Evaluating the usability of recycled aggregates as fill materials depending on the composition and strength of their grains

Abstract

While recycling is a topic of contemporary relevance, there is a scarcity of research on the engineering characteristics of construction and demolition wastes with different levels of grain strength and composition of debris, which impose constraints on their potential for reuse. This study aims to increase the use of recycled aggregates in fillings, addressing a gap in the literature. For this purpose, large-scale direct shear and California bearing ratio tests were conducted on nine diverse recycled aggregates from different construction works. The test outcomes were compared to those obtained from natural aggregates (NA) to draw a meaningful conclusion. The impact of the specimens’ water content and relative density on the findings was discussed. Results demonstrated that the shear strength of recycled aggregates is significantly affected by the compressive strength of the concrete within the recycled aggregates. Besides, increasing the percentage of NA or relative density improved the specimen’s shear strength. On the other hand, it was determined that the high water content of the crushed bricks reduced the fill’s quality. As a result of the study, equations were suggested for use in filling design.

Graphical abstract

Keywords

recycled aggregates / fillings / sustainability / granular soils / large-scale direct shear test

1 Introduction

The Industrial Revolution has increased the human population improperly worldwide, and to meet the needs of population growth, increasing construction work has accelerated the consumption of natural resources. This circumstance has caused an increase in global warming and environmental pollution. Therefore, sustainable use of natural resources in the construction industry has become a topic of interest. One of the most commonly used natural resources in the construction industry is natural (virgin) aggregates (NA) originating from quarries. NA are extensively utilized in the construction industry for various applications, such as concrete, mortar, and road infrastructure. Nevertheless, the extraction of materials from quarries has the potential to cause adverse environmental impacts. Moreover, the processes used to obtain NA are expensive, and the energy used in such processes contributes to global warming [1]. Furthermore, NA are not limitless and may run out or become more expensive in the near future. For these reasons, the sustainable use of NA is significant for reducing costs, energy consumption, and global warming, as well as for the conservation of the environment [2]. Accordingly, societies and governments have strongly supported all efforts to increase the usability of recycled aggregates rather than NA [3–5].

Construction works such as renovation, repair, or demolition carried out on the buildings generate wastes known as debris, also called construction and demolition wastes (CDW). The CDW may include different types of wastes like concrete, brick, tile, ceramic, wood, glass, plastic, bituminous mixtures, soil, metals, etc. [4]. Many researchers agreed that these wastes have the potential to be used as filling material due to the fact that they are generated in large quantities and that they include partially resistant grains [6–16]. On the other hand, on February 6, 2023, earthquakes measuring 7.7 and 7.6 on the Richter scale struck the Kahramanmaraş province in southern Turkey and near Syria, causing many buildings to collapse or be damaged. After the 2023 Kahramanmaraş earthquakes, the amount of CDW produced is predicted to be around 950 million tons [17]. This CDW is roughly twice Europe’s annual CDW [4]. Earthquakes and other reasons increase the amount of CDW produced worldwide. All or part of the CDW should be recycled, and ways to reuse it should be considered. For this reason, new, advanced studies about the reuse of CDW will be necessary.

Some researchers looked into recycled concrete aggregates (RCAs), made by separating waste concrete particles from debris (CDW), when they investigated how debris can be used as filler in places like the base or subbase layers of roads [18,19]. Furthermore, a cohort of researchers conducted an investigation on crushed bricks (CB), whereas another group of scholars undertook studies on CDW. In the studies that investigated the CDW, various laboratory tests such as sieve analyses, Los Angeles tests, pycnometer tests, compaction tests, unconfined compression tests, direct shear tests, and California Bearing Ratio (CBR) tests were carried out on the CDW, supplied from a location, and then the results of these tests were compared with those of NA. Recent research has focused on enhancing problematic soils with recycled aggregates [16,20–22]. Ujile and Abbey [20] stressed the significance of using recycled aggregates to enhance expansive soil. Al-Baidhani and Al-Taie [22] conducted further research on the improvement of expansive soils with recycled aggregates. Meanwhile, Agarwal et al. [23] determined the pullout capacity of strips embedded in recycled aggregates to ascertain their pullout capacity. On the other hand, there has been a growing emphasis on investigating the utilization of recycled aggregates as a filling material in road bases in recent years [24–26]. In studies in the literature, it was stated that although CDWs tend to have slightly lower performance than those of NA, they still have the potential to be used as filling materials [9,10,12,27–29]. Nevertheless, it was also emphasized that the performances of CDWs depended on their composition. It is clear that the structural characteristics of the buildings, where CDWs are obtained, could affect the composition of CDWs. Accordingly, the performance of CDWs may be valid for the locations where they are supplied since the structural characteristics of the buildings change worldwide. Therefore, in order to obtain a more general and realistic conclusion about the performance of CDWs, the effects of both the strengths of their different grains and their compositions on their engineering properties have to be determined. Previously, some studies evaluated the effect of the compositions of CDWs on their engineering properties [7,30–33]. Molenaar and van Niekerk [30] investigated the impacts of the compositions of CDWs on the compaction degree and emphasized that their compositions were a significant factor. Bareither et al. [34] stated that the shear strengths of granular filling materials may not be accurately determined by the widely used direct shear test device with a small-scale shear box since they contain gravels whose maximum particle size is greater than 4.75 mm. The recycled aggregates to be used as granular filling materials should also contain gravel-size particles. For this reason, the appropriate large-scale shear box might be needed to determine the shear strength of recycled aggregates [19,33,35–40]. One of the most remarkable works that included large-scale direct shear (LSDS) tests performed on a few different recycled aggregates was conducted by Arulrajah et al. [33]. On the other hand, Nataatmadja and Tan [40] revealed that the compressive strengths of concretes from which RCAs were supplied could be a critical parameter for the strength of filling built with the RCAs. However, no comprehensive studies investigating both the strengths of different grains in CDWs and the composition of CDWs together have been found in the literature. It was considered that examining these two parameters together could enable CDWs to be evaluated independently from the locations where they were obtained. Besides, analyzing the impact of these two factors on the performance of CDWs can lead to a safer filling design and a valuable contribution to the literature.

