**Microwave Healing Performance of Asphalt Mixture Containing Electric Arc Furnace (EAF) Slag and Graphene Nanoplatelets (GNPs)**

#### **Federico Gulisano 1,\*, João Crucho 2, Juan Gallego <sup>1</sup> and Luis Picado-Santos <sup>2</sup>**


Received: 24 January 2020; Accepted: 18 February 2020; Published: 20 February 2020

**Abstract:** Pavement preventive maintenance is an important tool for extending the service life of the road pavements. Microwave heating seems to be a promising technology for this application, as bituminous materials have the potential to self-repair above a certain temperature. As ordinary asphalt mixture has low microwave absorbing properties, some additives should be used to improve the heating efficiency. In this paper, the effect of adding Electric Arc Furnace (EAF) slag and Graphene Nanoplatelets (GNPs) on the microwave heating and healing efficiency of asphalt mixtures was evaluated. Microwave heating efficiency was assessed by heating the specimens using several heating times. In addition, the electrical resistivity of the mixtures was measured to understand its possible relationship with the microwave heating process. Furthermore, the healing rates of the asphalt mixtures were assessed by repeated Indirect Tensile Strength (ITS) tests. The results obtained indicate that the additions of graphene and EAF slag can allow important savings, up to 50%, on the energy required to perform a good healing process.

**Keywords:** graphene nanoplatelets (GNPs); EAF steel slag; asphalt mixtures; microwave heating; self-healing

#### **1. Introduction**

Cracking is one of the most common signs of asphalt pavement deterioration, producing a reduction of the mechanical strength and durability of the road pavement over time [1] affecting driving comfort and safety [2].

Generally, the cracks observed at the pavement surface can be caused by two major mechanisms, the bottom-up cracks and the top-down cracks. The bottom-up cracks are initiated by tensile strains at the bottom of the asphalt layer and the top-down cracks are initiated by surface tensile and shear stresses, environmental effects and ageing [3]. Fortunately, the asphalt mixture is a self-healing material, and if enough energy is applied, original mechanical properties can be partially or nearly totally restored. Such technology can enable an important reduction in the consumption of natural resources, saving aggregates and bitumen that would be used in reconstruction or repair/maintenance actions in the road network. By extending the service life of the current pavement, it may occur an overall reduction of the maintenance interventions, thus saving the corresponding costs and CO2 emissions, as well as minimizing the traffic disruptions caused by such actions [4].

From a molecular point of view, the self-healing phenomenon is due to the wetting and interdiffusion of material between the two faces of a microcrack to achieve properties of the original material [5]. Sun et al. developed a healing function of asphalt material based on molecular diffusion theory [6]. Healing activation energy was found to be a promising parameter for evaluating self-healing ability as if appreciable energy equal to or greater than the healing activation energy exists at the damage faces, the self-healing reaction will start. Molecular diffusion models can only be used for describing the microcracks healing, as in the case of macrocracks, the molecular interdiffusion cannot occur due to a wider gap between the faces. In this case, the capillary flow healing model can be used to describe the self-healing phenomenon. Garcia explained that above a specific temperature, the bitumen behaves as a Newtonian fluid, and can fill the cracks in a sort of capillarity flow [7]. Different types of bitumen exhibit different threshold temperatures for flow, depending on their rheological properties, usually ranging from 30 ◦C to 70 ◦C. The flow behavior index of the bitumen was found to be appropriate to characterize the threshold for the initial self-healing temperature of the bitumen [8–10].

The healing capability of asphalt mixtures depends on several internal and external factors [11]. As for the internal factors, bitumen properties have a strong influence on healing capability. Bitumen with lower flow behavior index not only need less energy for starting healing, but also produce better healing levels [10]. At the microscale level, chemical composition has a strong influence on the healing properties of asphalt [12,13]. Cheng et al. demonstrated the influence of the surface energy on the healing capability of bitumen [14]. According to the findings of these authors, the most efficient healers should have relatively lower Lifshitz–van der Waals components and higher acid–base components of surface energy. The amount of bitumen content in the mixture increases the healing capability of asphalt pavement [15]. Some volumetric properties also influence the self-healing properties [16], as well as the type of mixture. Garcia et al. found that porous asphalt mixtures heal faster than the dense mixture, and provide better healing levels [10]. Between the external factors, rest time, temperature and damage degree are the most influential [9,17]. Many researchers have focused on finding the optimal temperature to maximize the healing effect. If the temperature is too low, bitumen cannot flow through the cracks. However, if the temperature is too high, the healing level decreases, probably due to the expansion of the asphalt mixture, which could cause structural defects in the pavement [17–20].

In the field, due to the continuous traffic flow, usually, the rest periods are not long enough to allow the self-healing to occur, and the pavement temperature rarely reaches the temperature needed for flowing. For this reason, in the last years, researchers have studied several technologies to promote the self-healing process. An example is the capsule healing, in which capsules containing rejuvenator oil are mixed with the asphalt materials [21,22]. When crack damage appears next to the capsules, they open and release the oil. The bitumen will be rejuvenated and the life of the asphalt mixture extended [23]. Another type of technology consists of heating the pavement, through induction or microwave heating, in order to reduce the viscosity of the bitumen and heal the cracks.

In the case of induction heating, electrical currents are induced by adding conductive particles in the composition of the mixture, and the heat is generated by the Joule effect [24–26]. Several additives and respective dosages were studied in order to maximize the conductivity of the mixture, such as steel wool, steel fibers, graphite, carbon black and carbon fibers [24,27–30]. Another heating technique is microwave heating, that was found to be more effective than induction heating to heal cracks in asphalt roads [19]. Microwaves are electromagnetic waves with frequencies ranging from 300 MHz to 300 GHz, and wavelengths from 1 m to 1 mm. In industrial applications, the frequencies 915 MHz and 2.45 GHz are the most commonly used, but for special applications, the frequency 5.8 GHz is increasingly used [31]. When microwave radiation is applied, the polar molecules of the asphalt mixture attempt to line up (polarization) with the alternating electromagnetic field. The inability of this polarization to follow the extremely rapid reversals of the electromagnetic field generates random motion and inter-molecular friction that produces heat [32,33]. Microwave heating basically depends on the strength and the frequency of electromagnetic field, the dielectric properties of the matter, which represents the efficiency of material in absorbing microwave energy, the conductivity losses and some thermal properties [34].

Several authors studied the dielectric properties of asphalt mixture for different applications, such as deicing [35], density measurement [36], recycling [33], and maintenance purposes [34,37]. Since the microwave susceptibility of conventional asphalt mixture is low, some authors have added microwave absorbing additives to the mixture for healing purposes [38], such as steel wool fibers [1,19,39,40], steel slag [2,37,41], steel shaving [42], ferrite [17,18] and carbon nanotubes [43]. Trigos et al. proposed a classification of different aggregates frequently used for pavements construction in terms of microwave heating efficiency [44].

Steel slag is a byproduct of the steel production process and is widely used in road pavements, due to his excellent mechanical properties, in terms of roughness, shape, angularity, hardness, polishing and wear resistance [45]. From the environmental point of view, the use of steel slag allows to reduce the amount of material to dispose of, and therefore, the incorporation in asphalt mixtures permits to convert a waste material into a resource. Additionally, the inclusion of slag does not imply any additional cost because the cost of the slag is similar to the prices of natural aggregates [38]. However, only a few studies focused on its microwave absorbing properties and its use for healing purposes in asphalt mixtures. Liu et al. proposed a method to increase the content of ferric oxide of steel slag particles, improving the microwave heating efficiency of asphalt mixture [37]. Li et al. studied the influence of steel slag filler on the self-healing properties of the asphalt mixture using a fatigue-healing-fatigue test. The results show an enhancement in the healing properties of the mixture containing steel slag [2]. Phan et al. used coarse steel slag and steel wool fibers in the asphalt mixture and evaluated the healing rate through three-point bending test [41]. The main result was that the addition of 30% steel slag increased the healing properties of the asphalt mixture.

Recently, with the advent of nanotechnology, some nanomaterials were used in asphalt mixtures. Several mechanical properties can be improved by adding some nanomaterials, such as nanosilica, nanoclay and nanoiron [46]. Recently, carbon and graphene family nanomaterials have been used for asphalt modification [47], such as graphene nanoplatelets (GNPs). The addition of GNP in the mixture leads to an improvement in flexural strength at low temperatures, better performance at high temperatures [48] and easier compaction [49]. However, only a few studies focused on the microwave heating and healing efficiency of asphalt mixture containing graphene nanomaterials. Li et al. used graphene to improve the microwave heating and healing properties of bitumen [8]. The results showed that graphene provides benefits in terms of heating and healing performance.

