Experimental investigation and multi-objective optimization of fracture properties of asphalt mixtures containing nano-calcium carbonate

https://doi.org/10.1016/j.conbuildmat.2021.122876Get rights and content

Highlights

  • The fracture toughness in modes I and II and two mixed modes I/II was investigated.

  • The effect of four NCC contents and two bitumen types was explored.

  • Prediction models of KIf and KIIf were provided using MVR, GMDH and GP methods.

  • Optimization of NCC percentages were done to maximize KIf and KIIf simultaneously.

Abstract

The low temperature fracture is one of the most important challenges in asphalt mixtures, which, if not paid, will lead to high maintenance costs. Therefore, researchers are looking for different materials to enhance the fracture behavior of mixtures. As the asphalt surface is affected by different types of loading throughout their operational lifetime, this study explores the impact of nano-calcium carbonate (NCC) on the fracture behavior of mixtures in different fracture modes. For this purpose, semi-circular bending (SCB) tests were applied to specify the fracture toughness. Two bitumen types of PG 64-22 and PG 58-28 were modified with 1%, 3%, 5% and 7% NCC. Finally, the fracture toughness of samples in pure mode-I, pure mode-II and mixed-mode I/II was investigated at −10 °C. Moreover, the prediction models of multivariate regression (MVR), group method of data handling (GMDH) and genetic programming (GP) were provided to present the best model with higher accuracy in order to obtain optimum NCC content with two objectives of KIf and KIIf (the fracture toughness in modes I and II). The results indicated that NCC had a notable influence on the fracture toughness. However, in each mode and bitumen, the additive percentage that is associated with the highest fracture toughness was different. From the fracture tests, the most optimal percentage of NCC was determined between 3% and 7% and 3% to 5% for mixtures made with bitumen types of PG 64-22 and PG 58-28, respectively. Among the various models, GMDH had the greatest R2 so that the R2 amount of GMDH for KIf and KIIf was 98.68% and 99.02%, respectively. The two-objective optimization results showed that 4.17%, 3.62% and 6.29% NCC were the best optimal amounts to maximize KIf and KIIf amounts simultaneously for all mixtures and mixtures made with bitumen types of PG 64-22 and PG 58-28, respectively.

Introduction

Nowadays, with the increase in maintenance costs as well as increased traffic loads on asphalt pavements, the attention of researchers is changed on reducing breakdowns, including rutting, fatigue cracks, low-temperature cracks and so on. The above-mentioned breakdowns reduce the safety and service level of roads during the lifetime of pavement operation and impose a lot of costs on governments [1], [2], [3], [4]. So many studies have been conducted to reduce or retard the aforementioned distresses. One of the most degradation and failure modes in the asphalt pavement, especially in cold areas, is low-temperature cracks, in which high amounts of costs are paid for repairing the road pavements annually. Two principal parameters in the start and propagation of crackings are traffic loadings and thermal stresses in asphalt pavements [5]. Temperature variations can cause tensile thermal stresses in asphalt pavements, thus the asphalt may be put in mode I deformation (crack opening). Nevertheless, traffic loads may cause various deformation combinations of tension (mode-I) and shear (mode-II) based on the vehicle wheels’ location related to the cracks [6].

As asphalt pavements must today have the durability and stability needed to withstand the stresses of traffic loads and heat stresses, various additives have been used to enhance the mechanical characteristics of asphalt mixtures. Recently, the utilization of different nano-materials, including nano-titanium, nano-silicon dioxide [7], nano-clay [8], nano-zinc oxide [9], carbon nano-fiber [10] and so on has been extensively increased. The properties of additives added to bitumen must be such that they can be compatible with the chemical composition of bitumen and asphalt. This adaptation depends on the composition of base asphalts, polarity, particle size, the molecular structure of additives, interfacial interactions between asphalt-additive and so on. The type of additive, preparation temperature and modification steps are also effective in improving asphalt performance [11], [12], [13].