The main purpose of this study is to eliminate some uncertainties, such as the effect of the strengths of different grains and their composition on the engineering properties of debris, so that they could be more re-used in the fillings. Besides, the evaluation of the effects of relative densities and water contents on the investigated recycled aggregates is one of the main subjects of this paper. Nine different recycled aggregates were obtained from waste generating various construction projects for this research. Six of these recycled aggregates were from concrete waste with varying strengths. Two types of recycled aggregates containing waste, such as concrete, brick, etc., in different proportions were generated from the debris of the various demolished buildings. The last one was also created from waste bricks. Initially, the recycled aggregates obtained were made suitable for use as fill material. After that, LSDS and CBR tests were used to determine recycled aggregates’ shear strengths and CBR values. All results were presented by comparing them to similar tests on NA. In addition, the CBR tests determined the CBR values of recycled aggregates.

2 Materials

In this study, test results carried out on ten different types of aggregates were presented. Nine of these aggregates were different types of recycled aggregates, while the last one was NA. Investigated types of aggregates can be categorized into four groups: the CDWs contain the various wastes that emerge as a result of the demolition of old buildings; the RCAs consist of waste concrete with different compressive strengths; the CB includes waste bricks only; and the NA stems from natural rocks.

2.1 The construction and demolition waste and the recycled crushed bricks

First, two different types of CDW aggregates called CDW1 and CDW2 were obtained as debris from various locations in Turkey and then recycled. Initially, wastes that emerged due to the demolition of buildings were supplied. These buildings were older than 30 years and were not found to be resistant to earthquakes based on modern earthquake codes. Subsequently, the recycling process of wastes was carried out by removing the metal parts from the debris and then crushing the remaining wastes (Fig.1) [12]. Lastly, the obtained recycled aggregates of different sizes were mixed to have a gradation suitable for filling materials based on ASTM D1241-07 [41]. As a result of these processes, the recycled aggregates of CDW1 and CDW2 were grouped depending on the locations where they were obtained. The compressive strengths of the waste concrete from the different demolished buildings were tested to evaluate their earthquake performance. The average compressive strengths of the concretes from the demolished buildings, where the CDW1 and the CDW2 were supplied, were 13.4 and 14.5 MPa, respectively.

CDWs could include many different wastes, such as concrete, brick, glass, and trace amounts of other materials. Except for those, the CDWs could contain the NA classified as aggregate, aggregate without binder, or aggregate with hydraulic binder based on BS EN 933-11 [42]. These NA have no binder (or mortar) on their surfaces, and they are produced after the crushing process of the debris [9,10]. According to BS EN 933-11 [42], the proportions of different types of aggregates in CDW1 and CDW2 were determined and presented in Tab.1.

The brick manufacturer supplied waste bricks from their dump to obtain recycled CB. The waste bricks supplied were crushed to obtain a suitable gradation for filling materials [41]. The CDW1, the CDW2, and the CB are shown in Fig.2.

2.2 The recycled concrete aggregates

RCAs are one of the most significant components of the CDWs, but RCAs could also be used as an alternative to NA such as filling materials [6]. In both cases, the compressive strength of concrete from which recycled aggregates were obtained becomes an essential parameter for evaluating the strength of the filling built with the recycled aggregates. Fresh concrete specimens are received during the construction of the buildings and tested after a certain cure period to evaluate their compressive strength. After these tests, the specimens become waste. Besides, many concrete-based elements, such as interlocking concrete pavers (ICP), etc., could become waste after the end of their economic lives. Within the scope of the study, waste concrete specimens with different strengths were obtained, and specimens with similar strengths were grouped. After that, a crusher individually crushed each of the group specimens. Thus, RCAs of different sizes were obtained. Subsequently, the RCAs were mixed to have a gradation suitable for filling materials based on ASTM D1241-07 [41]. As a result of this recycling process, six different RCAs with different strengths have been produced. The five RCAs generated from the waste concrete with concrete compressive strength values of 25, 30, 35, 40, and 45 MPa were named the RCA25, the RCA30, the RCA35, the RCA40, and the RCA45, respectively. Moreover, one type of RCA produced from waste ICP having a concrete compressive strength value of approximately 19 MPa was called the ICP19. They are shown in Fig.3.

2.3 The natural (virgin) aggregates

To compare the performance of recycled aggregates investigated in this study, NA (limestone), widely used in fillings, were supplied from a quarry in the north-western part of the Mesopotamia region. After that, grains of different sizes were mixed to obtain a suitable gradation for fillings, so the NA (Fig.4) was produced [41].

2.4 Geotechnical/Engineering properties of recycled and natural aggregates

The geotechnical and physical properties of recycled and NA were determined by conducting sieve analysis, flatness index, Los Angeles abrasion, water absorption, pycnometer tests, and compaction tests [43–48]. The results of the tests are listed in Tab.2. The gradations, together with upper and lower limits based on ASTM D1241-07 [41], are presented in Fig.5. Furthermore, all samples investigated in this study were classified as GW (well-graded gravel) based on the Unified Soil Classification System [49].

3 Methodology

The CBR is a widely used parameter to compare the load-deformation behaviors of different filling materials with those of the standard filling material. Furthermore, it is known that shear strength is also an essential factor in comparing the mechanical properties of different filling materials. On the other hand, Dawson and Kolisoja [50] classified the potential occurrence of excessive settlement, also known as failure, in the road base layer into three distinct categories (Fig.6). The primary factor leading to the formation of the first group is the inadequate compression of the fill material. Within this particular group, it can be observed that the fill material exhibits an abundance of voids, resulting in an excessive amount of settlement when subjected to a load. In the second cohort, the compaction of the fill material is deemed satisfactory; however, the shear strength of said fill material is deemed inadequate. Shear failure occurs when the applied load surpasses the shear strength of the fill material. Within the third group, it can be observed that the subgrade soil exhibits a notable weakness, thereby lacking the requisite strength necessary for optimal performance [51]. The subgrade and fill layer experience significant settlement when subjected to applied loads. As a result, as shown in the second case in Fig.6, the fill material should have enough shear strength to prevent the excessive settlement that the fills in the second group experienced. For these reasons, CBR and shear strength have commonly been used to evaluate the performance of various filling materials.

