used alongside the layout of the road. Figure 9 shows the control section and the location of the trial section. location of the trial section. Figure 9. General location of the trial sections. Figure 9. General location of the trial sections. The final solutions were as follows: - Alternative section 1 (AS-1). Thirty centimetres of CS (40%) stabilised with 1.5% lime, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out accord- ing to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% quicklime. - Alternative section 2 (AS-2). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.056% N1, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out according to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% lime and 0.056% N1. - Alternative section 3 (AS-3). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.12% N2&N3, and SS (60%) stabilised with 1% CEM-II on the total soil (CS+SS) made according to the process described in Figure 1 on a 20-cm layer of CS stabilised with 1.5% lime and 0.12% N2&N3. Figure 10 shows a scheme of the developed sections. Appl. Sci. 2021, 11, 9780 13 of 20 The final solutions were as follows: - Alternative section 1 (AS-1). Thirty centimetres of CS (40%) stabilised with 1.5% lime, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out according to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% quicklime. - Alternative section 2 (AS-2). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.056% N1, and SS (60%) stabilised with 1% CEM-II over the total soil (CS+SS) carried out according to the process described in Figure 1 over a 20-cm layer of CS stabilised with 1.5% lime and 0.056% N1. - Alternative section 3 (AS-3). Thirty centimetres of CS (40%) stabilised with 1.5% lime and 0.12% N2&N3, and SS (60%) stabilised with 1% CEM-II on the total soil (CS+SS) made according to the process described in Figure 1 on a 20-cm layer of CS stabilised Appl. Sci. 2021, 11, x FOR PEER REVIEW 14 of 21 with 1.5% lime and 0.12% N2&N3. Figure 10 shows a scheme of the developed sections. Figure 10. Trial sections scheme. Figure 10. Trial sections scheme. 6.2. Trial Sections Survey 6.2. Trial Sections Survey According to Spanish specifications [33], three categories of subgrades, according to According to Spanish specifications [33], three categories of subgrades, according to the equivalent elastic modulus, are stabilised: low quality, 60–120 MPa; medium quality, the equivalent elastic modulus, are stabilised: low quality, 60–120 MPa; medium quality, 120–300 MPa and high quality, greater than 300 MPa. 120–300 MPa and high quality, greater than 300 MPa. For rural and low-volume roads, a low-quality subgrade with a modulus between 60 For rural and low-volume roads, a low-quality subgrade with a modulus between 60 and 120 MPa is structurally valid; however, due to the control section being designed as and 120 MPa is structurally valid; however, due to the control section being designed as a medium-quality subgrade, this criterion was maintained for the test sections, setting a a medium-quality subgrade, this criterion was maintained for the test sections, setting a minimum of a 120-MPa equivalent modulus. minimum of a 120-MPa equivalent modulus. To determine and compare the equivalent modulus value, the following methods To determine and compare the equivalent modulus value, the following methods were analysed. were analysed. Analysis of the theoretical deflection produced by a 500-kPa applied on a 300-mm- Analysis of the theoretical deflection produced by a 500-kPa applied on a 300-mm- diameter plate in a multilayer elastic model, the method specified in Spanish regula- diameter plate in a multilayer elastic model, the method specified in Spanish regulations tions [40]. The theoretical equivalent modulus value per section was obtained. [40]. The theoretical equivalent modulus value per section was obtained. Analysis of the deflections measured in the falling weight deflectometer (FWD) test. Analysis of the deflections measured in the falling weight deflectometer (FWD) test. The average equivalent modulus value of the section was obtained. The average equivalent modulus value of the section was obtained. Analysis of the second load cycle in the plate bearing test. A point