Many studies have been conducted on the fracture behavior of asphalt samples. Behbahani et al. (2013) explored the impact of various additives, including polyphosphoric acid, anti-stripping agent, styrene-butadienestyrene, sasobite and crumb rubber on the mode-I fracture of asphalt samples. They illustrated that these additives increased the fracture toughness of mixtures at a low temperature and using sasobite and crumb rubber resulted in the maximum increment of the fracture toughness value in pure mode-I (KIf) amounts [14]. Aliha et al. (2014) examined the impact of asphalt properties on the fracture behavior of asphalt samples in mixed-mode I/II using asymmetric semi-circular bending (SCB) samples. They illustrated that mixtures constructed with bitumen PG 64-22 had higher peak loadings compared to bitumen PG 52-28 and so, greater fracture toughness was achieved at the start of first fracture increase. Also, the effective fracture toughness amounts in the mixed-mode (Keff) were less than the fracture toughness amounts in pure mode-I or pure mode-II (KIf and KIIf) [15]. Ameri et al. (2016) explored the impact of carbon nanotubes (CNT) at five different contents of 0.2% to 1.5% on the fracture energy of asphalt mixtures by the use of SCB samples. Results indicated that CNT enhanced the fracture resistances of mixtures, especially at greater CNT contents [16]. Mansourian et al. (2016) studied the impact of sasobite and jute fiber in four different percentages on the low temperature fracture behavior of asphalt samples at −20, −10 and 0°. The finite element method was performed to achieve the geometry parameters of SCB samples in pure mode I, pure mode II and mixed-mode I/II loadings. The fracture resistance was then calculated. They illustrated that jute fiber enhanced the fracture resistance in pure mode I and mixed-mode with a greater tension proportion and the optimal percentage of jute fibers was approximately 0.3% [17]. Fakhri et al. (2018) examined the fracture energy of asphalt mixtures in mixed-mode I/II by the use of SCB samples at a moderate temperature. The impact of bitumen and aggregate types and air voids on the fracture behavior was explored. They indicated that these factors could significantly influence on the fracture resistance of mixtures at the intermediate temperature [18]. Ziari et al. (2019) examined the impact of three rejuvenators, such as Cyclogen, Rapiol, and waste cooking oil (WCO), on the asphalt fracture behavior at 25 °C. Also, they explored the impact of short and long-term aged samples on the fracture behavior of recycled asphalt mixtures. They showed that aging had a negative impact on asphalt fracture resistance and asphalt samples containing WCO had the greatest reduction in fracture energy after aging [19]. Pirmohammad et al. (2019) explored the impact of CNT and nano Fe2O3 at four various contents on fracture characteristics of asphalt mixtures by the use of SCB samples subjected to mixed-mode I and II loadings. They illustrated that both additives remarkably enhanced the fracture characteristics of mixtures. However, the fracture resistance of asphalt samples was more enhanced by CNT than nano Fe2O3. Also, samples in mode-I loading had the greatest enhancement in the fracture resistance. Results also indicated that the appropriate percentage of CNT for modifying asphalt mixtures was 1.2%. [20]. Pirmohammad et al. (2020) explored the carbon and kenaf fibers on the fracture resistance of asphalt samples under modes I and II and mixed mode through SCB test at a low temperature (-15 °C). Results indicated that both the additives enhanced the fracture toughness of mixtures. Using carbon fiber (as a superior material) up to 42% increased the fracture toughness in mode-I. Also, increasing the shear mode’s proportion at the crack front of mixtures resulted in a reduction of the positive impact of the additives [21].

For determination of the optimum amounts, one may create a mathematical relationship firstly which can be obtained by system identification models such as regression, neural network and evolutionary algorithms [22]. In multi-objective optimization (MO) problems, there is a set of optimal solutions, called Pareto fronts [23]. For solving these problems, the genetic algorithm can be used as an evolutionary algorithm. The non-dominated sorting genetic algorithm II (NSGA-II) method which was formed according to a genetic algorithm can properly deal with MO problems [24].

