Shakedown analysis of PET blends with demolition waste as pavement base/subbase materials using experimental and neural network methods

Highlights

• Shakedown analysis of plastic waste and demolition waste blends under traffic loading.

• Neural network-based shakedown analysis and permanent deformation characterization.

• Effect of plastic waste on permanent deformation behavior of demolition wastes.

Abstract

Recycling and reusing construction and demolition (C&D) wastes for civil engineering construction activities has been identified as an energy-saving and sustainable solution. The purpose of this study is to evaluate the deformation behavior of two types of C&D waste materials, namely recycled concrete aggregate (RCA) and crushed brick (CB) when mixed with polyethylene terephthalate (PET) plastic waste. RCA and CB were mixed with 1%, 3%, 5%, and 7% of PET, and the permanent deformation behavior of the blends was evaluated using repeated load triaxial (RLT) test. The shakedown criterion was utilized for identifying the deformation behavior of blends. Most of the PET/RCA and PET/CB blends exhibited Range B (plastic creep) and Range C (incremental collapse) response, respectively, in the investigated stress levels. Shakedown analysis of the test results indicated that up to 3% PET could be mixed with RCA for base/subbase applications, while CB should be mixed with 1% PET, in the subbase layer. Artificial neural network (ANN) method was next used to simulate the permanent strain and shakedown behavior of the blends with consideration of the physical properties and stress states. The ANN model was found to be highly efficient for simulating the permanent strain graph and identifying the shakedown behavior of the blends. A sensitivity analysis was subsequently performed to investigate the impact of input variables on the permanent deformation behavior and the results indicated that number of cycles and confining stress were the most important factors.

Introduction

The increasing rate of solid waste generation in the modern world as a result of growing construction activities and industrial developments has raised serious environmental and social concerns [1]. The generated waste is usually disposed into the landfills which requires a large volume of landfills and has detrimental effects on the environment. Therefore, proper waste management has been identified as a priority by many countries and environmental agencies across the world [2]. Recycling and reusing wastes for construction projects is a sustainable alternative to landfilling which has gained growing interest over the past few years, with several advantages such as reducing the extraction of non-renewable natural resources and saving energy [1], [3]. The efficient use of recycled waste materials is a sustainable solution toward achieving a low-carbon future.

Construction and demolition (C&D) waste materials have been used for various civil engineering applications [3]. Recycled concrete aggregate (RCA) and crushed brick (CB) are two major streams of C&D materials, accounting for more than 50% of the C&D waste generated annually in Australia [3]. It is estimated that nearly 8.7 million tons of RCA and 1.3 million tons of CB are disposed into the landfills annually in Australia [3]. The potential usage of C&D materials as pavement base/subbase material has been examined by several researchers. For example, Arulrajah et al. [3] investigated the geotechnical and geoenvironmental properties of C&D materials. Ghorbani et al. [2], [12] investigated the effect of environmental factors such as freeze-thaw cycles and temperature on the deformation response of C&D materials. The shear strength response of C&D materials with geosynthetic inclusion was investigated by Vieira and Pereira [4]. Puppala et al. [5] studied the resilient modulus of cement-treated reclaimed asphalt pavement for usage in pavement base layer.

Plastic waste generation rate is increasing at an alarming rate due to the wide use of plastics in everyday life of humans. In Australia, almost 2.24 million tons of plastic waste is produced each year [6]. Polyethylene terephthalate (PET) is a type of plastic used for making bottles and containers. Recently, attempts have been made to use the generated plastic wastes for different applications in civil engineering to divert the plastic wastes from landfills. Choudhary et al. [7], Yaghoubi et al. [8], and Perera et al. [9] used plastic wastes for the sustainable construction of flexible pavements. The addition of plastic waste to pavement granular materials can considerably influence the strength and deformation properties of materials, which need to be fully understood prior to their wide usage for construction of pavement base/subbase layers.

One of the key factors in the design of flexible pavements is the permanent deformation (rutting) of the granular layers. The accumulation of permanent strain in the pavement base/subbase layer must be within the allowable limits to ensure the safe design and stability of the pavement system. The most utilized approach for evaluating the permanent deformation of unbound granular materials is by performing repeated load triaxial (RLT) test. Recently, Soliman and Shalaby [10] and Ghorbani et al. [2] investigated the permanent deformation behavior of pavement materials using a single-stage RLT test, in which a constant stress level was applied to the specimen. Ghorbani et al. [12] and Chen et al. [13] used a multi-stage RLT test for characterizing the permanent deformation behavior of pavement base materials in different stress levels. Pavement materials are exposed to different magnitudes of dynamic loads because of the moving traffic. On the other hand, inclusion of plastic wastes in combination with demolition aggregates changes the nature of the rigid mixture into a soft-rigid structure. In this soft-rigid structure, applied loads are mainly carried by the rigid demolition aggregates. The difference in the rigidity of demolition aggregates and plastic waste could highly affect the permanent deformation behavior of the mixture which might not be apparent if the magnitude of the applied load is low. Therefore, it is important to evaluate the performance of C&D materials in different stress levels by performing a multistage RLT test, particularly when blended with soft materials such as plastic waste. However, limited studies have employed a multistage RLT test for characterizing the permanent deformation behavior of plastic waste/pavement granular materials.