3.1 The large-scale direct shear tests

ASTM D3080 [52] states that the width of the shear box must be at least ten times the maximum particle diameter, and its height must be at least six times the maximum particle diameter. Considering this situation, the width of the conventional shear box (50 mm) is not enough for granular filling materials with particles that are coarser than 4.75 mm [39,53]. One of the main subjects of this study is the investigation of the effects of various parameters on the shear strengths of recycled aggregates containing coarse particles (GW). For this reason, to more realistically determine the shear strength of the recycled aggregates whose maximum particle diameter is 15 mm, direct shear tests have been conducted on a test device that has a shear box (150 mm × 150 mm × 120 mm), which is larger than the conventional shear boxes (Fig.7).

For the investigation presented in this paper, 90 direct shear tests were carried out on ten different specimens. While the first nine specimens were of various types of recycled aggregates, the last specimen was obtained from NA. Test samples were prepared in two groups: wet specimens with the optimum water content and dry specimens with relative densities of 35% and 95% (Tab.3). All specimens were reconstituted in the shear box based on achieving the desired relative density value. The specimens were compacted into three layers in the shear box [54]. The compaction process was carried out with a vibratory hammer that gives each layer a vibration (frequency, 22–55 Hz) and an additional load (impact energy, 12 mN) to the ground for 60 s [48]. By taking samples and weighing them before the tests, it was possible to determine the compacted samples’ water contents and dry unit weights for compaction control. In the tests, shear stress was applied by pushing the upper half of the box relative to the lower half until the failure of the specimen, and three normal stresses (21.8, 43.6, and 87.2 kPa) were applied to the top cap. Moreover, the shearing speed in the test was set to 1 mm/min [52]. Furthermore, displacements (vertical and horizontal) and shear forces were measured by the transducers and load cell connected to the data logger during the tests. Then, all readings were recorded via special software. After the tests, shear stress–strain data were plotted, and the peak friction angles were determined using the maximum shear stress obtained and normal stress values applied. It has been known that cohesion has a negligible effect on coarse aggregates’ shear strength [55]. Because of that, the cohesion of granular materials in this study was not considered when evaluating the data. A similar approach was used in several other studies investigating the shear strengths of various granular soils [56–60].

3.2 The California bearing ratio tests

The CBR test could be a test to be preferred for many cases that must be evaluated for the load−displacement behavior of fillings in particular. The test is carried out by penetrating a steel cylindrical piston of 50 mm diameter into the specimen at a rate of 1.27 mm/min. The specimen is prepared in the test mold with a diameter of 152.4 mm [61]. The CBR value was obtained as a result of the test and is a parameter that could be employed to determine the thickness of the base and subbase layers on the roads [12]. In this study, CBR tests were performed on ten different specimens. Like the LSDS tests, nine specimens were created from various recycled aggregates. In addition, the last specimen was also obtained from NA. Furthermore, all test specimens were prepared at their optimum water content and placed in the mold by providing the unit weight values necessary for their conditions. After the preparation, the tests were performed. During the tests, vertical displacements and loads were measured by transducers and a load cell, and all readings were recorded by a computer using special software. The test results were presented as a percentage relative to the reference value by ASTM D1883-99 [61].

4 Results and discussion

4.1 The effect of relative densities on the recycled aggregates

Based on the results of the LSDS tests, the shear stress–horizontal strain curves (τ–ε) for the CDW1 and the RCA45, and the shear strength envelopes for the CDW1, the CDW2, the CB, the RCA45, and the ICP19, which have relative densities of 35% and 95%, are shown in Fig.8. Besides, their internal friction angles (φ) are presented in Tab.4. It was seen that the shear stress–horizontal strain curves had the peaks for the firm state (D_r = 95%), but they didn’t have a clear peak point for the loose state (D_r = 35%), as being in the natural granular soils (coarse-grained soils). On the other hand, it has been known that when loose granular soils are compacted, their internal friction angles also increase. This phenomenon could be associated with the high compressive energy needed to obtain firm-granular soil. In this way, the granular soil becomes more compact. By this, the most elevated internal friction angle for each specimen with recycled aggregates was obtained when this specimen was in a firm state (D_r = 95%). Furthermore, the difference between the internal friction angles of the RCA45 and the CDW1 in the loose state was about 2°, while this difference reached 4.5° in the firm state. The increase in the difference between the internal friction angles as the compression increases could be explained by the increase in the particles’ strength because the specimens’ gradations were the same. The contact areas between particles would also increase as the compaction of coarse-grained soil increased. In this case, the effect of the particle’s strength on the internal friction angle would increase. Considering the results, the strengths of RCA45’s particles should be greater than those of CDW1. The Los Angeles abrasion test results validated the implication regarding the strength of particles. On the other hand, it was observed that the internal friction angles at each relative density were sorted from greatest to smallest as the RCA45, the ICP19, the CB, the CDW2, and the CDW1. These outcomes demonstrated that the performance of recycled aggregates produced from waste concrete was greater than that of recycled aggregates obtained from waste CB or debris. When considering the Los Angeles test results in conjunction, it is believed that the disparity in strength between the particles plays a significant role in the performance of the internal friction angle. The observed escalation in performance disparity with increasing relative density can be attributed to an increase in the contact area of the particles.

4.2 The effect of optimum water content on the recycled aggregates

For recycled aggregates compacted at their optimum water content, the shear stress–horizontal strain curves for the CDW1 and the RCA45 and the shear strength envelopes for the CDW1, the CDW2, the CB, the RCA45, and the ICP19 are shown in Fig.9. In addition, their internal friction angles and CBR values are presented in Tab.5. Accordingly, the shear stress–horizontal strain curves of the recycled aggregates at optimum water content behave between the firm state (D_r = 95%) and the loose state (D_r = 35%). These curves have a little peak at first, including the increase and then a decrease, and then become roughly parallel to the x-axis. The RCA45, on the other hand, was found to have a higher internal friction angle than those of other recycled aggregates, with a difference in internal friction angles ranging from 5° to 1°. Furthermore, the internal friction angle of the CDW2 became about 2° greater than that of the CB at the optimum water content, even though the internal friction angle of the CB was greater than that of the CDWs in different relative densities. It has been considered that this phenomenon, observed only in the wet case, is related to the CB’s high optimum water content based on its excessive water absorption, which is approximately two times that of the CDW2.