Various researches have been performed by the use of the NSGA-II method in pavements that were often in pavement management field [25], [26], [27], [28], [29], as well as other fields in pavements [30], [31], [32], [33]. But limited studies have been performed to explore the exact optimum amount of an additive (i.e., NCC) with two objectives of KIf and KIIf simultaneously. In this study, a precise amount of NCC was obtained to improve the fracture behavior of asphalt samples. On the other, various researches have been done on the utilization of nano-materials in asphalts. But limited researches have been performed on the examination of fracture toughness of asphalt mixtures containing nano-calcium carbonate (NCC). So in this research, the effect of NCC as a bitumen modifier with weight percentages of 1%, 3%, 5% and 7% was investigated on low-temperature asphalt fracture behavior in mode-I, mode-II and mixed-mode I/II using SCB samples made with two bitumen types of PG 64-22 and PG 58-28 and linear elastic fracture mechanic (LEFM) method. Also, statistical analysis and prediction models of KIf and KIIf were performed using multivariate regression (MVR) analysis and two neural network methods of group method of data handling (GMDH) and genetic programming (GP) in order to determine the best method with greater accuracy and lower error, and then by the use of the best method relation, NCC additive content was optimized to maximize KIf and KIIf values simultaneously.

Section snippets

Fracture tests and finite element (FE) analysis

Various types of tests used to determine the fracture resistance of asphalt samples can be SCB [17], [18], [20], [21], [34], disk-shaped compact tension (DC (T)) [35], [36], [37], [38], [39], edge-notched disc bend (ENDB) [40], [41], [42], indirect tension test (IDT) [43], [44], [45], [46] and TTI upgraded overlay tester [47], [48], [49], [50], [51]. Nevertheless, some of the aforementioned tests have their own weaknesses and faults. However, the SCB sample looks to be a good candidate sample in

Aggregate

The applied aggregate in this research was limestone. The aggregate gradation was based on ASTM standard that the nominal and maximum sizes of aggregate are 1.25 cm and 1.9 cm, respectively. The aggregate gradation is represented in Fig. 3. Moreover, the physical properties of aggregate are illustrated in Table 1.

Bitumen

In this paper, two bitumen types, PG (64-22) with 60/70 penetration grade and PG (58-28) with 85/100 penetration grade, were used as base bitumens, the characteristics of which are

Modeling methods

MVR, GMDH and GP models were used in this study to present the fracture toughness models in modes I and II, and finally, the optimum additive content was represented in the optimization procedure. MVR was used for investigating the relation of independent and dependent parameters and comparing new models with classical ones and for further exact predictions, GMDH and a type of genetic algorithm called GP were applied.

Results and discussions

The results of fracture toughness tests performed in pure modes I and II and two mixed modes loadings on SCB samples containing 1%, 3%, 5% and 7% NCC as well as unmodified mixtures considering two bitumen types of PG 64-22 and PG 58-28 at −10 °C are presented in this section. Moreover, statistical analysis and prediction methods of KIf and KIIf were performed using MVR analysis and two neural network models of GMDH and GP for determining the best model with greater accuracy and lower error, and

Conclusion

In this research, the effect of NCC as a bitumen modifier with weight percentages of 1%, 3%, 5% and 7% was investigated on low-temperature asphalt fracture behavior in pure mode-I, pure mode-II and two mixed-modes I/II using SCB samples constructed with two bitumen types (PG 64-22 and PG 58-28) and then by selecting the best model relationship for two-objective optimization of KIf and KIIf, the optimal NCC content was determined. The main findings of the research are:

  • -

    The fracture toughness of

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Ahmad Ghasemzadeh Mahani: Data curation, Investigation. Payam Bazoobandi: Data curation, Writing - original draft. Seyed Mohsen Hosseinian: Methodology, Writing - review & editing. Hassan Ziari: Methodology.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (79)

  • H. Ziari

    The effect of rejuvenators on the aging resistance of recycled asphalt mixtures

    Constr. Build. Mater.

    (2019)
  • S. Pirmohammad et al.

    Laboratory investigations on fracture toughness of asphalt concretes reinforced with carbon and kenaf fibers

    Eng. Fract. Mech.

    (2020)
  • H. Jia

    LID-BMPs planning for urban runoff control and the case study in China

    J. Environ. Manage.

    (2015)
  • Z. Luo et al.

    Robust design approach for flexible pavements to minimize the influence of material property uncertainty

    Constr. Build. Mater.

    (2019)
  • C.M. Stewart et al.

    Effect of specimen thickness on the fracture resistance of hot mix asphalt in the disk-shaped compact tension (DCT) configuration

    Constr. Build. Mater.

    (2018)
  • B. Behnia et al.