value of equivalent Analysis of the second load cycle in the plate bearing test. A point value of equivalent modulus per section was obtained. modulus per section was obtained. Table 7 shows the results obtained in the Everstress 5.0 program, analysing the mul- tilayer elastic model with a stress of 500 kPa applied on a 300-mm-diameter plate. Apply- ing Formula (4), the equivalent modulus of compressibility of each section, Ev, was deter- mined. (MPa) = 13.150/ (mm/100)[40] (4) Table 7. Theorical deflection and modulus of compressibility obtained in Everstress 5.0. Everstress Results AS-1 AS-2 AS-3 Theorical deflection (mm/100) 78.02 73.51 75.63 Theorical modulus of compressibility, Ev (MPa) 168.55 178.89 173.87 As can be observed in Table 7, the three alternative sections present a sufficient the- oretical equivalent modulus to be considered a medium-quality subgrade. Appl. Sci. 2021, 11, 9780 14 of 20 Table 7 shows the results obtained in the Everstress 5.0 program, analysing the mul- tilayer elastic model with a stress of 500 kPa applied on a 300-mm-diameter plate. Ap- plying Formula (4) [40], the equivalent modulus of compressibility of each section, E , was determined. E (MPa) = 13.150/d (mm/100) (4) v 0 Table 7. Theorical deflection and modulus of compressibility obtained in Everstress 5.0. Everstress Results AS-1 AS-2 AS-3 Theorical deflection (mm/100) 78.02 73.51 75.63 Theorical modulus of compressibility, E (MPa) 168.55 178.89 173.87 Appl. Sci. 2021, 11, x FOR PEER REVIEW 15 of 21 As can be observed in Table 7, the three alternative sections present a sufficient theoretical equivalent modulus to be considered a medium-quality subgrade. Once the theoretical equivalent modulus values for each section have been obtained, Once the theoretical equivalent modulus values for each section have been obtained, they are verified with the values obtained in the in situ tests, FDW and plate bearing test. they are verified with the values obtained in the in situ tests, FDW and plate bearing test. As Figure 11 shows, the climatology of the area shows two distinct annual seasons, As Figure 11 shows, the climatology of the area shows two distinct annual seasons, with 6 months with abundant rainfall and 6 dry months [43]. Therefore, two in situ test with 6 months with abundant rainfall and 6 dry months [43]. Therefore, two in situ test campaigns were carried out: the first in July 2020 in the dry season and the second in campaigns were carried out: the first in July 2020 in the dry season and the second in March 2021 in the wet season. March 2021 in the wet season. Figure 11. Annual rainfall distribution in Villacarrillo, Spain. Figure 11. Annual rainfall distribution in Villacarrillo, Spain. Once the data for both periods was obtained, the average annual value of equivalent Once the data for both periods was obtained, the average annual value of equivalent modulus was obtained for each of the methods. modulus was obtained for each of the methods. 6.2.1. Deflection Measurements by FWD 6.2.1. Deflection Measurements by FWD The deflection measurement enables characterising the structural capacity of the The deflection measurement enables characterising the structural capacity of the formed subgrade, as well as its layers along the road layout. A Dynatest HDW 8081 falling formed subgrade, as well as its layers along the road layout. A Dynatest HDW 8081 falling weight deflectometer (FWD) was used. weight deflectometer (FWD) was used. A pressure of 850 kPa was applied through a 300-mm-diameter plate. The surface A pressure of 850 kPa was applied through a 300-mm-diameter plate. The surface deformation due to the application of this load was measured by seven geophones located deformation due to the application of this load was measured by seven geophones located at 0, 300, 450, 600, 900, 1200 and 1500 mm from the application of the load. at 0, 300, 450, 600, 900, 1200 and 1500 mm from the application of the load. Surface deformation measurements were taken every 10 m in sections AS-1 and AS-2 Surface deformation measurements were taken every 10 m in sections AS-1 and AS- and every 5 m in section AS-3 in both lanes, obtaining the average per section. 