The objective of this paper is to evaluate the effect of adding Electric Arc Furnace (EAF) slag and Graphene Nanoplatelets (GNPs) on the microwave heating and healing efficiency of asphalt mixtures. This research is the continuation of the study carried out by Gallego et al. [50], where preliminary results of the heating efficiency of these additives were obtained. The graphene nanoplatelets were incorporated as a binder additive, while the EAF slag was added as partial replacement of the natural aggregates. The asphalt mixtures heating efficiency was evaluated using the ratio ◦C/kWh/kg and, in addition, the electrical resistivity was measured to understand the effect of conductivity losses in the heat generation process. The healing efficiency was studied by applying microwave energy to damaged specimens and evaluating the healing recovery using the Indirect Tensile Strength (ITS) test.

#### **2. Materials and Methods**

#### *2.1. Materials*

A conventional dense asphalt mixture AC20 35/50 (EN 13108-1:2007) was the mixture type selected to conduct the experimental study. The mixture particle size distribution is presented in Table 1. Limestone aggregates, limestone filler, and 35/50 conventional bitumen were the materials chosen for the production of the mixtures. According to the bitumen specification, the temperatures of 165 ◦C and 155 ◦C were adopted for the asphalt mixture production and compaction, respectively. The bitumen content, 4.7% by total weight of the mixture, was previously determined using the Marshall method. The mixing process was conducted using a laboratory mixer (EN 12697-35:2016). Cylindrical specimens

of 100 mm in diameter and approximately 63.5 mm in height were compacted using a Marshall hammer (EN 12697-30:2004) applying 75 blows on each side of the specimen.


**Table 1.** Grading curve of the asphalt mixtures.

Six types of asphalt mixtures were produced in this study: one conventional reference mixture (with no additives), two mixtures with graphene (with 1% and 2% dosage by mass of modified binder) and three mixtures with EAF slag (with 3%, 6% and 9% of aggregate replacement).

Graphene nanoplatelets, commercially designated as GRAPHENIT-XL, were used to make the asphalt mixtures susceptible to microwaves. Regarding its chemical composition, graphene is essentially carbon (96.41%) with traces of other elements, such as oxygen (1.05%), sulphur (0.48%), nitrogen (0.48%), hydrogen (0.07%) and others. The graphene presents a bulk density of 0.04 g/cm3. Similarly to other nanomaterials, the nanoscale of the graphene platelets enables a high specific surface area, thus occupying considerably more volume than conventional macroscopic particles. Figure 1 presents a sample of 2.50 g of graphene nanoplatelets and, by the left side, for comparison, 2.50 g of conventional limestone filler (fraction passing sieve 0.063 mm).

**Figure 1.** Mass of 2.50 g of limestone filler under 0.063 mm (**left**) and graphene nanoplatelets (**right**).

The Graphene nanoplatelets were dispersed in the bitumen matrix by adding the nanomaterial to the bitumen heated at 160 ◦C and applying high-speed mechanical stirring (2000 rpm) during 60 min. Additional details about the mixing process can be found elsewhere [51]. After each bitumen modification (with 1% and 2% graphene), the asphalt mixtures were produced and compacted as described for the conventional mixture.

The EAF slag used in this study is produced in Spain, and its chemical composition is reported in Table 2. After the hydration process carried out by the producer, the free calcium oxide becomes almost zero. This procedure prevents expansion problems associated with the presence of CaO. Two fractions of slag were used as a replacement of limestone aggregates, 0.5/2 mm and 0.063/0.5 mm. Figure 2 presents a sample of 50 g of both fractions. Although many investigations report that the substitution of coarse aggregates improves the mechanical properties of the mixture [45], in this research only fine aggregates were replaced, because this provides more homogeneous heating throughout the mixture, as reported by other authors [52]. Volumetric replacement principle was used in order to take into account the different bulk density of slag and natural aggregates [45].


**Table 2.** Chemical composition of EAF slag.

**Figure 2.** Mass of 50 g of slag, fraction 0.063/0.5 (**left**) and 0.5/2 mm (**right**).

#### *2.2. Bulk Density of the Asphalt Mixtures*

Bulk density of the asphalt mixtures (EN 12697-6:2012) was calculated in order to evaluate the physical properties of the mixtures with and without the addition of slag or graphene. Saturated surface dry (SSD) procedure was applied, in which the specimen is first saturated with water, and then its surface is blotted dry with a towel. The bulk density ρbssd is calculated as:

$$
\rho\_{\text{bessel}} = \frac{\mathbf{m}\_1}{\mathbf{m}\_3 - \mathbf{m}\_2} \times \rho\_\mathbf{w} \tag{1}
$$

where m1 is the mass of the dry specimen in g; m2 is the mass of the specimen in water in g; m3 is the mass of the saturated surface-dried specimen in g and ρ<sup>w</sup> is the density of the water at the test temperature.

#### *2.3. Indirect Tensile Strength (ITS) Test*

The effect of additives on the mechanical properties of the mixtures was evaluated with the Indirect Tensile Strength (ITS) test (EN 12697-23:2018). The selected test temperature was 15 ◦C. In the ITS test, a diametrical load was applied at a constant deformation rate of 50 ± 2 mm/min till the rupture of the specimen. Such loading produces tensile stress through the vertical diametral plane, as shown in Figure 3. To prevent excessive deformation of the specimens, the test was interrupted when the measured load dropped by 20% after the peak load. The Indirect Tensile Strength (ITS), in MPa, was calculated as:

$$\text{LTS} = \frac{2 \cdot \text{P}\_{\text{max}}}{\pi \cdot \text{d} \cdot \text{h}} \tag{2}$$

where Pmax is the peak load, in KN, d is the diameter of the specimen in mm and h is the height of the specimen, in mm.

**Figure 3.** Indirect Tensile Strength (ITS) test.

#### *2.4. Microwave Heating*

To heat the samples of asphalt mixture, a conventional microwave oven (with maximum power of 700 W and a frequency of 2.45 GHz) was used. In this study, the medium power level (350 W) was selected and used through all of the study. The heating efficiency of the mixtures was calculated as follows. The cylindrical specimens were cut into two pieces and then were conditioned at 25 ◦C for 2 h. Then, both pieces were placed in the microwave oven and heated for five heating times: 30 s, 60 s, 90 s, 120 s and 150 s. After each heating time, the two halves of the specimen were separated and an infrared thermometer was used to measure the internal temperature, as the average of eight randomly measurements, as shown in Figure 4. Additionally, the energy consumption during heating was measured with an electricity meter. Linear regression analysis was used to model the relationship between the internal temperature (◦C) of the asphalt mixtures and the total energy consumption during the heating process (kWh/Kg).

**Figure 4.** Internal temperature measurement.

#### *2.5. Electrical Resistivity Measurement*

The electrical resistance of the asphalt mixture was measured with the two-probe method, by using a megohmmeter with 5 ranges (50 V–1000 V).

The asphalt specimen, with a height of about 4 cm, was placed between two copper plate electrodes with dimensions of 15 × 15 cm connected with the megohmmeter, as shown in Figure 5. In order to ensure perfect contact, graphene powder was used to fill the gaps between the plate electrodes and the specimens. The electrical resistivity was calculated applying the second Ohm's law:

$$
\rho = \frac{\text{R} \cdot \text{S}}{\text{l}} \tag{3}
$$

where R is the electrical resistance of each specimen in Ω, S is the electrode-specimen contact area measured in m<sup>2</sup> and l is the thickness of the asphalt sample in m.

**Figure 5.** Electrical resistivity measurement.

#### *2.6. Self-healing Test Procedure*

The self-healing performance of the asphalt mixtures was evaluated as follow. First, each cylindrical specimen was tested under the Indirect Tensile Strength (ITS) test, according to Section 2.3. Then, the specimens were left at room temperature until they reached a temperature of 25 ◦C, and elastic rubber bands were used to tight the specimens before the heating treatment, as shown in Figure 6. A similar approach was used by other authors [53], which used a plastic collar to tight the specimen.

**Figure 6.** Simulation of in situ confining conditions of the asphalt pavements.

Preliminary tests conducted in the laboratory showed that the absence of the confining elastic rubber bands produces an enlargement of the crack during the heating process, due to the collapse of the mixture, that makes impossible the healing process. In situ, this cannot occur due to the lateral confinement of the asphalt mixture in the pavement. This phenomenon can be observed in Figure 7, where the broken specimen (Figure 7a) was heated without the inclusion of the rubber bands, causing an enlargement of the crack (Figure 7b).

**Figure 7.** The broken specimen before heating (**a**), and the enlargement of the crack after the heating without rubber bands (**b**).