Arulrajah et al. [33] performed tests on a type of RCA and CB to determine their shear strength. They found that the CB and RCA’s internal friction angles were 52° and 53°, respectively, disregarding cohesion for coarse materials. When considering the gradation differences between the recycled aggregates and the different structures where samples were taken, it is possible to conclude that the findings of this study are practical and comparable to similar studies.

The CBR values of RCAs were found to be higher than those of other recycled aggregates. The internal friction angles obtained from the LSDS tests showed a similar relationship between specimens. Nonetheless, it was determined that the CBR value of the CB is the lowest, having an opposite relationship seen in the internal friction angles. This phenomenon could be explained by the mechanism of the CBR test and the CB’s high optimum water content, which occurs depending on its high water absorption. The load values corresponding to the displacements (2.5 and 5.0 mm) used for the CBR calculation were low due to the high optimum water content of the CB (21.5%). Therefore, the CB’s CBR value, calculated by dividing these loads by the standard loads, was lower than that of other specimens. The observed outcome is attributable to reduced friction between grains caused by elevated water content within the grains. CB’s high water absorption value puts it in a disadvantaged position compared to other recycled aggregates. Indeed, a circumstance close to this phenomenon described for the CB was noted before. Poon and Chan [6] prepared a new specimen by mixing the RCA with the CB at 50%. They determined the optimum water content and the CBR values of the RCA as 11.8% and 66%, respectively. However, they also noted that the optimum water content increased to 16%, and the CBR value decreased to 43% when the CB and the RCA were mixed. Poon and Chan’s [6] finding corroborates the outcome observed in this research.

4.3 The effect of concrete’s compressive strength in recycled aggregates

The shear strength envelopes of the RCA45 and the ICP19 were compared in Fig.10. These envelopes were separately given for wet specimens (ω_opt, D_r = 75%–85%) and dry specimens (D_r = 35% and D_r = 95%). In this way, the effect of the compressive strength of concrete on the friction angle of recycled aggregates was discussed. In addition, the internal friction angles and the CBR values of all RCAs investigated in this study are presented in Tab.6. The results showed that as the compressive strengths of concrete decreased, the internal friction angles and CBR values of the recycled aggregates produced also decreased. In the loose state (D_r = 35%), when the concrete’s compressive strength decreased from 45 to 19 MPa, RCA’s internal friction angle decreased by 1°. In the firm state (D_r = 95%), this decrease reached 2°. When the specimens had optimum water content, the same decline in concrete’s compressive strength decreased the CBR value by about 17%. The reduction in the compressive strength of concrete resulted in a corresponding decrease in the strength of the individual grains. The decline in the strength of the specimen’s grains resulted in increased grain breakdown due to inter-particle friction when shear stress was applied to the specimen, reducing the specimen’s shear strength.

The changes in the internal friction angles of RCAs depending on relative densities or optimum water content are shown in Fig.11. Besides, the changes in the CBR values of RCAs with optimum water content are also shown in this figure. Considering the results of the tests, it was noticed that there was an approximately linear relationship between the concrete’s compressive strengths and the internal friction angles, or CBR values, of the recycled aggregates generated from those concretes. The linear equations with a relatively higher correlation coefficient (R > 0.89) were obtained from these relationships. Equations related to the internal friction angle were presented in Eqs. (1)–(3). Equations (1) and (2) were developed based on dry specimens with a D_r of 35% and a D_r of 95%, while Eq. (3) was developed for the specimens at optimum water content. In addition, an equation related to CBR values was given in Eq. (4). Equation (4) is only valid at the optimum water content. Given that the specific circumstances, such as gradations, crushing processes, and so on, are taken into account, these four equations could be used to help with the preliminary design of the RCA filling. The compressive strength of concrete, from which recycled aggregates are made, is represented by the (RCA)_MPa in these equations. As an example of how to use these equations, suppose a designer would like to create a fill with RCA, and the internal friction angle of the fill material should be 45°. The designer can determine the compressive strength of the concrete from which the RCA will be obtained by using suggested equations depending on the relative density or water content. Accordingly, the designer can select the appropriate waste material for the design. In other words, through Eqs. (1)–(4), it will be possible to estimate the shear strength parameter φ and the bearing capacity (CBR) of RCAs to be used as filling material in any geotechnical project.

(1)\phi \left({}^{\circ }\right)=0.0449{\left(RCA\right)}_{\mathrm{M}\mathrm{P}\mathrm{a}}+54.462,\mathrm{f}\mathrm{o}\mathrm{r}\omega =0,D_{\mathrm{r}}=35\mathrm{\%},

(2)\phi \left({}^{\circ }\right)=0.0833{\left(RCA\right)}_{\mathrm{M}\mathrm{P}\mathrm{a}}+59.885,\mathrm{f}\mathrm{o}\mathrm{r}\omega =0,D_{\mathrm{r}}=95\mathrm{\%},

(3)\phi \left({}^{\circ }\right)=0.0672{\left(RCA\right)}_{\mathrm{M}\mathrm{P}\mathrm{a}}+57.828,\mathrm{f}\mathrm{o}\mathrm{r}{\omega }_{\mathrm{o}\mathrm{p}\mathrm{t}},D_{\mathrm{r}}=75\mathrm{\%}-85\mathrm{\%},

(4)\begin{array}{rl}& CBR\left(\mathrm{\%}\right)=0.6908{\left(RCA\right)}_{\mathrm{M}\mathrm{P}\mathrm{a}}+60.27,\\ & \mathrm{f}\mathrm{o}\mathrm{r}{\omega }_{\mathrm{o}\mathrm{p}\mathrm{t}},D_{\mathrm{r}}=75-85\mathrm{\%}.\end{array}

Equation (5), derived from Eqs. (3) and (4), also helps estimate the internal friction angle of RCA material with its own CBR values. For example, to use Eq. (5), suppose a designer would like to create a fill with RCA, and the designer supplies the RCA from somewhere. The designer could have an opinion about the internal friction angle of the RCA by determining its CBR value. As a result, the designer bases the design on the RCA’s internal friction angle.