2 and every 5 m in section AS-3 in both lanes, obtaining the average per section. Figures 12–14 show the mean deflection values measured in the dry and wet seasons Figures 12–14 show the mean deflection values measured in the dry and wet seasons for each section. for each section. Figure 12. Deflection measurement in AS-1. Appl. Sci. 2021, 11, x FOR PEER REVIEW 15 of 21 Once the theoretical equivalent modulus values for each section have been obtained, they are verified with the values obtained in the in situ tests, FDW and plate bearing test. As Figure 11 shows, the climatology of the area shows two distinct annual seasons, with 6 months with abundant rainfall and 6 dry months [43]. Therefore, two in situ test campaigns were carried out: the first in July 2020 in the dry season and the second in March 2021 in the wet season. Figure 11. Annual rainfall distribution in Villacarrillo, Spain. Once the data for both periods was obtained, the average annual value of equivalent modulus was obtained for each of the methods. 6.2.1. Deflection Measurements by FWD The deflection measurement enables characterising the structural capacity of the formed subgrade, as well as its layers along the road layout. A Dynatest HDW 8081 falling weight deflectometer (FWD) was used. A pressure of 850 kPa was applied through a 300-mm-diameter plate. The surface deformation due to the application of this load was measured by seven geophones located at 0, 300, 450, 600, 900, 1200 and 1500 mm from the application of the load. Surface deformation measurements were taken every 10 m in sections AS-1 and AS- 2 and every 5 m in section AS-3 in both lanes, obtaining the average per section. Appl. Sci. 2021, 11, 9780 15 of 20 Figures 12–14 show the mean deflection values measured in the dry and wet seasons for each section. Appl. Sci. 2021, 11, x FOR PEER REVIEW 16 of 21 Appl. Sci. 2021, 11, x FOR PEER REVIEW 16 of 21 Figure 12. Deflection measurement in AS-1. Figure 12. Deflection measurement in AS-1. Figure 13. Deflection measurement in AS-2. Figure 13. Deflection measurement in AS-2. Figure 13. Deflection measurement in AS-2. Figure 14. Deflection measurement in AS-3. Figure 14. Deflection measurement in AS-3. Figure 14. Deflection measurement in AS-3. Analysing Figures 12–14, two patterns of behaviour were observed: firstly, the addi- Analysing Analysing Figures Figures 1 12 2–1 –14 4, t , two wo pa patterns tterns of of be behaviour haviour we wer re e obs observed: erved: firfirstly stly, the a , thed ad- di- tion of nanomaterials, both sodium silicates and organosilanes, together with acrylic co- dition tion of na of nanomaterials, nomaterials, both sodi both sodium um silica silicates tes and organosilanes, to and organosilanes,ge together ther with with acryli acrylic c co- polymers, improved the bearing capacity of the soil, reducing the deflections measured copolymers, polymers, im impr proved t ovedh the e bear bearing ing ca capacity pacity of ofthe soil, reduc the soil, reducing ing the the deflections deflections measur measured ed during the test. during the test. during the test. Secondly, the sections with nanomaterials showed a drop in the bearing capacity due Secondly, the sections with nanomaterials showed a drop in the bearing capacity due to rainfall, in contrast to the AS-1 section, which remained practically constant after a to rainfall, in contrast to the AS-1 section, which remained practically constant after a year’s weather. year’s weather. However, the AS-1 section showed greater dispersion in the data, which could be However, the AS-1 section showed greater dispersion in the data, which could be attributed to problems during the execution of the work, presenting over-compacted areas attributed to problems during the execution of the work, presenting over-compacted areas that reduce the average deflection of the section. that reduce the average deflection of the section. Table 8 shows the average annual deflection of each section, as well as the weighted Table 8 shows the average annual deflection of each section, as well as the weighted deflection for a load of 500 kPa, which enables calculating the elastic modulus through deflection for a load of 500 kPa, which enables calculating the elastic modulus through Equation (4). Equation (4). Table 8. Modulus