The rubber bands were, therefore, used to approximately simulate the in situ confining conditions of the asphalt mixture in the pavement, and to obtain a more realistic measurement of the healing performances. Then, microwave radiation was applied. In order to evaluate the effect of temperature on the healing efficiency, the specimens were heated at different internal temperatures, 40 ◦C, 60 ◦C, 80 ◦C, 100 ◦C. Times required to reach these internal temperatures were found using the linear models obtained as described in Section 2.4. After the heating process, the specimens rested at 25 ◦C for 24 h, and then, after removing the rubber bands, Indirect Tensile Strength (ITS) test at 15 ◦C was repeated in order to evaluate the Healing Rate (HR):

$$\text{HR}(\%) = \frac{(\text{ITS}\_{\text{fin}})}{(\text{ITS}\_{\text{in}})} \cdot 100 \tag{4}$$

where ITSfin is the Indirect Tensile Strength of the sample after the healing process and ITSin is the Indirect Tensile Strength of the sample initially tested.

The schematic representation of the methodology is reported in Figure 8.

**Figure 8.** Flow chart of the healing process.

Furthermore, the analysis of variance (ANOVA) was performed to evaluate the effect of the temperature and the additive content on the healing rates of the asphalt mixtures.

#### **3. Results**

#### *3.1. Influence of Slag and Graphene on the Physical and Mechanical Properties of the Asphalt Mixtures*

The effect of additives on the bulk density of the asphalt mixtures is shown in Figure 9. The values are the average of 12 specimens, and the error bars represent the standard deviation. The reference mixture, without additives, has a bulk density of 2.452 g/cm3. It can be observed that by adding slag to the mixture, the bulk density increases, until 2.481 g/cm3, 2.503 g/cm3 and 2.526 g/cm<sup>3</sup> for mixtures with 3%, 6% and 9% of slag, respectively. This trend is due to the higher specific gravity of the slag respect to natural aggregate, as reported also by other authors [45,54,55]. In contrast, graphene addition has the effect of decreasing the bulk density of the mixtures, until 2.429 g/cm<sup>3</sup> and 2.412 g/cm3 for mixtures with 1% and 2% of graphene, respectively. This effect was probably due to the presence of graphene that increased the viscosity of the bitumen, as any powdered filler incorporated in the bitumen. Therefore, as the mixing and compaction temperatures were kept constant regardless of the content of graphene, for comparative purposes, the compaction was less effective when incorporating graphene.

**Figure 9.** Effect of slag (**a**) and graphene (**b**) on the bulk density.

The effect of additives on the initial Indirect Tensile Strength (ITSin) of the mixtures is shown in Figure 10. The values are the average of 12 specimens, and the error bars represent the standard deviation. It can be observed that the addition of slag or graphene has no important effect on the ITSin of the mixtures.

**Figure 10.** Effect of slag (**a**) and graphene (**b**) on the Indirect Tensile Strength ITSin.

#### *3.2. Influence of Slag and Graphene on the Heating E*ffi*ciency of the Asphalt Mixtures*

The effect of adding slag or graphene to the mixture is an increase in the heating rates, as shown in Figure 11. The higher the amount of additive, the faster the temperature increase with energy. Even the mixture without additives can be heated by microwaves, although more energy, and consequently more heating time, must be applied to reach the same temperature. This means that microwave heating technique can also be used for existing pavements without additives, as also reported by other authors [56]. It can be observed in Figure 11 and Table 3 that the lineal models fit well the data, in terms of R2. Nevertheless, it is interesting to analyze the effect of adding slag or graphene in terms of energy saving. The addition of 3 %, 6 % and 9 % (o/aggregates) of slag allows to save, respectively, 29%, 37% and 45% of the heating energy, respect to the ordinary asphalt mixture, while the addition of 1% and 2% (o/binder) of graphene allows to save 29% and 50% of the heating energy. This improvement of energy efficiency can produce several benefits in terms of CO2 emissions and maintenance costs.

**Figure 11.** Effect of slag (**a**) and graphene (**b**) on the microwave heating consumption.


**Table 3.** Output of the linear regression models.

#### *3.3. Influence of Slag and Graphene on the Electrical Resistivity of the Asphalt Mixtures*

According to other studies [27–30], the electrical resistivity of the mixture slightly decreases with the additive content, until a critical value, called percolation threshold, where resistivity sharply decreases. The effect of adding slag or graphene on the electrical resistivity of the mixture is shown in Figure 12. In the case of slag addition, the percolation threshold is reached approximately between 6% and 9% of slag, when the electrical resistivity passes from 1.5 <sup>×</sup> <sup>10</sup><sup>8</sup> <sup>Ω</sup>·m to 1.7 <sup>×</sup> <sup>10</sup><sup>6</sup> <sup>Ω</sup>·m, corresponding to a reduction of 99%. Nevertheless, in the case of graphene, higher contents should be added to reach the percolation threshold. Comparing these results with the heating models (Figure 11), the contribution of the conductivity to the microwave heating can be analyzed. If the conductivity influenced the microwave heating, the sudden reduction of electrical resistivity (percolation threshold) would have led to a drastic increase in the heating rates. However, observing the heating curves of slag mixtures (Figure 11), the increase of the heating rate (◦C/ kWh/kg) with the slag content is

almost linear. This result can be referred to the fact that at microwaves frequencies, ranging from 300 MHz to 300 GHz, the conductivity contribution to heating is very low, and the heat is produced mostly by dipolar polarization rather than by the current created and the resulting Joule's effect. In this sense, the dielectric and thermal properties of the mixtures should be analyzed in future researches, in order to better understand the microwave heating phenomenon of asphalt mixtures and optimize the heating process.

**Figure 12.** Effect of slag (**a**) and graphene (**b**) on the electrical resistivity.

#### *3.4. Influence of Slag and Graphene on the Healing Properties of the Asphalt Mixtures*

The results of the healing test are shown in Figure 13. The values represent the average Healing Rate (HR) of 3 samples, and the error bars represent the standard deviation. A one-way analysis of variance (ANOVA) was performed to evaluate the effect of the temperature on the healing rate of the asphalt mixtures. Normality and homogeneity of variances assumptions were checked. The analysis showed that the effect of the temperature was significant, F(3,68) = 46.96, *p*-value = 0.000. Post hoc comparisons using the Tukey HSD test indicated that all the means were significantly different from each other (*p*-value < 0.05). Therefore, the effect of the temperature was an increment in the healing rate of the asphalt mixtures. In fact, as described in Section 1, by increasing the temperature, the bitumen reduces its viscosity and flows through the open cracks easier, healing them. At 100 ◦C, for all the mixtures under study, the average HR is 67% (SD = 3.9), while the average HR for the mixtures heated at 40 ◦C is 44% (SD = 2.9). These results are consistent with those obtained by other authors [20,40,42].

**Figure 13.** Effect of slag (**a**) and graphene (**b**) on the Healing Rate (HR)

A one-way analysis of variance (ANOVA) was performed to evaluate the effect of the additive content on the healing rates of the asphalt mixtures. Normality and homogeneity of variances assumptions were checked. The analysis showed that the effect of the additive content was not significant, F(5,66) = 0.15, *p*-value = 0.98. Therefore, the addition of slag or graphene did not produce substantial benefits in terms of the healing rate of the asphalt mixtures. However, the benefits in terms of energy savings are important. In this sense, Figure 14 shows the relationship between the total energy consumption during the heating and the healing rate of the mixtures. These curves should be interpreted as an indicator of the healing efficiency of the energy consumed by the microwave heating technique. The greater the slope of the curves, the higher the healing efficiency. In the case of slag, considerable benefits can be obtained even with the addition of 3% (o/aggregates), while with higher contents, no improvements are achieved. Similarly, in the case of graphene, the improvement of the healing efficiency is obtained with the addition of 2% (o/binder). For example, in order to obtain HR = 60%, the healing efficiency of the mixtures with 3% of slag and 2% of graphene is about double compared to the reference mixture. In this way, approximately half of the energy for pavements maintenance operations would be saved.

**Figure 14.** Effect of slag (**a**) and graphene (**b**) on the Healing Efficiency

#### **4. Conclusions**

In this paper, the effect of adding Electric Arc Furnace (EAF) slag or Graphene Nanoplatelets (GNPs) on the microwave heating and healing efficiency of asphalt mixtures was evaluated. The following conclusions can be drawn:


oscillating electromagnetic fields that excite molecules rather than to eventual electrical currents generated in the mixture and the resulting Joule's effect.

The use of GNPs in the asphalt mixtures is an innovative field, and the results of this work can be a starting point for further investigations in this area. In future research, other types of nanomaterials and the optimum dosage can be analyzed and compared with other traditional materials used in asphalt pavements, such as EAF slag.