(5)\begin{array}{rl}& \phi \left({}^{\circ }\right)=0.0973CBR\left(\mathrm{\%}\right)+51.9637,\mathrm{f}\mathrm{o}\mathrm{r}{\omega }_{\mathrm{o}\mathrm{p}\mathrm{t}},\\ & D_{\mathrm{r}}=75\mathrm{\%}-85\mathrm{\%}.\end{array}

It could be put forward that the relationships between the concrete’s compressive strengths and the internal friction angles, or the CBR values, depend on the particle strength largely due to their similarity in the relative densities and particle shapes. Following this, it was seen that as the concrete’s compressive strength increased, the Los Angeles abrasion loss (LA) percentage of RCA approximately linearly decreased (Fig.12). Equation (6) shows the relationship between the LA and the concrete’s compressive strength.

(6)LA\left(\mathrm{\%}\right)=-0.22{\left(RCA\right)}_{\mathrm{M}\mathrm{P}\mathrm{a}}+44.97.

4.4 The effect of content in recycled aggregates

The CDW includes a considerable quantity of the RCA, the NA (aggregate without binder, in other words, natural aggregate), and the CB. However, it could have a small quantity of many other wastes. Besides, it is clear that the CDW’s shear strength could change depending on its composition since it includes various wastes with different characteristics. The compositions of the CDWs investigated in this study were different, although the average compressive strengths of the concrete from which CDWs were obtained were similar. The CDW2 contained 63% aggregate without a binder (it could be called the NA or the CDW2-NA) and 29% waste concrete (CDW2-RCA), while the CDW1 contained 30% NA (CDW1-NA) and 62% waste concrete (CDW1-RCA). Furthermore, the internal friction angle of CDW2 was determined to be greater than that of CDW1 under the same conditions. To clarify the shear strength difference between CDW1 and CDW2, the Los Angeles abrasion tests have been conducted on the specimens of the CDW2-NA, the CDW2-RCA, the CDW1-NA, and the CDW1-RCA, which have been prepared by separating them from the CDW2 and CDW1, respectively. As a result, it was identified that the LA values of the CDW2-NA and the CDW1-NA were 20.9% and 27.5%, while those of the CDW2-RCA and the CDW1-RCA were 48.4% and 50.1%, respectively. That means the LA value of the natural aggregate in CDWs was approximately two times less than that of RCA. In addition, the natural aggregate percentage in the CDW2 was about two times greater than that of the CDW1. It was considered that these reasons caused the difference between the shear strengths of the CDW1 and the CDW2. Accordingly, the internal friction angle of the CDW2 with the optimum water content was approximately 4° greater than that of the CDW1. Besides, the CBR value of CDW2 was about 7% greater than that of CDW1. These results were due to a difference of approximately 33% in natural aggregate content between CDW1 and CDW2. For all the cases investigated in this study, the effect of the percentage of NA in the CDWs on the internal friction angle is shown in Fig.13 by comparing them with the result of the NA obtained from a quarry. Besides, the changes in the CBR values for specimens with optimum water content are also given in Fig.13. As a result, it could be seen that as the percentage of natural aggregate in the CDWs increased, their internal friction angles and CBR values increased. The main reason for this increase is that the NA in CDWs have higher strength than other components. It has been observed that these high-strength grains have high friction between grains and less fragmentation during shearing.

Equations with relatively higher correlation coefficients were developed between the percentages of NA in the CDWs ((CDW)_NA) and their internal friction angles and CBR values. Equations related to the internal friction angle were presented in Eqs. (7)–(9). Equations (7) and (8) were developed for dry specimens with a D_r of 35% and a D_r of 95%, while Eq. (9) was developed for the specimens at optimum water content (ω_opt, D_r = 70%–80%). Besides, an equation related to CBR values was given in Eq. (10).

(7)\phi \left({}^{\circ }\right)=0.0449{\left(CDW\right)}_{\mathrm{N}\mathrm{A}}+52.858,\mathrm{f}\mathrm{o}\mathrm{r}\omega =0,D_{\mathrm{r}}=35\mathrm{\%},

(8)\phi \left({}^{\circ }\right)=0.0{858\left(CDW\right)}_{\mathrm{N}\mathrm{A}}+55.845,\mathrm{f}\mathrm{o}\mathrm{r}\omega =0,D_{\mathrm{r}}=95\mathrm{\%},

(9)\phi \left({}^{\circ }\right)=0.0692{\left(CDW\right)}_{\mathrm{N}\mathrm{A}}+54.642,\mathrm{f}\mathrm{o}\mathrm{r}{\omega }_{\mathrm{o}\mathrm{p}\mathrm{t}},D_{\mathrm{r}}=70\mathrm{\%}-80\mathrm{\%},

(10)\begin{array}{rl}& CBR\left(\mathrm{\%}\right)=0.3968{\left(CDW\right)}_{\mathrm{N}\mathrm{A}}+60.296,\\ & \mathrm{f}\mathrm{o}\mathrm{r}{\omega }_{\mathrm{o}\mathrm{p}\mathrm{t}},D_{\mathrm{r}}=70\mathrm{\%}-80\mathrm{\%}.\end{array}

4.5 The comparison of recycled and natural aggregates

The internal friction angles in all the cases examined in this study are shown in Fig.14. From this figure, it can be seen that the NA had the best performance in terms of internal friction angle for all the cases. That means all recycled aggregates had internal friction angles lower than those of the NA, although their internal friction angles were close to those of the NA. In all cases studied in this paper, the RCAs’ internal friction angle values (RCA45, RCA40, RCA35, RCA30, RCA25, and ICP19) were proportional to the concrete’s compressive strengths. A difference of approximately 1° was obtained between the internal friction angles of the RCA45 and the NA, while this difference became about 3° for the ICP19 and the NA. On the other hand, it has been determined that the internal friction angle values of the CB and the CDWs were less than those of the RCAs. Besides, the difference between the internal friction angles of the NA and the CB was approximately 4°. However, it was seen that the difference in internal friction angle reached about 6° when comparing the internal friction angles of the NA and the CDWs. As a result, the CB and CDWs could need to be improved since their performances were poorer than the NA, even if their internal friction angles were not below 54°. Based on the results of this research, a flowchart was created to help engineers better incorporate recycled aggregates into filling designs. This flowchart is shown in Fig.15.