of the compressibility mean in each section. Table 8. Modulus of the compressibility mean in each section. Anual Average Deflection Weighted Deflection Anual Average Deflection Weighted Deflection Trial Section Ev (MPa) Trial Section Ev (MPa) (0.01 mm) (0.01 mm) (0.01 mm) (0.01 mm) AS-1 183 108 122 AS-1 183 108 122 AS-2 168 99 133 AS-2 168 99 133 AS-3 159 93 142 AS-3 159 93 142 Appl. Sci. 2021, 11, 9780 16 of 20 Secondly, the sections with nanomaterials showed a drop in the bearing capacity due to rainfall, in contrast to the AS-1 section, which remained practically constant after a year ’s weather. However, the AS-1 section showed greater dispersion in the data, which could be attributed to problems during the execution of the work, presenting over-compacted areas that reduce the average deflection of the section. Table 8 shows the average annual deflection of each section, as well as the weighted deflection for a load of 500 kPa, which enables calculating the elastic modulus through Equation (4). Table 8. Modulus of the compressibility mean in each section. Anual Average Weighted Deflection Trial Section E (MPa) Deflection (0.01 mm) (0.01 mm) AS-1 183 108 122 AS-2 168 99 133 AS-3 159 93 142 As is shown in Table 8, the average values of the sections show a reduction in the average annual deflections and an increase in the equivalent modulus of the sections made with nanomaterials. Comparing the deflection results obtained for the new hybrid stabilised sections, a bet- ter behaviour was observed than the alternative section analysed by Rosales et al. [32], and a slight drop was observed when compared to the 80-cm-thickness control section. However, a section reduction of approximately 40% and a similar mechanical behaviour confirmed the suitability of the proposed method of hybrid stabilised solutions for road layers. All the sections can be classified as in the medium category, although with lower values than those obtained theoretically, due to the irregularities of the terrain and construction peculiarities of the linear works. 6.2.2. Plate-Bearing Test The plate loading test is an in situ test used to measure the final bearing capacity of the subgrade built from the sections designed through its compressibility modulus. The plate bearing test was performed according to UNE 103808:2006, and it consists of measuring the settlement of a rigid circular plate resting on the ground, subjected to different loads in a staggered manner, called the load cycle. This circular-shaped plate has a surface area of 700 cm (diameter 298.5 mm), and the measurements enable determining the compressibility modulus in the first load cycle (E ) and in the second load cycle (E ). v 1 v 2 The plate bearing test enables the calculation of the punctual behaviour of the section in the place where the load was applied, so its result was not as representative as those obtained in FDW; however, it enables verifying the behaviour of the subgrade and analysing its evolution over time. The results of the plate loading test are shown in Figures 15–17. As is shown in Figure 15, all the sections presented a medium quality, which verified the adequate structural behaviour in relation to the volume of the loads the sections will support during their useful life. An improvement in the quality of subgrades AS-2 and AS-3 was observed due to the addition of nanomaterials, similar to that observed in the FDW results. Analysing Figure 16, a drop in the equivalent modulus was observed in all three sections due to the rainy period; however, this drop was greater in sections AS-2 and AS-3 than AS-1. This sharp drop may be due to a lack of bearing capacity in some of the layers that make up the section because of the rain or perhaps a specific problem associated with the uncertainty of the plate loading test. Appl. Sci. 2021, 11, x FOR PEER REVIEW 17 of 21 As is shown in Table 8, the average values of the sections show a reduction in the average annual deflections and an increase in the equivalent modulus of the sections made with nanomaterials. Comparing the deflection results obtained for the new hybrid stabilised sections, a better behaviour was observed than the alternative section analysed by Rosales et al. [32], and a slight drop