**Author Contributions:** Conceptualization, J.G. and L.P.-S.; methodology, J.G. and L.P.-S.; validation, F.G and J.C.; formal analysis, F.G., J.C.; investigation, F.G., J.G. J.C., and L.P.-S.; resources, J.G. and L.P.-S.; writing—original draft preparation, F.G.; writing—review and editing, F.G., J.C., J.G. and L.P.-S.; supervision, J.G. and L.P.-S. All authors have read and agreed to the published version of the manuscript.

**Funding:** Fundación Agustín de Betancourt and Spanish Ministry of Education.

**Acknowledgments:** This investigation was possible thanks to Grant for PhD students by the Agustín of Betancourt Foundation and Grant n. PDI-18-0HXIUI-55-QZMWQL for Foreing Investigation Stays by the Spanish Ministry of Education.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Experimental Study on Photocatalytic E**ff**ect of Nano TiO2 Epoxy Emulsified Asphalt Mixture**

#### **Ming Huang \* and Xuejun Wen**

Shanghai Municipal Engineering Design Institute (Group) Co., Ltd., Shanghai 200092, China; 13719145727@163.com

**\*** Correspondence: huangming@tongji.edu.cn

Received: 14 May 2019; Accepted: 13 June 2019; Published: 17 June 2019

**Featured Application: A new emulsified asphalt mixture, to which a specially applicable epoxy curing system was added, was used in this study; four key influence factors on the photocatalytic e**ff**ect were investigated to guide application of TiO2 to asphalt pavements; and average illumination of underground road surface as a design index for xenon lamp lighting systems was proposed.**

**Abstract:** The two major problems that have plagued urban underground roads since their introduction are the harmful emissions caused by hot mix paving and vehicle exhaust accumulation during operation. In order to solve these two problems at the same time, a new asphalt mixture degrading automobile exhaust, which has the advantage of cold mix and cold-application, was presented and studied. A considerable amount of research shows that the use of titanium dioxide (TiO2) for pavements has received considerable attention in recent years to improve air quality near large metropolitan areas. However, the proper method of applying TiO2 to asphalt pavements is still unclear. The new mixture presented in this article contains epoxy emulsified asphalt as the binder; therefore, how to apply TiO2 in the special asphalt mixture proves to be the main focus. By experimental design, four influence factors on the photocatalytic effect, which are the nano-TiO2 particle sizes, dosage, degradation time, and light intensity, have been investigated. The experimental results showed that the 5-nm particle size of TiO2 is better than 10–15 nm for exhaust gas degradation, especially for HC and NO; with an increase in the amount of photocatalytic material, the degradation of CO and CO2 in the exhaust gas did not increase obviously, while the degradation effects of HC and NO were remarkable; in the 4-h time extended degradation test, the experimental data show that the extended time has little effect on the degradation rate of CO2 and CO, and the general trend is that the degradation of exhaust became significant with the extension of time; while setting a 2-h NO degradation rate as an indicator, to make the index more than 50% or 25%, the average illumination of the road surface cannot be less than 60 lx or 40 lx.

**Keywords:** nano titanium dioxide; epoxy emulsified asphalt; photocatalysis; exhaust gas degradation

#### **1. Introduction**

Urban underground roads have initiated a new and convenient way for rapid growth of vehicle flows in metropolitan areas. They will play a major role in changing urban traffic conditions, reducing noise and the destruction of the urban three-dimensional space. However, they have a fundamental problem, that is, space is relative airtight, which will cause two problems in both the construction and operation periods [1,2]. Firstly, heavy emissions accumulate in underground roads and construction air conditions deteriorate. Secondly, in later stages of operation, there will be higher concentrations of automobile exhaust and more danger to the health of the citizens surrounding the air vents [3].

In order to solve the emission problem in construction, current researchers mostly advocate the use of warm mixing technology. This technology can usually bring the temperature of paving asphalt mixture from 170–190 ◦C down to 135–150 ◦C [4,5], which can slightly reduce the emissions in construction.

In order to solve the vehicle exhaust accumulation problem in the stage of operation, in recent years, researchers in road work have considered vehicle carrier-road pavement materials and developed many new degradable automobile exhaust pavement materials [6–8]. These photocatalysts are dissolved in a solution and sprayed on the surface of the road to be exposed to automobile exhaust and sunlight. However, the durability of these methods is insufficient because of the thin structure thickness [9]. The photocatalyst cannot stay on the road for a long time, and the road will soon lose the function of degrading the tail gas.

Hence, a high-performance nano titanium dioxide epoxy emulsified asphalt mixture, instead of a solution, has been introduced as a new pavement material to solve the two problems simultaneously. In this study, it is tentatively applied to a surface wear layer with a thickness of 1–3 cm. The objective of the current research was to investigate the effects of different influencing factors on asphalt pavement degradation exhaust, taking into account the nano material's particle size, dosage in the binder, degradation time, and light intensity [10–12]. The findings can be seen as important reference indexes in the production of nano titanium dioxide epoxy emulsified asphalt.

#### **2. Materials and Mixture Design**

#### *2.1. TiO2 Powder*

HC, NO and CO compounds can be transformed into salt and water by nano-TiO2 under photocatalysis that is an irradiation by a light source with a wavelength less than 387.5 nm [7,8]. Therefore, we needed to select a scheme to make titanium dioxide more in contact with air and ultraviolet light in the design. In this study, Anatase phase nano titanium dioxide was used as it has the best degradation effect among several known phases [13]. The size range of anatase titanium dioxide is large. In this study, 5 nm and 10–15 nm levels were selected. Their specific surface areas were 240 and 60–100 respectively, purities are both 99%.

#### *2.2. Binder*

The binder of epoxy asphalt was divided into two parts, Part A and Part B. Part A was epoxy resin, and Bisphenol A epoxy resin (type E-51) was chosen in this study [14]. Part B was a mixture of emulsified asphalt, titanium dioxide (TiO2) powder, and the curing system. Table 1 presents the basic properties of Part B, with the test method according to ASTM specifications [15]. The basic properties of E-51 epoxy resin are shown in Table 2. In addition, the curing system in Part B comprises an amine curing agent, compatibilizer, and additives [16].



**Table 2.** Main chemical properties of E-51 type epoxy resin.

#### *2.3. Mixture Design*

All aggregates used in this study were basalt because such aggregates produced by most ore fields exhibit a better shape and strength than others. The filler was limestone, and the vast majority of mineral fillers are made of limestone mainly because limestone powder combines well with asphalt [16], which can produce an effect similar to that of asphalt mastic, thereby effectively reducing the bleeding. The test results of the basic properties of the aggregates are summarized in Tables 3 and 4. Each aggregate was studied separately to fulfill the requirements of the material specifications in China [17,18].

**Table 3.** Aggregate gravity.


**Table 4.** Property index of aggregates.


Table 5 presents the gradation of AC-13, which was referenced from the Technical Specifications [17].

**Table 5.** Gradation of asphalt mixture used in test.


The optimum asphalt contents for the different modified asphalt mixtures have been determined using the Marshall mixture design (optimum asphalt content is 4.2% and the average of air void is 4.2%). Then, all of the above materials were supplied to produce all of the asphalt mixture specimens tested, in 30 × 30 cm rutting test form, using the Marshall design [18].

#### **3. Experimental Design**

#### *3.1. Preparation of Carriers and Mixtures*

The process of adding nano-titanium dioxide into the carrier of asphalt mixture is described as follows: Titanium dioxide is first dispersed in water, and eventually fused with epoxy emulsified asphalt, the mixture and specimens required for the test were prepared by adding aggregate. The process is shown in Figure 1 (part of the process and data are patented technology and appropriate concealments have been made).

**Figure 1.** Preparation process of nano TiO2 epoxy emulsified asphalt and mixtures.

#### *3.2. Optical Parameters of Laboratory Simulated Light Source*

Titanium dioxide needed to be excited under a certain condition of illumination. Xenon lamps which are similar to the solar spectrum can be chosen as a light source. They have 400 nm–315 nm–280 nm wavelengths with a width of 3.10 to 4.43 eV ultraviolet light. The xenon lamp (power 25 W, the luminous flux 3200 lm), which can generate excited electrons and holes, was used as a safe photocatalytic light source [19]. The xenon lamp and the rutting test specimen were placed in the sealed tank of the automobile exhaust analyzer, as shown in Figure 2. For the calculation of the spatial distribution of light intensity, luminaire efficiency, luminance distribution, and shading angle, etc., the spatial arrangement is simplified as a mathematical geometric model (see Figures 3–5 below). The experiments were carried out at nighttime to simulate the dark environment of an underground road.