Unbound granular materials experience accumulation of the permanent strain under the applied repeated load which results in gradual deterioration and ultimately failure of the sample [13]. Shakedown limit is defined as the maximum stress level below which the accumulation of the permanent strain stabilizes after a certain number of load cycles [13], [14]. The shakedown theory [15] has been applied by several researchers to evaluate the deformation response of pavement materials [10], [12], [14], [16]. Werkmeister [14] proposed the following shakedown ranges for classifying the material behavior based on the results of the RLT test [14]:

Range A (plastic shakedown): Large accumulation of permanent strain in initial hundreds of cycles (post-compaction period), which then becomes constant and response of material becomes resilient over the next cycles, i.e., there is a steep decrease in the permanent strain rate [14].

$${\epsilon }_{p-5000}-{\epsilon }_{p-3000}<0.045\times {10}^{-3}$$

Range B (plastic creep): Large accumulation of permanent strain in the post-compaction period, followed by a decrease in the permanent strain rate to a low or constant value [14]. Material might exhibit Range C response after a large number of load cycles.

$${0.045\times {10}^{-3}<\epsilon }_{p-5000}-{\epsilon }_{p-3000}<0.4\times {10}^{-3}$$

Range C (incremental collapse): Accumulation of permanent strain increases significantly with increasing the number of loads cycles, and the permanent strain rate decreases very slowly, leading to the failure of specimen [14].

$${\epsilon }_{p-5000}-{\epsilon }_{p-3000}>0.4\times {10}^{-3}$$

where ${\epsilon }_{p-5000}-{\epsilon }_{p-3000}$ is the difference between the accumulated permanent strain of cycles 3000 and 5000.

Determination of shakedown ranges of pavement materials using shakedown concept is a practical approach to examine the performance of materials under traffic loading. The permanent deformation behavior of material can be classified as either Range A (plastic shakedown), Range B (plastic creep), and Range C (incremental collapse). Range A and Range B response are preferred in the pavement base/subbase layers, while material with Range C response should be avoided to ensure the stability and durability of the pavement system [16].

The permanent deformation behavior of unbound granular materials is a complex phenomenon with several influential factors. Due to the high cost and time-consuming nature of RLT tests, several models have been developed to characterize the permanent deformation characteristics of pavement materials, relating the permanent strain of materials to the number of cycles (N) or a combination of N and stress state parameters through conventional regression analysis [17], [18], [19], [20]. With recent developments in the software and hardware systems, more sophisticated computational approaches could be employed for modeling the deformation behavior of pavement materials. Machine learning techniques are computational approaches that have been utilized for tackling sophisticated engineering problems. Several machine learning techniques have been recently used for solving several complicated problems in various engineering fields [21], [22], [23]. The main advantage of machine learning techniques is their ability to efficiently find the nonlinear relationships between variables in a database. Despite the accuracy and high precision of machine learning methods, their application for evaluating the deformation behvaior of C&D materials is limited to date. Recently, Oskooei et al. [24] used artificial neural networks for prediction of the resilient modulus of unbound and bound C&D materials. Ghorbani et al. [12] performed an experimental and ANN analysis for investigating the effect of temperature on the permanent deformation of recycled concrete aggregate and reclaimed asphalt pavement blends. Application of ANNs for simulating the deformation behvaior of PET/C&D blends and predicting the shakedown behavior in a multi-stage RLT test has not been investigated yet.

Despite several studies have evaluated the potential usage of plastic waste in pavement base/subbase applications, the permanent deformation behavior of PET mixed with demolition waste has not been investigated. Previous studies (e.g., [9], [8]) have focused primarily on the stiffness and strength properties of the plastic waste blends and limited studies have investigated the permanent deformation behavior of plastic wastes mixed with C&D materials. The permanent deformation behavior of PET/C&D blends needs to be investigated in different stress levels before these materials can be readily used in pavement base/subbase layers. Further characterization of PET/C&D blends in various stress levels and precise modeling of the permanent deformation behavior is necessary, which will enable practitioners and industry to utilize them as sustainable pavement materials.

This research evaluates the performance of C&D materials including RCA and CB mixed with up to 7% PET. A suite of multi-stage RLT tests with different stress levels was conducted to investigate the effect of confining stress (σ_c) and cyclic stress (σ_d) on the permanent deformation behavior of blends. Shakedown theory was utilized to classify the deformation behavior of the investigated blends and recommendations were made. Moreover, one of the main motivations of this research is to use a machine learning approach, i.e., ANN, for modeling the permanent deformation behavior and identifying the shakedown ranges of C&D materials blended with plastic waste.