The CBR values of various types of ten specimens investigated in this study, prepared at the optimum water content, are shown in comparatively Fig.16. As a result of the tests, it was determined that the CBR value of NA was greater than 100%. In contrast, those of various recycled aggregates could not reach 100%. However, the CBR values of recycled aggregates, except for the CB, were close to 100%. The CBR value of the CB was also approximately 60%. Because of that, it has been predicted that the CBR values of the recycled aggregates, except for the CB, could reach the NA with a simple improvement, while a more advanced improvement could be needed for the CB. Furthermore, it could be stated that since the RCAs did not include the particles of the CB, their engineering parameters improved. Accordingly, it would be beneficial to have a low percentage of CB particles in the CDW. In many studies, such as this study, the percentage of CB particles in CDWs has been seen to be less than 12% [9,12,27].

4.6 The sensitivity analysis of independent variables

Sensitivity analysis is a method used to assess the degree of change in an output variable (y) in response to variations in an input variable (x). One-at-a-time sensitivity analysis (OAT) was used on the variables in the proposed equations as part of this study to find out how sensitive these variables were. The OAT sensitivity study figures out the extent of change in the output as a response to a change in only one input at a time [62]. A coefficient-normalized sensitivity index (S_{ijk}^{\mathrm{n}}) was calculated for variables to assess the sensitivity of each variable quantitatively and compare the magnitudes of sensitivities among themselves (Eqs. (11)).

(11)

where Y_ji and X_ki are the values of the internal friction angle prediction j and variable k. The partial derivative in the coefficient of sensitivity index can be approximated into a standard central difference approximation (Eq. (11)). For each variable, X_k, two coefficient-normalized sensitivity indices (S_{ijk}^{\mathrm{n}}) were calculated using the 20% increased and 20% decreased values of variables (X_j,1.2i > X_j,i and X_j,0.8i < X_j,i) [62]. The average of the two coefficient-normalized sensitivity indices calculated was used to find a coefficient-normalized sensitivity index for each variable [63]. Tab.7 presents these normalized sensitivity indexes in order. Based on these indexes, it can be concluded that (CDW)_NA (for D_r = 95% and ω = 0) is the most efficient variable that affects the internal friction angle.

5 Conclusions

The engineering properties of heterogeneous recycled aggregates rely on their particle strength and composition. This paper focuses on determining the effects of these two factors on the engineering parameters of recycled aggregates. For this purpose, the LSDS and CBR tests were carried out on the nine different recycled aggregates. The five test specimens (the RCA45, the RCA40, the RCA35, the RCA30, the RCA25, and the ICP19) were created from waste concrete. The two test specimens (the CDW1 and the CDW2) were obtained from debris. The last test specimen (the CB) originated from waste CB. The specimens were prepared in wet and dry conditions. The wet specimens had the optimum water content, and the dry specimens had relative densities of 35% and 95%. Furthermore, the results of the tests were evaluated by comparing them with the those of similar tests on NA obtained from a quarry. The valuable results obtained from the study are summarized in the following items.

1) For the dry specimens (D_r = 35% and D_r = 95%) and wet specimens (ω_opt), the behaviors of the shear stress–horizontal strain curves of natural and recycled aggregates tended to be similar. This phenomenon was valid for natural and recycled aggregate specimens with the same condition.

2) The internal friction angles and CBR values of the RCAs were strongly affected by particle strength. On the other hand, the results of the Los Angeles tests demonstrated that there could be an approximately linear relationship between the compressive strength of concrete, from which the RCA was produced, and the strength of particles. Accordingly, it was determined that the internal friction angles and CBR values of the RCA increased with the increase in the concrete’s compressive strength. Besides, this increase was approximately linear.

3) The percentages of various particles in the CDWs considerably change their internal friction angles and CBR values. The internal friction angle and the CBR value of the CDW increased with the increase in the percentage of NA existing in the CDW. This increase was approximately linear. Meanwhile, having a high percentage of CB particles (approximately 10%–12%) in the CDWs would not be beneficial for the engineering parameters of the CDWs due to the CB’s high water absorption.

4) Using the equations proposed in this study, the strength of the recycled aggregates can be estimated for use in the preliminary design of the fillings based on the percentage of the NA contained in the recycled aggregates or the average compressive strength of concrete that the recycled aggregates produce. However, it has been recommended that designers should compare the conditions in this paper with the conditions in their own case when using these equations.

5) For all cases (i.e., for both the dry specimens and the wet specimens) investigated in this study, it was observed that the RCAs have greater internal friction angles and CBR values than those of the CB and CDWs. Therefore, the RCAs exhibit the best performance among the recycled aggregates.

6) The internal friction angles and the CBR values of the NA were greater than those of all the recycled aggregates studied in this paper. Nevertheless, the RCAs show a similar performance to the NA.

7) Considering the internal friction angles, it could be specified that the internal friction angles of the CB and the CDWs could be acceptable for filling materials. However, their internal friction angels are less than those of the RCAs. Nevertheless, they could need improvement if the same performance compared to the NA is expected

8) Considering the CBR values, it has been stated that only using the CB as a filling material in fillings may not be appropriate since its CBR value is very low (approximately 60%), and its optimum water content is very high (21.15%). For this reason, the CB could need to be improved or mixed with other filling materials at a low percentage if it is desired to be used in fillings.

9) Although the study demonstrates inclusivity by incorporating low- and high-quality recycled aggregates and examining a wide range of recycled material mixtures, the comprehensive findings from this study may apply to recycled materials that share similar characteristics, such as compression levels, water content, etc. In this case, the fill design made from recycled aggregates can be more reliable by increasing laboratory studies and testing it in the field to see how well it works.