Three lighting arrangement schemes were used to obtain different photocatalytic reaction effects. Scheme A: The number of xenon lamps: 1, position, top, height H = 0.2 m.

Scheme B: The number of xenon lamps: 2, position, both sides, height H = 0.2 m, elevation angle = 30◦.

Scheme C: The number of xenon lamp: 3, position, both sides + top, height H = 0.2 m, elevation angle = 30◦.

**Figure 2.** Rutting specimen and light source arrangement.

**Figure 3.** Illumination geometric model of Scheme A.

**Figure 4.** Illumination geometric model of Scheme B.

**Figure 5.** Illumination geometric model of Scheme C.

As shown in Figures 2–4, P1 is the maximum point of illumination, P2 is the weakest one, and P3 is the midpoint of the edge. The illumination value needs to be calculated constantly.

According to Reference [20], the illumination of the rutting panel is:

$$E\_{pi} = \frac{I\_{C\prime}}{H^2} \cdot \cos^3 \gamma \frac{\Phi}{1000} M \tag{1}$$

In the formula,

*Epi* –illumination of point P generated by the particular lamp (lx); γ–light incident angle of point P from the particular lamp (◦); *Ic*γ–light intensity value of point P (cd); *M*–maintenance correction coefficient of the particular lamp, usually 0.6–0.7;

Φ–rated luminous flux of the particular lamp (lm);

*H*–the height of lamp light center to the road surface (m).

The single light source illuminance is calculated by Equation (1), while Equation (2) calculates the total illumination generated by several light sources.

$$E\_p = \sum\_{i=1}^{n} E\_{pi} \tag{2}$$

According to Equations (1) and (2), total illuminance of P1, P2, and P3 are calculated respectively. The calculation results are shown in Table 6 (unit, lx):


**Table 6.** Illuminance value of P1, P2, and P3 (unit, lx).

According to the illuminance design of conventional underground roads, luminosity is about 40–50 lx [20], Scheme B is consequently closer to the actual situation. In the following experimental studies, the xenon lamp model of Scheme B will be adopted.

#### *3.3. Influence Factors*

In this section, the impact of different factors on the degradation of vehicle exhaust will be studied, in order to determine the engineering design instructions. There are 4 influencing factors proposed, which are particle size of nano-TiO2, dosage of nano-TiO2, the duration and the illuminance of light [21,22]. Influence factors and related experimental design at different levels are shown in Table 7, and the parallel experiments are conducted three times in each grade. If the coefficient of variation of the result is greater than 10%, it will be removed, and more experiments will be carried out until three valid datasets were obtained. The whole experiment was carried out in an opaque closed room. The xenon lamp specified in the test is used for lighting.

**Table 7.** Experimental design of 4 influence factors.


#### **4. Results and Discussion**

#### *4.1. Comparative Study on Two Particle Sizes*

The 5 nm and 10–15 nm nano TiO2 particles were tested for a two-hour test. The results are shown in Figure 6.

**Figure 6.** The comparison chart of exhaust degradation different particle size mixture.

With the smaller particle size of 5 nm, the degradation effect of exhaust is better than the 10–15 nm size, especially for HC and NO's degradation. The rest results will be presented for nano-TiO2 in particles size of 5 nm.

#### *4.2. Di*ff*erent Nano-TiO2 Dosage*

Nano-TiO2 emulsion (content of 5%) was added during the asphalt emulsification process. Among the emulsions, solid content nano-TiO2 emulsion dosages were 10%, 20% and 30% (a higher content of solid nano-TiO2 cannot be dispersed, and a lower dosage will lead to less photocatalytic effect). This section mainly investigated the exhaust degradation of different dosages of 5 nm nano-TiO2 added in the mixture. The test results of two hours are shown in Figure 7.

**Figure 7.** Exhaust degradation effect in different content of nano-TiO2.

Through the analysis of experimental data, we can conclude that with the increase of the amount of nano-TiO2, the degradation rate of exhaust gas of asphalt mixture tended to increase. However, the degradation of CO and CO2 in the exhaust gas was not obvious. With the increase of the amount of nano-TiO2, the degradation rate curves of the two were almost flat. With the amount of nano-TiO2 increased from 10% to 20%, the degradation rate of NO and HC greatly increased. Relatively, with the content of nano-TiO2 increased from 20% to 30%, the increase rate decreased.

#### *4.3. Degradation E*ff*ect Changes with the Length of Degradation Time*

According to the properties of nano-TiO2 photocatalytic materials, nano-TiO2 will not decrease in the chemical reactions. In this section, as the degradation reaction time is prolonged, the main study target is on the changes of degradation performance on exhaust gas. The exhaust degradation test sustained 4 h. The xenon lamp configuration is Scheme B, and the particle size is 5 nm. The test results are shown in Figure 8.

**Figure 8.** The effects of duration on degradation rate.

From the above experimental results, the overall trend is that the degradation rate of the exhaust gas increases with the extension of the reaction time, and the degradation rate curves of each gas generally agrees with the different amounts of variation of the nano-TiO2. The degradation rate of CO and HC gas did not change much, but for the overall upward trend. From the trend line it can be seen that the three curves will eventually stay at a steady degradation rate, that is, CO2 and CO will be close to 20%, HC close to 30%; the NO concentration will have a significant effect, more sustained and eventually will reach 85%. There is a more obvious linear correlation on the NO curve, accordingly, in the next study, the NO degradation rate is used as the recommended index.

The fact is that when the traffic flows in and out, greenhouse gases or harmful gases such as vehicle exhaust will be maintained at a specific concentration dynamically in a relatively confined space such as underground roads. The degradation effect of photocatalytic pavement will tend toward the best reduction rate as time goes by.

#### *4.4. Di*ff*erent Light Intensity*

The photocatalytic material nano-TiO2 is added in the asphalt pavement. Ultraviolet irradiation is needed as an elementary condition for the degradation reaction. In this section, we continue to use the Section 3.2. light source arrangement; various numbers of xenon lamps were designed to simulate different light intensity to achieve different luminosity on the impact of exhaust degradation. The tests lasted 2 h. The degradation rates of the four kinds of harmful exhaust gases at different light intensities are shown in Figure 9.

**Figure 9.** Fitting curves of degradation rates at different light intensities.

The statistical curves in Figure 9 suggest that illuminance and degradation rate have positive correlation, and the positive correlation tendency is very obvious on HC and NO. After 2 h, for Scheme C, the degradation rate of HC and NO reached 32.8% and 64.71%, respectively, which were considerable. Further, the proposed illumination index of lighting arrangement should meet some requirements (set NO degradation rate at 2 h as an indicator). To make the index more than 50% or 25%, the average illumination of road surface needs to be not less than 60 lx or 40 lx, respectively. The data show that the NO degradation rate of this method is higher than that described in other literature, under the basic similar and convertible conditions [7–9].

#### **5. Conclusions**

In this research program, test schemes were designed for the factors that may affect the performance of asphalt mixture on exhaust gas degradation, and the different influence factors were tested. According to the experimental test results and the statistical analysis findings, the following conclusions can be drawn:

1. The 5-nm particle size of TiO2 is better than 10–15 nm on exhaust gas degradation, especially for HC and NO.

2. The experimental data showed that with an increase of the amount of photocatalytic material, the degradation of CO and CO2 in the exhaust gas did not obviously increase, while the degradation effects of HC and NO were remarkable.

3. In the 4-h time extended degradation test, the experimental data show that the extended time has little effect on the degradation rate of CO2 and CO, and the general trend is that the degradation of exhaust became significant with the extension of time.

4. 2 h NO degradation rate is set as an indicator. In order to make the index more than 50% or 25%, the average illumination of road surface needs to be not less than 60 lx or 40 lx.

**Author Contributions:** conceptualization, X.W.; methodology, M.H.; validation, M.H.; formal analysis, Ming Huang; resources X.W.; data curation, M.H.; writing—original draft preparation, M.H.; writing—review and editing, M.H.; visualization; supervision, X.W.; project administration, X.W.

**Funding:** 2015 and 2018 Technology Program of Shanghai Municipal Engineering Design Institute (Group) Co., Ltd. No. K2015K024 and K2018K081.

