*3.1. Mixture's Details*

Two mixtures were designed in compliance with a gradation band suggested by Italian technical specifications for semi-porous layers. The SPT mix was produced with 100% pale limestone aggregates. The SPS mix replaced 21% of the natural aggregate with synthetic aggregates (Figure 2).

**Figure 2.** Pale limestone aggregates (**a**), synthetic aggregates (**b**) and transparent polymeric binder (**c**).

The synthetic aggregates were sieved in order to substitute only the natural aggregate particles with the same dimensions (6.3/12.5 mm). Therefore, the gradations of the two mixtures were kept constant in order to have comparable grading curves (Figure 3). After several laboratory trials, the optimum binder content was determined to be 5.5% by weight of the aggregates. In this case, the evaluation of the optimum binder content was carried out in compliance with the Marshall mix design method [40]. The mixing procedure provided for the heating of aggregates at 170 ◦C and the addition of the polymeric binder chips at ambient temperature directly into the laboratory mixer together with the hot aggregates.

**Figure 3.** SPT and SPS particles distributions and gradation limits.

As showed in Figure 3, both the grading distributions fit the reference gradation band and there is no remarkable difference between the two mixtures.

The final aspect of the samples after compaction and its surface texture as well as the porous structure are shown in Figure 4. From a visual analysis the synthetic aggregates seem to be well distributed into the mixture, highlighting no issues in terms of workability and segregation during the mixing and compaction process.

**Figure 4.** Gyratory sample with synthetic aggregates.

#### *3.2. Air Voids Content*

In order to verify the quality of the mix design, the air voids (Av) content (EN 12697-8) [41] was evaluated for the two experimental mixtures. Four gyratory samples were produced for each mix in compliance with EN 12697-31 standard (80 gyrations). The bulk density of the specimens was calculated using the geometrical procedure, as suggested by the EN 12697-6 [42] standard for open graded bituminous mixtures. The results are presented in Table 3.


**Table 3.** Air voids content.

Both mixtures show a significant Av content. It is worth noting that the Italian technical specification suggests a lower Av limit of 16% for semi-porous mixtures. The remarkable porosity of the mixture is probably related to the adopted particles size distribution, which was close the lower limit of the gradation band for the material passing the 1 mm sieve and to the upper limit for the retained material at 4 mm sieve. In this case, the obtained aggregates distribution is more open graded.

The higher Av content for the SPS mixture is probably related to the rounded shape of the handcrafted artificial aggregates.

#### *3.3. Mechanical Characterization: ITS and ITSM*

The mechanical characterization was based on both static and dynamic tests: Indirect Tensile Strength (ITS) and Indirect Tensile Stiffness Modulus (ITSM).

The ITS test is generally useful for evaluation of the strength of cohesion between aggregates and mastic [43]. According to EN 12697-23, a load with a constant velocity of 50 mm/min is applied diametrically to a cylindrical specimen until failure.

The test was carried out using 3 gyratory specimens (80 gyrations) conditioned at 25 ◦C for 4 hours. The results are shown in Table 4. Both experimental mixtures comply with the minimum ITS value suggested by the reference technical specification (ITS ≥ 0.50 MPa). The SPS mix shows better performance despite its higher porous structure. In wider terms, a suitable ITS is reached despite the considerable amount of air voids content in the mixtures.


**Table 4.** Indirect Tensile Strength (ITS) test results.

These results represent also a further validation of the quality of the mastic formed by the polymeric transparent binder and the finest part of the aggregates distribution.

An advanced mechanical characterization was based on evaluation of mixture behavior under dynamic load using the ITSM test carried out on 3 gyratory samples (80 gyrations) for each mixture at 3 different temperatures: 10, 20 and 30 ◦C. According to the EN 12697-26 part C standard, the Modulus is determined through a pulse loading with a rise-time of 124 ms, to generate a predefined horizontal deformation of 7±2 μm in the core of the cylindrical specimen. The tests were carried out on 3 samples conditioned at 3 temperatures in order to verify the thermal sensitivity of the mixtures and how the presence of synthetic aggregates could affect this property.

The average results are presented in Figure 5 and summarized in Table 5.

**Figure 5.** Average Indirect Tensile Stiffness Modulus (ITSM) results at 10, 20 and 30 ◦C.


As overall results, both mixtures show a consistent stiffness at each test temperature.

If 20 ◦C is considered as the reference temperature, there are no substantial differences in stiffness between the two mixtures. The adopted technical specification does not suggest any limitations in terms of stiffness moduli. However, according to the scientific literature and real applications, the achievement of 3000 MPa at 20 ◦C can be considered as a suitable requirement for porous asphalts, considering their relatively weak structure.

In terms of thermal sensitivity, the two mixtures show a different variation in stiffness in relation to test temperature. The SPT mixture would appear to have a mechanical behavior that is more influenced by test temperature than the SPS mixture. Typically, for asphalt concretes an excessive stiffness at low temperatures and a low mechanical response at high temperatures could result in detrimental issues in terms of durability. In light of the above, the experimental mixture shows a positive increase in stiffness at high temperature and this could have a favorable effect in terms of rutting resistance. Still, the SPS mixture shows a stiffness trend line that corresponds to a reduced thermal sensitivity. This might be a consequence of the partial substitution of natural aggregates with synthetic materials. Future testing will assess the level of thermal transmittance for these artificial aggregates in order to validate this speculation.