References

[1] Arulrajah A, Piratheepan J, Aatheesan T, Bo M W. Geotechnical properties of recycled crushed brick in pavement applications. Journal of Materials in Civil Engineering, 2011, 23(10): 1444–1452

[2] Cabalar A F, Abdulnafaa M D, Isik H. The role of construction and demolition materials in swelling of a clay. Arabian Journal of Geosciences, 2019, 12(11): 361

[3] Han J, Thakur J K. Sustainable roadway construction using recycled aggregates with geosynthetics. Sustainable Cities and Society, 2015, 14: 342–350

[4] Vieira C S, Pereira P M. Use of recycled construction and demolition materials in geotechnical applications: A review. Resources, Conservation and Recycling, 2015, 103: 192–204

[5] Saribas I, Ok B. Seismic performance of recycled aggregate-filled cantilever reinforced concrete retaining walls. Advances in Mechanical Engineering, 2019, 11(4): 1–11

[6] Poon C S, Chan D. Feasible use of recycled concrete aggregates and crushed clay brick as unbound road sub-base. Construction and Building Materials, 2006, 20(8): 578–585

[7] Arulrajah A, Piratheepan J, Disfani M M, Bo M W. Geotechnical and geoenvironmental properties of recycled construction and demolition materials in pavement subbase applications. Journal of Materials in Civil Engineering, 2013, 25(8): 1077–1088

[8] Arulrajah A, Piratheepan J, Disfani M M. Reclaimed asphalt pavement and recycled concrete aggregate blends in pavement subbases: Laboratory and field evaluation. Journal of Materials in Civil Engineering, 2014, 26(2): 349–357

[9] Jiménez J R, Ayuso J, Agrela F, López M, Galvín A P. Utilisation of unbound recycled aggregates from selected CDW in unpaved rural roads. Resources, Conservation and Recycling, 2012, 58: 88–97

[10] Cerni G, Cardone F, Bocci M. Permanent deformation behaviour of unbound recycled mixtures. Construction and Building Materials, 2012, 37: 573–580

[11] Gabr A R, Cameron D A. Properties of recycled concrete aggregate for unbound pavement construction. Journal of Materials in Civil Engineering, 2012, 24(6): 754–764

[12] Ok B, Sarici T, Talaslioglu T, Yildiz A. Geotechnical properties of recycled construction and demolition materials for filling applications. Transportation Geotechnics, 2020, 24: 100380

[13] Guo Q, Zhan L, Shen Y, Wu L, Chen Y. Classification and quantification of excavated soil and construction sludge: A case study in Wenzhou, China. Frontiers of Structural and Civil Engineering, 2022, 16(2): 202–213

[14] Kaloop M R, Gabr A R, El-Badawy S M, Arisha A, Shwally S, Hu J W. Predicting resilient modulus of recycled concrete and clay masonry blends for pavement applications using soft computing techniques. Frontiers of Structural and Civil Engineering, 2019, 13(6): 1379–1392

[15] Ying Y, Huang Y, Gao X, Qi X, Sun Y. Effects of coarse and fine aggregates on long-term mechanical properties of sea sand recycled aggregate concrete. Frontiers of Structural and Civil Engineering, 2021, 15(3): 754–772

[16] Bağriaçik B, Beycioğlu A, Topolinski S, Akmaz E, Sert S, Güner E D. Assessment of glass fiber-reinforced polyester pipe powder in soil improvement. Frontiers of Structural and Civil Engineering, 2021, 15(3): 742–753

[17] Xiao J, Deng Q, Hou M, Shen J, Gencel O. Where are demolition wastes going: refection and analysis of the February 6, 2023 earthquake disaster in Turkey. Low-Carbon Materials and Green Construction, 2023, 1: 17

[18] Park T. Application of construction and building debris as base and subbase materials in rigid pavement. Journal of Transportation Engineering, 2003, 129(5): 558–563

[19] Arulrajah A, Rahman M A, Piratheepan J, Bo M W, Imteaz M A. Interface shear strength testing of geogrid-reinforced construction and demolition materials. Advances in Civil Engineering Materials, 2013, 2(1): 189–200

[20] Ujile M C, Abbey S J. The use of fine portions from construction and demolition waste for expansive soil stabilization: A review. Frontiers of Structural and Civil Engineering, 2022, 16(7): 803–816

[21] Amakye S Y, Abbey S J, Booth C A, Mahamadu A M. Enhancing the engineering properties of subgrade materials using processed waste: A Review. Geotechnics, 2021, 1(2): 307–329

[22] Al-Baidhani A F, Al-Taie A J. Recycled crushed ceramic rubble for improving highly expansive soil. Transportation Infrastructure Geotechnology, 2020, 7(3): 426–444

[23] Agarwal A, Ramana G V, Datta M, Soni N K, Satyakam R. Pullout behaviour of polymeric strips embedded in mixed recycled aggregate (MRA) from construction & demolition (C&D) waste––Effect of type of fill and compaction. Geotextiles and Geomembranes, 2023, 51(3): 405–417

[24] Esfahani M A. Evaluating the feasibility, usability, and strength of recycled construction and demolition waste in base and subbase courses. Road Materials and Pavement Design, 2020, 21(1): 156–178

[25] Xue Y, Arulrajah A, Narsilio G A, Horpibulsuk S, Chu J. Washed recycled sand derived from construction and demolition wastes as engineering fill materials. Construction and Building Materials, 2022, 358: 129433

[26] Nasiri A, Eslami A, Fahimifar A. Assessment of construction and demolition waste depot materials for filling. Transportation Infrastructure Geotechnology, 2024, 11: 236–262

[27] Vieira C S, Pereira P M. Damage induced by recycled construction and demolition wastes on the short-term tensile behaviour of two geosynthetics. Transportation Geotechnics, 2015, 4: 64–75

[28] Mehrjardi G T, Azizi A, Haji-Azizi A, Asdollafardi G. Evaluating and improving the construction and demolition waste technical properties to use in road construction. Transportation Geotechnics, 2020, 23: 100349

[29] Cabalar A F, Abdulnafaa M D, Isbuga V. Plate loading tests on clay with construction and demolition materials. Arabian Journal for Science and Engineering, 2021, 46(5): 4307–4317

[30] Molenaar A A A, van Niekerk A A. Effects of gradation, composition, and degree of compaction on the mechanical characteristics of recycled unbound materials, Transportation Research Record. Transportation Research Record: Journal of the Transportation Research Board, 2002, 1787(1): 73–82

[31] Leite F C, Motta R S, Vasconcelos K L, Bernucci L. Laboratory evaluation of recycled construction and demolition waste for pavements. Construction and Building Materials, 2011, 25(6): 2972–2979