**Acknowledgments:** The authors would like to thank He Changxuan (SHMPI), Lv Weimin (Tongji University) and Dr. Zheng Xiaoguang (SMEDI) for helpful and constructive prophase studies.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Life Cycle Assessment for the Production Phase of Nano-Silica-Modified Asphalt Mixtures**

**Solomon Sackey 1, Dong-Eun Lee <sup>2</sup> and Byung-Soo Kim 1,\***


**\*** Correspondence: bskim65@knu.ac.kr, Tel.: +82-10-6205-5348

Received: 8 March 2019; Accepted: 26 March 2019; Published: 29 March 2019

**Featured Application: The application of LCA to NMAM has the potential to guide decision-makers on the selection of pavement modification additives to realize the benefits of using nanomaterials in pavements while avoiding potential environmental risks.**

**Abstract:** To combat the rutting effect and other distresses in asphalt concrete pavement, certain modifiers and additives have been developed to modify the asphalt mixture to improve its performance. Although few additives exist, nanomaterials have recently attracted significant attention from the pavement industry. Several experimental studies have shown that the use of nanomaterials to modify asphalt binder results in an improved oxidative aging property, increased resistance to the rutting effect, and improves the rheological properties of the asphalt mixture. However, despite the numerous benefits of using nanomaterials in asphalt binders and materials, there are various uncertainties regarding the environmental impacts of nano-modified asphalt mixtures (NMAM). Therefore, this study assessed a Nano-Silica-Modified Asphalt Mixtures in terms of materials production emissions through the Life Cycle Assessment methodology (LCA), and the results were compared to a conventional asphalt mixture to understand the impact contribution of nano-silica in asphalt mixtures. To be able to compare the relative significance of each impact category, the normalized score for each impact category was calculated using the impact scores and the normalization factors. The results showed that NMAM had a global warming potential of 7.44563 × <sup>10</sup><sup>3</sup> kg CO2-Eq per functional unit (FU) compared to 7.41900 × 103 kg CO2-Eq per functional unit of the conventional asphalt mixture. The application of LCA to NMAM has the potential to guide decision-makers on the selection of pavement modification additives to realize the benefits of using nanomaterials in pavements while avoiding potential environmental risks.

**Keywords:** nanomaterials; life cycle assessment; nano-modified asphalt materials; environmental impact

#### **1. Introduction**

Asphalt is the most widely used pavement layer in the world. It consists of a binding material called bitumen and crushed or natural aggregates. The mixture of these materials forms asphalt mixtures. Demand for paved roads exceeded the supply of lake asphalts in the late 1800s and led to the use of petroleum asphalts [1]. Asphalt is often used as a shortened form of asphalt concrete which is the material of choice in the pavement sector. In the United Kingdom and the rest of Europe, the term 'bitumen' is used as a synonym for the term 'asphalt binder' while 'asphalt cement' is often used in the United States [2]. Asphalt cement or bitumen is used to bind the aggregates together to provide the required strength and stiffness to transfer vehicular loads. In addition to its strength and stiffness, asphalt pavements offer a damping ability due to the viscous-elastic nature of the bitumen [3]. Consequently, asphalt mixtures are qualified to provide optimal driving comfort as well as flexible maintenance actions. Asphalt pavements are designed to provide maximum performance throughout the design life. Bitumen (asphalt binder) performs two functions: Binding aggregates together and protecting the aggregates from distortions. However, unlike concrete pavements, asphalt pavements experience deformations over short periods of time. This, coupled with increased traffic loads and extreme weather conditions have resulted in asphalt pavement authorities seeking alternative solutions to improve the resistance of the road pavements to the adverse effects of mechanical and environmental loading [4].

Currently, several additives and modifiers produced commercially are used to modify the properties of the asphalt binder. Ref. [3] stated that additives and other modifiers are added in asphalt mixtures to lower mixing and compaction temperatures. This was found to improve adhesion and increase resistance against cracking and rutting. Regarding the viscosity of bitumen, Ref. [5] studied the effects of asphaltene on rheological properties of diluted Athabasca bitumen. Nanotechnology and nanomaterials have recently attracted significant attention from the pavement industry. Nanomaterial application is considered to have the potential to improve asphalt binder properties. As mentioned by Ref. [6], the application of nanomaterials as asphalt modifiers is growing rapidly in popularity due to its unique characteristics that significantly improve the performance of asphalt binder. It has been shown in several studies that the addition of nano-silica in asphalt mixtures improve the oxidative aging property, increases resistance to the rutting effect, improves the rheological properties of asphalt mixture and decreases the interaction between asphalt molecules [7–10]. In addition, Ref. [11] investigated and found that increasing nano-silica content in asphalt mixtures decreases the ductility and temperature sensitivity of the asphalt mixture.

It is becoming increasingly important to explore the full benefits of additives and modifiers on the long-term performance of asphalt pavements. With sustainability in mind, and also embracing the global effort to reduce the environmental impacts associated with these newly perforated materials, being able to make decisions and judge the benefits and environmental friendliness linked to the long-term pavement performance has become important. Consequently, having a life-cycle assessment (LCA) tools available to assess modified-asphalt materials on a life-cycle basis becomes necessary. Due to the concern of global warming and resource depletion, LCAs for different materials and products and systems have gained significant popularity with researchers. LCA studies can help to determine and minimize the energy consumption, use of resources, and emissions to the environment by providing a superior understanding of the systems [3]. LCA studies can help to consider different alternatives if the environmental performance of a particular material or product is not favorable. There have been several studies that attempt to assess the environmental impact of asphalt materials and some studies have also been conducted on asphalt binders modified with additives [12–15]. However, to the author's knowledge, no studies that assess the complete LCA for the production phase of nano-silica-modified asphalt mixtures have previously been published. A new material being used as a modifier, there are uncertainties regarding the environmental impacts associated with nanomaterials. Therefore, it is of paramount importance to investigate the extent to which the use of nano-silica-modified asphalt mixtures for asphalt concrete pavement is beneficial from an environmental perspective.

This study presents the assessment of a Nano-Silica-Modified Asphalt Mixtures in terms of materials production emissions through LCA methodology. The environmental impacts of a conventional asphalt mixture were assessed so that a comparison could be made to understand the impact contribution of nano-silica in the asphalt mixture. In addition, to be able to compare the relative significance of each impact category, the normalized score was computed for each impact category using impact scores and normalization factors. The application of LCA to NMAM has the potential to guide decision-makers on the selection of pavement modification additives to realize the full benefits of the use of nanomaterials in pavements while avoiding potential environmental risks.

#### **2. Literature Review and Definitions**

#### *2.1. Life Cycle Assessment*

LCA is described by Ref. [16] as a tool for systematically analyzing the environmental performance of products or processes over their entire life cycle, which includes raw material extraction, manufacturing, use, end-of-life disposal, and recycling. LCA is described as a 'cradle to grave' method for the evaluation of environmental impacts [17]. In a similar description, Ref. [18] defines LCA as a methodology that quantifies the environmental impacts of a process or a product. In their study, Ref. [19] stated that most of the environmental impacts do not occur in the use, maintenance, and repair of the product but during the manufacturing, transportation, and disposal stages. Ref. [20] claimed that it would be premature to make any claims on the environmental benefits of a particular product or manufacturing process without first considering its consequences in a life cycle context. LCA methodology includes the establishment of an inventory of all types of emissions and waste products [21,22]. LCA studies are conducted in accordance with the specification and standards of the International Organization for Standardization (ISO). The four major components of an LCA study according to Ref. [23] are illustrated in Figure 1. The inventory analysis part is made up of material extraction phase, manufacturing or production phase, use or operational phase and disposal phase. However, it is quite difficult to effectively assess the environmental impact of a product during its in-service life. Therefore, the analysis of this study does not include the operational phase and/or the disposal phase of the inventory analysis.

**Figure 1.** Structure of LCA study.

#### *2.2. Nanomaterials and their Application as a Modifier in Asphalt Mixtures*

Nanotechnology is an emerging technology and is regarded as a key enabling technology due to its numerous associated benefits to many areas of society. Nanotechnology is defined as the use of very small particles of materials (either by themselves or by their manipulation) to create new large materials [24]. The author added that nanotechnology is not a new science or technology, but an extension of the science and technology that has been in development for many years, and is used to examine nature at an ever-smaller scale. Ref. [25] defines nanomaterials as those physical substances with at least one dimension between 1 and 150 nm (1 nm = 10−<sup>9</sup> m). With reference to the European Commission's recommended definition of nanomaterials, Ref. [26] defines nanomaterial as a "natural, manufactured material containing particles, in an unbound state or as an aggregate or as an agglomerate and where, for 50% or more of the particles in the number size distribution, one or more external dimensions is in the size range 1–100 nm". The application of nanomaterials in the field of construction is growing rapidly. Ref. [27] mentioned that nanotechnology is a rapidly expanding area of research where novel properties of materials manufactured at the nanoscale can