    Self-healing of thermal cracks in asphalt pavements

    Constr. Build. Mater.

    (2019)
  • S. Son et al.

    Fracture properties of asphalt concrete under various displacement conditions and temperatures

    Constr. Build. Mater.

    (2019)
  • A. Bahmani

    An extended edge-notched disc bend (ENDB) specimen for mixed-mode I+II fracture assessments

    Int. J. Solids Struct.

    (2020)
  • H. Motamedi

    Evaluation of temperature and loading rate effect on fracture toughness of fiber reinforced asphalt mixture using edge notched disc bend (ENDB) specimen

    Constr. Build. Mater.

    (2020)
  • P.J. Haghighat Pour et al.

    Evaluating mode I fracture resistance in asphalt mixtures using edge notched disc bend ENDB specimen with different geometrical and environmental conditions

    Eng. Fract. Mech.

    (2018)
  • M. Ameri

    Mixed mode fracture resistance of asphalt concrete mixtures

    Eng. Fract. Mech.

    (2012)
  • M. Ameri

    Cracked asphalt pavement under traffic loading – a 3D finite element analysis

    Eng. Fract. Mech.

    (2011)
  • H.C. Rhee et al.

    Mixed-mode stress intensity factor solutions of a warped surface flaw by three-dimensional finite element analysis

    Eng. Fract. Mech.

    (1987)
  • S. Forth

    Three-dimensional mixed-mode fatigue crack growth in a functionally graded titanium alloy

    Eng. Fract. Mech.

    (2003)
  • R. Sezavar et al.

    New model of moisture susceptibility of nano silica-modified asphalt concrete using GMDH algorithm

    Constr. Build. Mater.

    (2019)
  • A.H. Alavi

    Formulation of flow number of asphalt mixes using a hybrid computational method

    Constr. Build. Mater.

    (2011)
  • H. Behbahani

    Improving the moisture performance of hot mix glass asphalt by high-density polyethylene as an asphalt binder modifier

    Int. J. Sustainable Build. Technol. Urban Develop.

    (2019)
  • N.M.V. Gilani

    Prediction and pareto-based multi-objective optimization of moisture and fatigue damages of asphalt mixtures modified with nano hydrated lime

    Constr. Build. Mater.

    (2020)
  • S.M. Hosseinian et al.

    Investigation of moisture sensitivity and conductivity properties of inductive asphalt mixtures containing steel wool fiber

    Adv. Civil Eng.

    (2020)
  • N.M.V. Gilani et al.

    Presentation of a new deicer with the least moisture and fatigue failures in asphalt mixtures

    Arabian J. Sci. Eng.

    (2021)
  • C. Ye et al.

    Study on road performance of nano-SiO2 and nano-TiO2 modified asphalt

    New Build. Mater.

    (2009)
  • M. Sadeghnejad, et al. Experimental evaluation of fatigue life of asphalt mixture modified with nano zinc oxide. In 7th...
  • H. Behbahani

    Experimental fracture toughness study for some modified asphalt mixtures

    Advanced Materials Research

    (2013)
  • M. Ameri

    Investigation of fatigue and fracture properties of asphalt mixtures modified with carbon nanotubes

    Fatigue Fract. Eng. Mater. Struct.

    (2016)
  • S. Pirmohammad et al.

    Experimental investigation of fracture properties of asphalt mixtures modified with Nano Fe2O3 and carbon nanotubes

    Road Mater. Pavem. Des.

    (2019)
  • M.H. Shojaeefard et al.

    CFD simulation and Pareto-based multi-objective shape optimization of the centrifugal pump inducer applying GMDH neural network, modified NSGA-II, and TOPSIS

    Struct. Multidiscip. Optim.

    (2019)
  • C.A.C. Coello

    A comprehensive survey of evolutionary-based multiobjective optimization techniques

    Knowl. Inf. Syst.

    (1999)
  • K. Deb

    A fast and elitist multiobjective genetic algorithm: NSGA-II

    IEEE Trans. Evol. Comput.

    (2002)
  • A. Khavandi Khiavi et al.

    Multiobjective optimization in pavement management system using NSGA-II method

    J. Transp. Eng., Part B: Pavements

    (2018)
  • Cited by (28)

    View all citing articles on Scopus
    View full text