was observed when compared to the 80-cm-thickness control section. However, a section reduction of approximately 40% and a similar mechanical behaviour confirmed the suitability of the proposed method of hybrid stabilised solutions for road layers. All the sections can be classified as in the medium category, although with lower values than those obtained theoretically, due to the irregularities of the terrain and con- struction peculiarities of the linear works. Appl. Sci. 2021, 11, 9780 17 of 20 6.2.2. Plate-Bearing Test The plate loading test is an in situ test used to measure the final bearing capacity of Finally, Figure 17 shows the average annual behaviour of the sections. Despite the large the subgrade built from the sections designed through its compressibility modulus. drop observed in the rainy weather for sections AS-2 and AS-3, an adequate mechanical The plate bearing test was performed according to UNE 103808:2006, and it consists behaviour was observed, being practically identical to that obtained in the theoretical of measuring the settlement of a rigid circular plate resting on the ground, subjected to analysis using the multilayer elastic model, which confirms the validity of the multilayer different loads in a staggered manner, called the load cycle. This circular-shaped plate has elastic method used and the validity of the sections constructed. a surface area of 700 cm (diameter 298.5 mm), and the measurements enable determining Comparing the results with those obtained by Rosales et al. [32], an increase in the the compressibility modulus in the first load cycle (Ev1) and in the second load cycle (Ev2). modulus of the compressibility of the current hybrid sections was observed. This increase The plate bearing test enables the calculation of the punctual behaviour of the section confirmed the adequate behaviour of the new solutions constructed with a reduction of the in the place where the load was applied, so its result was not as representative as those input materials. obtained in FDW; however, it enables verifying the behaviour of the subgrade and ana- Additionally, the evolution of the E /E ratio was shown, limited by Spanish v2 v1 lysing its evolution over time. specifications to 2.2, a limitation that was met by all sections in both the dry and wet seasons. The results of the plate loading test are shown in Figures 15–17. Appl. Sci. 2021, 11, x FOR PEER REVIEW 18 of 21 Figure Figure 15. 15. Plat Plate e bearing bearing test test rr eesults sults in the in thedr dry y season. season. Figure 16. Plate bearing test results in the wet season. Figure 16. Plate bearing test results in the wet season. Figure 17. Annual average plate bearing test results. As is shown in Figure 15, all the sections presented a medium quality, which verified the adequate structural behaviour in relation to the volume of the loads the sections will support during their useful life. An improvement in the quality of subgrades AS-2 and AS-3 was observed due to the addition of nanomaterials, similar to that observed in the FDW results. Analysing Figure 16, a drop in the equivalent modulus was observed in all three sec- tions due to the rainy period; however, this drop was greater in sections AS-2 and AS-3 than AS-1. This sharp drop may be due to a lack of bearing capacity in some of the layers that make up the section because of the rain or perhaps a specific problem associated with the uncertainty of the plate loading test. Finally, Figure 17 shows the average annual behaviour of the sections. Despite the large drop observed in the rainy weather for sections AS-2 and AS-3, an adequate me- chanical behaviour was observed, being practically identical to that obtained in the theo- retical analysis using the multilayer elastic model, which confirms the validity of the mul- tilayer elastic method used and the validity of the sections constructed. Appl. Sci. 2021, 11, x FOR PEER REVIEW 18 of 21 Appl. Sci. 2021, 11, 9780 18 of 20 Figure 16. Plate bearing test results in the wet season. Figure 17. Annual average plate bearing test results. Figure 17. Annual average plate bearing test results. 