be utilized for the benefits of constructing infrastructure. Although some nanomaterials are already being used in the concrete industry, their application as a modifier in asphalt binder has attracted more interest recently. Several experimental studies have been conducted to determine the effect of nanomaterials, especially nano-silica on the properties of asphalt mixtures. Nano-silica materials are used as additives which are applied in small percentages by weight of the asphalt binder to improve the rheological and other properties of asphalt mixtures. Ref. [7] investigated the characteristics of asphalt binder and mixture containing nano-silica and found that the addition of nano-silica has a positive influence on different properties of the asphalt binder and mixture. Ref. [28] also studied the effect of nano-silica and rock asphalt on rheological properties of modified bitumen. In their study, Ref. [29] found that the inclusion of nano-silica reduces the rutting susceptibility of nano-modified asphalt mixtures. Ref. [30] studied laboratory evaluation of composed modified binder and mixture containing nano-silica/rock asphalt/SBS. In a similar experimental study, Ref. [31] found that increasing the percentage of nano-silica increases the Brookfield Rotational Viscosity (RV). Ref. [32] worked on the application of nano-silica to improve asphalt mixture self-healing. In another study, Ref. [33] investigated the effect of nano-silica on thermal sensitivity of hot-mix asphalt. Nano-silica increases the strength or durability of asphalt mixture [34,35]. Refs. [36–40] also made similar studies on the effect of nano-silica on asphalt binder and mixtures. Table 1 summarizes the review of previous studies on the characterization of asphalt binder modified with nano-silica. Regarding the cost of using nanomaterials, Ref. [41] provides the prices for almost all nanomaterials based on the quantity required. For example: precipitated calcium carbonate Nanopowder, 50 nm (100 g = \$45, 1 kg = \$85); nano-silica nanopowder, 60–70 nm (100 g = \$55, 1 kg = \$155); titanium oxide Nanopowder, 20 nm (100 g = \$165, 1 kg = \$468); zinc oxide Nanopowder, 80–200 nm (100 g = \$58, 1 kg = \$168). While some nanomaterials may seem costly, others may be cheap. However, on a large scale, an extensive economic analysis is required to determine the optimum cost for each nanomaterial based on the quantity required.


**Table 1.** Review of previous studies on modification of asphalt binder with nano-silica.

#### **3. Methodology**

LCA methodology was used (as standardized by the ISO in 2006) to assess the environmental impact of nano-silica-modified asphalt mixtures. There are numerous nanomaterials whose effect on asphalt binder and mixtures have previously been evaluated. However, based on the extensive literature review, the common nanomaterials which have been experimentally shown to have a greater impact on asphalt concrete performance include nanoclay and nano-silica. Consequently, nano-silica was used in this study. However, any other nanomaterial (especially nanoclay) which uses a similar production process could give similar results when modified with asphalt binder and materials. Also, the analysis of this study focused on only material extraction and production phases and does not include the operational or the disposal phase. The inclusion of the operational phase in the LCA analysis could change the inference about the conformity of nano-silica-modified asphalt mixtures.

The structure of LCA studies adopted includes goal and scope definition, inventory analysis, impact assessment, and improvement assessment or interpretation stages.

#### *3.1. Goal and Scope Definition*

#### 3.1.1. Goal and System Boundaries

The goal of this study is to assess the potential life-cycle environmental impacts resulting from modifying asphalt materials with nanomaterial (i.e., the environmental impacts of nano-silica-modified asphalt mixtures). Additionally, a comparison is made with the environmental impacts of unmodified asphalt mixture to provide a better understanding of the impact contribution of nanomaterials in asphalt materials to allow for informed decisions to be made. In other words, the extent to which the use of nano-silica-modified asphalt mixtures for asphalt concrete pavement is beneficial from the environmental perspective is evaluated.

Two alternative case scenarios were examined. In CASE 1A, the environmental impact of nano-modified asphalt material was assessed. The use of nanomaterial (nano-silica), asphalt materials, and the production processes of asphalt mixtures were considered. Modification of bitumen with nanomaterial is depicted in Figure 2.

**Figure 2.** Nano-modified bitumen (NMB) production.

In CASE 2A, the environmental impact of asphalt material production, excluding nanomaterial (conventional asphalt mixture), was assessed. The system boundaries which defines the unit process considered in the LCA studies [4] were limited to cover the following life cycle stages in this study: (1) raw materials extraction; (2) transportation of raw materials for a unit product manufacturing; (3) modification and production of asphalt materials in the plant. Transportation of asphalt materials

to the field, use, and the end-of-life were not included. The life cycle stages and key processes of nano-modified asphalt production in the plant are shown in Figure 3. The flow emissions and resource consumption (such as electricity and natural gas) for heating and the production processes were also included in the system boundaries.

**Figure 3.** Key processes of nano-modified asphalt materials.

The results of this study will help practitioners in the asphalt concrete pavement industry to make informed decisions by considering the numerous benefits of nanomaterials (nano-silica) and the environmental impacts resulting from modifying asphalt mixtures with the nano-silica material.

#### 3.1.2. Functional Unit (FU)

The FU is the heart of any LCA studies. The FU is a quantified performance of a product system for use as a reference unit in an LCA study [21] (referring to the Malaysian standards handbook on environmental management). A fixed value must be created and the output results of the environmental impacts of the impact categories depend on this selected FU. In this study, a FU of 1000 kg production of nano-silica-modified asphalt mixtures was assumed.

#### *3.2. Life Cycle Inventory (LCI)*

#### Material Extraction and Production Processes

The life cycle inventory stage is the stage of actual data collection and the modeling of the system product. For the data on material extraction, processing, and production, an openLCA database was used. OpenLCA is an open source LCA tool from GreenDeLTa located in Berlin, Germany. The software uses an Eco-invent 2.2 database and other proprietary databases and produces equally good results compared to other proprietary LCA tools such as SimaPro, Gabi, etc. The software allows the user to import any external database into its platform and it can be used to model any product. For the production process of nano-silica, silica gel and precipitated silica type, the process outlined by Ref. [42] was followed. A 1000 kg production of bitumen and aggregates was assumed. For the input amount of 1kg nano-silica production, the data provided by Ref. [43] were referred to. Additives are often applied in small percentages (1–10%) by weight of asphalt binder. This study used 3% of nano-silica for asphalt binder modification. Therefore, the input amount of 30 kg nano-silica was required to modify the bitumen. Data from Refs. [13,14] were used regarding the energy consumption data per kg of material required for bitumen production, aggregates, and the mixing of asphalt materials at the plant. In other studies, such as the one reported by Ref. [3], the modification of asphalt binder with additives results in an increase in fuel consumption by approximately 15%. Therefore, it was assumed that an increase of 15% in energy for bitumen production was required to modify the bitumen. Hence, to account for the asphalt binder modification with nano-silica in the analysis, an additional 15% increase in energy (fuel) was added to the 0.51 MJ energy for bitumen production. Transportation of nano-silica material to the milling terminal for modification was assumed as 100 km, while the total transportation for bitumen and to the asphalt plant was also assumed to be 100 km and that for aggregates was assumed as 5 km to the mixing plant. In Table 2, the material and energy requirements for the production of nanomaterial (nano-silica) and asphalt materials is summarized.

**Table 2.** Materials and energy requirements for 1 kg unit production of nano-silica and asphalt materials.


#### *3.3. Life Cycle Impact Assessment (LCIA)*

The life cycle impact assessment (LCIA) stage involves analyzing the data to evaluate the contribution to each impact category. LCIA consists of characterization, normalization, evaluation, and weighting depending on the LCIA used. In this study, the Tool for the Reduction and Assessment of Chemical and other Environmental Impacts (TRACI) method version 2.1 (provided in openLCA) was used to calculate the impact category indicator scores TRACI is a software from the US Environmental Protection Agency (EPA), Durham, NC, USA. TRACI uses Equation (1) [44] to determine the impact score for each individual environmental impact category.

$$I^i = \sum\_{xm} \mathbb{C}F^i\_{xm} \times M\_{xm} \tag{1}$$

where: *I<sup>i</sup>* is the potential impact of all substances (x) for a specific category of concern (*i*), *CF<sup>i</sup> xm* is the characterization factor for substance (*x*) emitted to media (*m*) for each impact category (*i*), and *Mxm* is the mass of the substance emitted to media (*m*). OpenLCA version 1.7.4 was then used for modeling the processes in this study.

Finally, to be able to compare the relative significance of each impact category, the normalized score for each impact category was calculated. Normalization is the ratio of the impact score in each category and the estimated impacts from a reference (often called normalization factors). These factors represent the impact produced by an average person in a reference place per year. Equation (2) was used for the computation:

$$NS\_{\bar{i}} = \frac{Env\_{\bar{i}}}{NF\_{\bar{i}}} \tag{2}$$

where *NSi* is the normalized score of impact category *i*, *EnvIi* is the environmental impact result of impact category *i*, and *NFi* is the normalization factor of impact category *i*. Regarding the normalization factors, US 2008 reference data was used, the impact per person-year updated in the research by Ref. [40]. Table 3 provides the details of the normalization factors. The units of four categories: ecotoxicity, carcinogenic, non-carcinogenic, and acidification are different from the reference units first converted to the reference units before computing the normalized score.