[32] Barbudo A, Agrela F, Ayuso J, Jiménez J R, Poon C S. Statistical analysis of recycled aggregates derived from different sources for sub-base applications. Construction and Building Materials, 2012, 28(1): 129–138

[33] Arulrajah A, Disfani M M, Horpibulsuk S, Suksiripattanapong C, Prongmanee N. Physical properties and shear strength responses of recycled construction and demolition materials in unbound pavement base/subbase applications. Construction and Building Materials, 2014, 58: 245–257

[34] Bareither C A, Benson C H, Edil T B. Comparison of shear strength of sand backfills measured in small-scale and large-scale direct shear tests. Canadian Geotechnical Journal, 2008, 45(9): 1224–1236

[35] Sivakumar V, McKinley J D, Ferguson D. Reuse of construction waste: Performance under repeated loading. Geotechnical Engineering, 2004, 157(2): 91–96

[36] Zekkos D, Athanasopoulos G A, Bray J D, Grizi A, Theodoratos A. Large-scale direct shear testing of municipal solid waste. Waste Management, 2010, 30(8–9): 1544–1555

[37] Soból E, Sas W, Szymański A. Scale effect in direct shear tests on recycled concrete aggregate. Studia Geotechnica et Mechanica, 2015, 37(2): 45–49

[38] Vieira C S, Pereira P M. Interface shear properties of geosynthetics and construction and demolition waste from large-scale direct shear tests. Geosynthetics International, 2016, 23(1): 62–70

[39] Saberian M, Li J, Perera S T A M, Zhou A, Roychand R, Ren G. Large-scale direct shear testing of waste crushed rock reinforced with waste rubber as pavement base/subbase materials. Transportation Geotechnics, 2021, 28: 100546

[40] Nataatmadja A, Tan Y L. Resilient response of recycled concrete road aggregates. Journal of Transportation Engineering, 2001, 127(5): 450–453

[41] ASTMD1241-07. Standard Specification for Materials for Soil-Aggregate Subbase, Base, and Surface Courses. West Conshohocken, PA: ASTM, 2007

[42] BSEN 933-11. Tests for Geometrical Properties of Aggregates Part 11: Classification Test for The Constituents of Coarse Recycled Aggregate. London: British Standards Institution, 2009

[43] ASTMC127-01. Standard Test Method for Density, Relative Density (Specific Gravity), and Absorption of Coarse Aggregate. West Conshohocken, PA: ASTM, 2001

[44] ASTMC128-01. Standard Test Method for Density, Relative Density (Specific Gravity), and Absorption of Fine Aggregate. West Conshohocken, PA: ASTM, 2001

[45] ASTMC131-03. Standard Test Method for Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in The Los Angeles Machine. West Conshohocken, PA: ASTM, 2003

[46] ASTMD422-63. Standard Test Method for Particle-Size Analysis of Soils. West Conshohocken, PA: ASTM, 2009

[47] ASTMD4791-10. Standard Test Method for Flat Particles, Elongated Particles, or Flat and Elongated Particles in Coarse Aggregate. West Conshohocken, PA: ASTM, 2010

[48] ASTMD7382. Test Methods for Determination of Maximum Dry Unit Weight and Water Content Range for Effective Compaction of Granular Soils Using a Vibrating Hammer. West Conshohocken, PA: ASTM, 2008

[49] ASTMD2487-00. Standard Practice for Classification of Soils for Engineering Purposes (Unified Soil Classification System). West Conshohocken, PA: ASTM, 2000

[50] DawsonAKolisojaP. Managing Rutting in Low Volume Roads. ROADEX II, Final Report Part 2.1. 2006

[51] KazmeeHTutumluerE. Evaluation of Aggregate Subgrade Materials Used as Pavement Subgrade/Granular Subbase. Illinois Center for Transportation Research Report FHWA-ICT-15-013. 2015

[52] ASTMD3080. Standard Test Method for Direct Shear Test of Soils Under Consolidated Drained Conditions. West Conshohocken, PA: ASTM, 2004

[53] Abozraig M, Ok B, Yildiz A. Determination of shear strength of coarse-grained soils based on their index properties: Comparison between diferent statistical approaches. Arabian Journal of Geosciences, 2022, 15(7): 593

[54] Ok B, Colakoglu H, Daglı U. Evaluation of the geogrid various sustainable geomaterials interaction by direct shear tests. Geomechanics and Engineering, 2023, 34(2): 173–186

[55] DasB M. Principles of Geotechnical Engineering. 7th ed. Eagan, MN: Thomson Corporation, 2010

[56] Park J W, Song J J. Numerical simulation of a direct shear test on a rock joint using a bonded-particle model. International Journal of Rock Mechanics and Mining Sciences, 2009, 46(8): 1315–1328

[57] Basudhar P K. Modeling of soil–woven geotextile interface behavior from direct shear test results. Geotextiles and Geomembranes, 2010, 28(4): 403–408

[58] Infante D J U, Martinez G M A, Arrua P A, Eberhardt M. Shear strength behavior of different geosynthetic reinforced soil structure from direct shear test. International Journal of Geosynthetics and Ground Engineering, 2016, 2(2): 1–16

[59] Bahaaddini M. Effect of boundary condition on the shear behaviour of rock joints in the direct shear test. Rock Mechanics and Rock Engineering, 2017, 50(5): 1141–1155

[60] Stathas D, Wang J P, Ling H I. Model geogrids and 3D printing. Geotextiles and Geomembranes, 2017, 45(6): 688–696

[61] ASTMD1883-99. Standard Test Method for CBR (California Bearing Ratio) of Laboratory-Compacted Soils. West Conshohocken, PA: ASTM, 1999

[62] NationalCooperative Highway Research Program Project. Sensitivity Evaluation of MEPDG Performance Prediction. Washington, D.C.: NCHRP, 2011

[63] KayaO. Investigation of AASHTOW are Pavement ME Design/Darwin-ME^TM performance prediction models for Iowa pavement analysis and design. Thesis for the Master’s Degree. Ames, IA: Iowa State University, 2015

RIGHTS & PERMISSIONS

Higher Education Press