7. Conclusions As is shown in Figure 15, all the sections presented a medium quality, which verified In this work, the construction of trial sections based on hybrid stabilised soils with the adequate structural behaviour in relation to the volume of the loads the sections will small amounts of lime, cement and nanomaterials was studied. For this purpose, an in- support during their useful life. depth laboratory study was carried out to determine the physicochemical properties of An improvement in the quality of subgrades AS-2 and AS-3 was observed due to the the materials and the mechanical behaviours of the mixtures of stabilised soil. A structural addition of nanomaterials, similar to that observed in the FDW results. design of the trial sections was carried out through a multilayer elastic model. Finally, the Analysing Figure 16, a drop in the equivalent modulus was observed in all three sec- sections were built in a real-scale application, and their short and medium-term in situ tions due to the rainy period; however, this drop was greater in sections AS-2 and AS-3 properties were monitored. than AS-1. The following conclusions have been drawn from this study: This sharp drop may be due to a lack of bearing capacity in some of the layers that The stabilisation of soils with nanosilica slightly increases the maximum dry density of make up the section because of the rain or perhaps a specific problem associated with the the samples and slightly reduces their optimum water content. A reduction in the optimum uncertainty of the plate loading test. water content implies a reduction in the consumption of natural resources (water, fuel, less Finally, Figure 17 shows the average annual behaviour of the sections. Despite the CO emissions, etc.) to achieve the same or higher degree of compaction. large drop observed in the rainy weather for sections AS-2 and AS-3, an adequate me- The addition of nanosilica improves the bearing capacity of the stabilised soils as mea- chanical behaviour was observed, being practically identical to that obtained in the theo- sured by the CBR index. The nanomaterial N1, composed of sodium silicate, shows a higher retical analysis using the multilayer elastic model, which confirms the validity of the mul- reactivity in combination with cement, increasing the unconfined compressive strength. tilayer elastic method used and the validity of the sections constructed. The proposed hybrid stabilisation solution reduces the thickness of the control section by 37.5% and increases the bearing capacity measured in the number of equivalent axles by up to 25%. The sections built with N1 and N2&N3 show an improvement in the annual mean equivalent modulus of compressibility, with these sections showing a greater drop in bearing capacity during the rainy season. The results support the application of the developed alternative solutions with a hybrid stabilisation process due to their increased bearing capacity for rural and low- volume roads. However, due to the typical climatic conditions in Southern Spain, there is no evidence of effectiveness in the other conditions. In countries with high humidity and large temperature differences, the solution should be verified. As a general conclusion, the use of nanomaterials in percentages between 0.06 and 0.12% improves the mechanical behaviour of stabilised soils and allows a reduction in the thickness of the road layer, improving its general structural capacity, although there is a drop in capacity during rainy periods, which will be the subject of future studies. Author Contributions: Conceptualization, F.A. and J.R.M.; methodology, J.R. and F.A.; investigation, J.L.D.-L. and M.C.; resources, F.A. and J.R.M.; data curation, J.L.D.-L. and J.R.; writing—original draft preparation, J.L.D.-L. and M.C.; writing—review and editing, M.C., J.R. and J.L.D.-L. and supervision, F.A. All authors have read and agreed to the published version of the manuscript. Appl. Sci. 2021, 11, 9780 19 of 20 Funding: This work was financed by the project “SLAKED-LIME REDUCTION ON EXPANSIVE SOILS BY MEANS OF NANOMATERIALS AND THE REUTILIZATION OF WASTED MATERIAL AND STABILIZING BY-PRODUCTS—ECARYSE. Ref. RTC-2017-62025”, granted by the call for “FEDER/Ministerio de Educacion Ciencia y Universidades-Agencia Estatal de Investigación”. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Acknowledgments: Call for Sub-modality 2.2. Pre-doctoral contracts UCO of the University of Cordoba’s Research Plan 2020. Conflicts of Interest: The authors declare no conflict of interest. References 1. Ali, Y.; Socci, C.; Pretaroli, R.; Severini, F. Economic and environmental impact of transport sector on Europe economy. Asia-Pac. J. Reg. Sci. 2018, 2, 361–397. [CrossRef] 2. Asomani-Boateng, R.; Fricano, R.J.; Adarkwa, F. Assessing the socio-economic impacts of rural road improvements in Ghana: A case study of transport sector program support (II). Case Stud. Transp. Policy 2015, 3, 355–366. [CrossRef] 3. Chamorro, A.; Tighe, S. Development and Application of a Sustainable Management System for Unpaved Rural Road Networks. Transp. Res. Rec. 2019, 2673, 891–901. [CrossRef] 4. Jiménez, J.; Ayuso, J.; Galvín, A.; López, M.; Agrela, F. Use of mixed recycled aggregates with a low embodied energy from non-selected CDW in unpaved rural roads. Constr. Build. Mater. 2012, 34, 34–43. [CrossRef] 5. Sirivitmaitrie, C.; Puppala, A.J.; Saride, S.; Hoyos, L. Combined lime–cement stabilization for longer life of low-volume roads. Transp. Res. Rec. 2011, 2204, 140–147. [CrossRef] 6. Lim, S.; Wijeyesekera, D.; Lim, A.; Bakar, I. Critical review of innovative soil road stabilization techniques. Int. J. Eng. Adv. Technol. 2014, 3, 2249–8958. 7. Firoozi, A.A.; Olgun, C.G.; Firoozi, A.A.; Baghini, M.S. Fundamentals of soil stabilization. Int. J. Geo-Eng. 2017, 8, 26. [CrossRef] 8. Afrin, H. A review on different types soil stabilization techniques. Int. J. Transp. Eng. Technol. 2017, 3, 19–24. [CrossRef] 9. Horpibulsuk, S.; Rachan, R.; Chinkulkijniwat, A.; Raksachon, Y.; Suddeepong, A. Analysis of strength development in cement- stabilized silty clay from microstructural considerations. Constr. Build. Mater. 2010, 24, 2011–2021. [CrossRef] 10. Khemissa, M.; Mahamedi, A. Cement and lime mixture stabilization of an expansive overconsolidated clay. Appl. Clay Sci. 2014, 95, 104–110. [CrossRef] 11. Olinic, T.; Olinic, E. The effect of quicklime stabilization on soil properties. Agric. Agric. Sci. Procedia 2016, 10, 444–451. [CrossRef] 12. Prusinski, J.R.; Bhattacharja, S. Effectiveness of Portland cement and lime in stabilizing clay soils. Transp. Res. Rec. 1999, 1652, 215–227. [CrossRef] 13. Costa, C. Hydraulic binders. In Materials for Construction and Civil Engineering; Springer: Cham, Switzerland, 2015; pp. 1–52. 14. Barišic, ´ I.; Netinger Grubeša, I.; Dokšanovic, ´ T.; Markovic, ´ B. Feasibility of agricultural biomass fly ash usage for soil stabilisation of road works. Materials 2019, 12, 1375. [CrossRef] [PubMed] 15. Cabrera, M.; Rosales, J.; Ayuso, J.; Estaire, J.; Agrela, F. Feasibility of using olive biomass bottom ash in the sub-bases of roads and rural paths. Constr. Build. Mater. 2018, 181, 266–275. [CrossRef] 16. Galvín, A.P.; López-Uceda, A.; Cabrera, M.; Rosales, J.; Ayuso, J. Stabilization of expansive soils with biomass bottom ashes for an eco-efficient construction. Environ. Sci. Pollut. Res. 2021, 28, 24441–24454. [CrossRef] 17. Vichan, S.; Rachan, R. Chemical stabilization of soft Bangkok clay using the blend of calcium carbide residue and biomass ash. Soils Found. 2013, 53, 272–281. [CrossRef] 18. Amrani, M.; Taha, Y.; Kchikach, A.; Benzaazoua, M.; Hakkou, R. Phosphogypsum recycling: New horizons for a more sustainable road material application. J. Build. Eng. 2020, 30, 101267. [CrossRef] 19. Zeng, L.-L.; Bian, X.; Zhao, L.; Wang, Y.-J.; Hong, Z.-S. Effect of phosphogypsum on physiochemical and mechanical behaviour of cement stabilized dredged soil from Fuzhou, China. Geomech. Energy Environ. 2021, 25, 100195. [CrossRef] 20. Manso, J.M.; Ortega-López, V.; Polanco, J.A.; Setién, J. The use of ladle furnace slag in soil stabilization. Constr. Build. Mater. 2013, 40, 126–134. [CrossRef] 21. Thomas, A.; Tripathi, R.; Yadu, L. A laboratory investigation of soil stabilization using enzyme and alkali-activated ground granulated blast-furnace slag. Arab. J. Sci. Eng. 2018, 43, 5193–5202. [CrossRef] 22. Wang, D.; Du, Y.; Xiao, J. Shear properties of stabilized loess using novel reactive magnesia-bearing binders. J. Mater. Civ. Eng. 2019, 31, 04019039. [CrossRef] 23. Gu, K.; Jin, F.; Al-Tabbaa, A.; Shi, B.; Liu, C.; Gao, L. Incorporation of reactive magnesia and quicklime in sustainable binders for soil stabilisation. Eng. Geol. 2015, 195, 53–62. [CrossRef] 24. Krishnan, J.; Shukla, S. The behaviour of soil stabilised with nanoparticles: An extensive review of the present status and its applications. Arab. J. Geosci. 