**Table 3.** Normalization factors for impact categories based on inventories from the US (2008) and US-Canada [45].


Acidification potential = 1.98 × <sup>10</sup>−<sup>2</sup> SO2/kg substance (multiplied its impact result by this value).

Ecotoxicity potential for rural air = 0.064 CTU eco/kg substance (multiplied its impact result by this value).

Human health cancer potential for rural air = 1.2 × <sup>10</sup>−<sup>7</sup> CTU canc/kg substance (multiplied its impact result by this value).

Human health non-cancer potential for rural air = 3.0 × <sup>10</sup>−<sup>8</sup> CTU canc/kg substance (multiplied its impact result by this value).

#### **4. Results and Discussion**

#### *4.1. CASE 1A: Impact Assessment of Nano-Silica-Modified Asphalt Mixtures Analysis*

OpenLCA version 1.7.4 was used to model and analyzed the environmental impacts of nano-modified asphalt materials and the analysis results are shown in Table 4.


**Table 4.** LCIA results of nano-silica-modified asphalt mixtures per FU.

Increase in the production of the raw materials and/or the FU results in an increase in fuel and energy usage and will cause a significant increase in the impact scores in each category. The environmental performance of 7.44563 × 103 kg CO2-Eq/FU global warming of nano-silica-modified asphalt mixture is better than the results of (Butt et al.) who found the modification of asphalt materials with a polymer to be 44.9 × <sup>10</sup><sup>3</sup> kg CO2-Eq per FU of 1 km by 3.5 km wide asphalt pavement.

#### *4.2. CASE 2A: Impact Assessment of Unmodified (Conventional) Asphalt Mixture Analysis*

The analysis of unmodified (conventional) asphalt materials was needed to better understand the environmental implication of modifying conventional asphalt with nanomaterials. The results of the analysis are shown in Table 5.


**Table 5.** LCIA results of unmodified (conventional) asphalt materials per FU.

Any increase in the production of raw materials or a change in the FU will result in an increase in the impact scores in each category and vice versa.

The modification of asphalt materials with nanomaterials results in an increase in environmental impacts, which is clear when comparing the results in Table 5 with that in Table 4. Across all impact categories, there is an increase in the impact scores. This fact is reinforced by Ref. [4] when the authors found that using Ethylene-Vinyl-Acetate (EVA) polymer as a modifier agent leads to a deterioration of the life cycle profile of the pavement compared to unmodified asphalt binder. However, the deterioration of the life cycle environmental profile with nano-modified asphalt materials is insignificant. Specifically, there was only a 0.4% increase in global warming, 0.8% increase in respiratory effects, 0.009% increase in ozone depletion, 0.98% increase in eutrophication, 1.0% increase in human health carcinogenic, 0.7% increase in photochemical oxidation, 0.96% increase in human health non-carcinogenic, 0.72% increase in ecotoxicity, and 0.85% increase in acidification. This means the modification of asphalt materials with nanomaterials (nano-silica) causes more impacts in human health carcinogenic than other impact categories. Apart from ozone depletion, the modification of asphalt materials with nano-silica contributes fewer impacts in global warming per 3% by weight of asphalt binder production of nano-silica.

#### *4.3. Computation of Normalised Score*

Table 6 and Figure 4 show the normalized score in each impact category of nano-modified asphalt materials. According to Ref. [46], by inspection, large values of normalized scores as compared to the total are classified as worse performing impact categories, while those with small normalized scores of approximately less than 2% of the total are classified as better performing impact categories. Table 6 shows that nano-modified asphalt materials only perform significantly better in four impact categories: photochemical oxidation (0.0217 pts/FU), ecotoxicity (0.0634 pts/FU), ozone depletion (0.2323 pts/FU), and global warming (0.3102 pts/FU).

However, to fully understand when and how an impact category is classified as either better or worse performing, a logarithmic scale criterion was used. This was especially useful in situations where there existed large variation in the normalization scores. It is argued by Ref. [47] that dimensionless data is more appropriately plotted on an arithmetic scale to clearly understand where the data points lie (better or worse trend). On a logarithmic scale, the center of gravity (where the eye is drawn) lies at the geometric mean, where the line starts at 1 and not 0. Hence, applying the logarithmic scale plot (see Figure 4), all the impact categories below the 1pts line are referred to as ZONE 1 (better performance zone). Hence, it can be said NMAM performs better in five categories: global warming (0.3102 pts/FU), ozone depletion (0.2323 pts/FU), eutrophication (0.6779 pts/FU), photochemical oxidation (0.0217 pts/FU), and ecotoxicity (0.0634 pts/FU).


**Table 6.** Normalized score per FU of the impact categories for NMAM.

Small values are better.

**Figure 4.** Environmental performance (normalized score) of NMAM.

All the impact categories above the 1 pts line are referred to as ZONE 2 (worse performance zone). NMAM performs worse in this zone in 4 categories: respiratory effects (36.9556 pts/FU), human health carcinogenic (5.8258 pts/FU), human health non-carcinogenic (182 pts/FU), and acidification (41.0101 pts/FU).

The worst performance in acidification, which is the increase in hydrogen ions (H+) concentration within the environment as a result of the presence of acids, can be attributed to the sulfuric acid used in the production of nano-silica and the cause of sulphur dioxide and nitrogen oxides released during transportation of the materials and including asphalt materials. As mentioned previously, the modification of asphalt materials with nanomaterial causes only 0.4% per unit increase in global warming. This is because carbon dioxide (the main cause of global warming) is released during the production of bitumen, aggregates, asphalt mixing, and also during transportation. In short, the fact that the modification of asphalt materials with nanomaterial causes just less than or equal to 1% increase in impact score across all impact categories suggests that modifying asphalt materials with nanomaterials does not cause an unreasonable risk to the environment. However, the results of this study using nano-silica does not conclude that all other nanomaterials may have very low impact. The impact on the environment and the combined impact when modified with asphalt materials depend on the production process of the nanomaterial. Therefore, it is expected that some nanomaterials may have a more negative environmental impact.

#### **5. Conclusions**

LCA is a tool that helps to assess the environmental impacts of materials and products so that decisions can be made not just on the benefits of using these materials but also considering their environmental contributions (especially to climate change and human health). This study assessed a Nano-Silica-Modified Asphalt Mixtures in terms of materials production emissions through the Life Cycle Assessment methodology (LCA), and the results were compared to a conventional asphalt mixture to understand the impact contribution of nano-silica in asphalt mixtures. The results showed that NMAM had a global warming potential of 7.44563 × 103 kg CO2-Eq per FU as compared to 7.41900 × <sup>10</sup><sup>3</sup> kg CO2-Eq per FU of unmodified asphalt mixture. The study also computed the normalized score for each impact category and the results showed NMAM performs better in five categories: global warming (0.3102 pts/FU), ozone depletion (0.2323 pts/FU), eutrophication (0.6779 pts/FU), photochemical oxidation (0.0217 pts/FU), and ecotoxicity (0.0634 pts/FU). NMAM performs worse in four categories: respiratory effects (36.9556 pts/FU), human health carcinogenic (5.8258 pts/FU), human health non-carcinogenic (182 pts/FU), and acidification (41.0101 pts/FU). The modification of asphalt materials with nano-silica causes less than or equal to 1% per unit increase in impact score across all impact categories. The application of LCA to NMAM has the potential to guide decision-makers on the selection of pavement modification additives to realize the benefits of nanomaterials in the pavement while avoiding potential environmental risks. Additionally, this study has shown that even though the modification of asphalt mixtures with nano-silica results in an increase in fuel consumption, it does not cause an unreasonable risk to the environment nor does its application as a modifier results in significant deterioration of the life cycle environmental profile. However, future research is required by considering the analysis of the whole life cycle for nano-modified asphalt materials using different nanomaterials as a modifier to confirm that nanomaterials are sustainable materials.

**Author Contributions:** Conceptualization, S.S.; Methodology, S.S.; Software, S.S.; Validation, S.S.; Formal Analysis, S.S.; Writing–Original Draft Preparation, S.S., D.E.L., B.S.K.; Writing–Review & Editing, B.S.K.; Funding Acquisition, D.E.L., B.S.K.

**Funding:** This work was supported by the National Research Foundation (NRF) of Korea grant funded by the Korea government (MSIT) (No.NRF-2018R1A5A1025137).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Applied Sciences* Editorial Office E-mail: applsci@mdpi.com www.mdpi.com/journal/applsci

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18