2019, 12, 436. [CrossRef] 25. Ghasabkolaei, N.; Choobbasti, A.J.; Roshan, N.; Ghasemi, S.E. Geotechnical properties of the soils modified with nanomaterials: A comprehensive review. Arch. Civ. Mech. Eng. 2017, 17, 639–650. [CrossRef] Appl. Sci. 2021, 11, 9780 20 of 20 26. Kawashima, S.; Hou, P.; Corr, D.J.; Shah, S.P. Modification of cement-based materials with nanoparticles. Cem. Concr. Compos. 2013, 36, 8–15. [CrossRef] 27. Kulanthaivel, P.; Soundara, B.; Velmurugan, S.; Naveenraj, V. Experimental investigation on stabilization of clay soil using nano-materials and white cement. Mater. Today Proc. 2021, 45, 507–511. [CrossRef] 28. Choobbasti, A.J.; Kutanaei, S.S. Microstructure characteristics of cement-stabilized sandy soil using nanosilica. J. Rock Mech. Geotech. Eng. 2017, 9, 981–988. [CrossRef] 29. Meeravali, K.; Ruben, N.; Rangaswamy, K. Stabilization of soft-clay using nanomaterial: Terrasil. Mater. Today Proc. 2020, 27, 1030–1037. [CrossRef] 30. Ghasabkolaei, N.; Janalizadeh, A.; Jahanshahi, M.; Roshan, N.; Ghasemi, S.E. Physical and geotechnical properties of cement- treated clayey soil using silica nanoparticles: An experimental study. Eur. Phys. J. Plus 2016, 131, 134. [CrossRef] 31. Bahmani, S.H.; Farzadnia, N.; Asadi, A.; Huat, B.B. The effect of size and replacement content of nanosilica on strength development of cement treated residual soil. Constr. Build. Mater. 2016, 118, 294–306. [CrossRef] 32. Rosales, J.; Agrela, F.; Marcobal, J.R.; Diaz-López, J.L.; Cuenca-Moyano, G.M.; Caballero, Á.; Cabrera, M. Use of Nanomaterials in the Stabilization of Expansive Soils into a Road Real-Scale Application. Materials 2020, 13, 3058. [CrossRef] [PubMed] 33. de Fomento, E. Instrucción de carreteras. Norma 6.1 IC: Secciones de firme. Norm. Instr. Construcción 2003, 41. 34. de Fomento, M. PG-3: Pliego de Prescripciones Técnicas Generales Para Obras de Carreteras y Puentes; Ediciones Liteam SL: Madrid, Spain, 2002. 35. Zhu, X.B.; Wang, H.; Zhang, Y.C. Evaluation of subgrade compactness. In Applied Mechanics and Materials; Trans Tech Publications Ltd.: Bäch, Switzerland, 2013; pp. 1663–1667. 36. Poon, C.S.; Chan, D. Feasible use of recycled concrete aggregates and crushed clay brick as unbound road sub-base. Constr. Build. Mater. 2006, 20, 578–585. [CrossRef] 37. Alireza, S.G.S.; Mohammad, M.S.; Hasan, B.M. Application of nanomaterial to stabilize a weak soil. In Proceedings of the 7th International Conference on Case Histories in Geotechnical Engineering, Chicago, IL, USA, 29 April–4 May 2013. 38. Mousavi, F.; Abdi, E.; Rahimi, H. Effect of polymer stabilizer on swelling potential and CBR of forest road material. KSCE J. Civ. Eng. 2014, 18, 2064–2071. [CrossRef] 39. Haeri, S.M.; Hosseini, A.M.; Shahrabi, M.M.; Soleymani, S. Comparison of strength characteristics of Gorgan loessial soil improved by nano-silica, lime and Portland cement. In From Fundamentals to Applications in Geotechnics; IOS Press: Amsterdam, The Netherlands, 2015; pp. 1820–1827. 40. Dutor, A.-B.; Castilla-Molina, J.; Gómez-Casado, J.-A. Instrucción para el diseño de firmes de la red de carreteras de Andalucía; ICAFIR: Sevilla, Spain, 2007. 41. Powell, W.; Potter, J.; Mayhew, H.; Nunn, M. The Structural Design of Bituminous Roads; Transport and Road Research Laboratory (TRRL): Crowthorne, UK, 1984. 42. Xuan, D.; Houben, L.; Molenaar, A.; Shui, Z. Mechanical properties of cement-treated aggregate material—A review. Mater. Des. 2012, 33, 496–502. [CrossRef] 43. de Andalucía, J. Instituto De Investigación Y Formación Agraria Y Pesquera (IFAPA). Available online: https://www. juntadeandalucia.es/agriculturaypesca/ifapa/riaweb/web/estacion/23/102 (accessed on 21 May 2021).

Journal

Applied SciencesMultidisciplinary Digital Publishing Institute

Published: Oct 20, 2021

Keywords: hybrid stabilisation; mechanical behaviour; real scale application; nanomaterials

There are no references for this article.