#### *3.4. Durability Evaluation: ITSR and Cantabro Tests*

Considering the porous structure of the material, the durability evaluation was based on the assessment of the water sensitivity of the mixtures, in terms of reduction in ITS and raveling resistance.

In the first case a set of 3 specimens for each mix were subjected to 10 freeze and thaw cycles, from −20 to 20 ◦C, for 5 days before being tested. In fact, moisture damage can be considered as one of the main forms of pavement deterioration, which is also promoted by the formation of ice [44]. According to well-established literature, high ITS and ITSR values could guarantee a good resistance to moisture damage [45]. According to the EN 12697-12, the reduction in ITS is calculated as the ratio between results obtained in wet and normal dry conditions (EN 12697-12).

The Cantabro test is typically used in Europe for evaluating the raveling resistance of porous asphalt concretes (EN 12697-17). The test enables the estimation of the abrasiveness of porous asphalt, as these mixtures have high air voids, the contact areas between aggregates, which guarantee cohesion of the asphalt concrete, are limited. It is worth noting that it does not reflect the abrasive effect by studded tires. Thus, the cohesion is evaluated in terms of particles loss (PL) when a set of Marshall samples (EN 12697-30, 50 blows per side) is placed in a Los Angeles machine for 300 revolutions, with a speed of 30 revolutions per minute. Four Marshall samples were tested for each mixture. In compliance with the standard, the specimens were stored for 2 days at a temperature of 25 ◦C prior to testing. Table 6 summarizes the results for ITSR and Cantabro tests.


**Table 6.** ITSR and Cantabro test results.

The technical specifications generally suggest a minimum ITSR value equal to 75%. Both experimental mixtures exceed this threshold value. It is worth noting that the reduction in ITS is generally evaluated for samples kept in a water bath (40 ◦C) for 72 hours prior testing. In the case under study, the ITSR results are in line with the suggested lower limit even if the mixtures have been subject to a considerable higher deterioration. The lower ITSR results for SPS are probably related to the higher air voids content that might had a detrimental effect during the freeze and thaw cycles.

In terms of particle loss, there is no significant difference between the two mixtures. The most common Italian technical specification suggests a maximum particle loss (PL) value equal to 20%, for porous asphalt, which is substantially higher if compared to test results. This is a further validation of the quality of the cohesion between particles guaranteed by the polymeric transparent binder considering the remarkable porosity of the two mixtures.

#### *3.5. Functional Properties: Skid Resistance and Vertical Permeability Test*

Surface friction and vertical permeability are important functional properties for porous layers. The friction between tires and road pavement involves two components: adhesion and hysteresis. The first phenomenon is connected to the microtexture of the pavement, which is generally evaluated by means of the skid resistance test [46]. The most common measure of the skid resistance is given in terms of Pendulum Test Value (PTV, EN 13036-4) using the British Portable Pendulum. According to the standard, the frictional force is the force acting tangentially in the contact area and it is measured

as the loss of energy of a standard rubber slider that slides across the test surface. The PTV value in given by the average result of five repetitions for each single test point. The final result is adjusted with specific factors depending on the surface temperature. It is worth mentioning that the test surface must be wetted prior to testing.

As for the permeability, it is the most important property for a porous asphalt. According to the EN 12697-19, it can be evaluated in lab in terms of vertical and horizontal permeability: in this research, the former property was assessed. The vertical permeability is considered as the water flowing in a vertical direction through the specimen thickness. The test procedure imposes that a water column of constant height is kept on the surface of the porous sample and the vertical permeability is evaluated in terms of the amount of water flowing through the sample in a specific range of time using the Darcy's Law.

The results of both tests are shown in Table 7.


**Table 7.** Average Pendulum Test Value (PTV) and vertical permeability results.

In terms of skid resistance, a small difference was found between the two mixtures. The SPS mixture has a higher friction, possibly related to the different surface texture affected by the higher porosity and the different micro and macro texture of the synthetic aggregates. Both PTV values are acceptable but the surface texture needs to be improved if compared to values generally suggested by technical specifications for asphalt pavements (PTV ≥ 50). However, an increase in PTV is expected after a primary polishing of the binder film that covers the aggregates by the traffic. Nevertheless, an optimization of the grading distribution might improve the macro-texture of the material.

No significant difference was found in terms of vertical permeability for the two mixtures. It is worth noting that the reference standard suggests a minimum value of 0.5·10−<sup>3</sup> m/s for traditional porous layers. The obtained values are remarkable, considering that the adopted gradation band is suggested for semi-porous layers. Nevertheless, taking into account the significant porosity of the two mixtures, the air voids interconnection must be improved. Future imaging tests with e.g. Nuclear Magnetic Resonance (NMR) technology will evaluate the inner structure of the samples in order to verify the interconnectivity and tortuosity of air voids.

#### **4. Conclusions**

In the present research, a low impact semi-porous concrete produced with transparent polymeric binder and pale limestone aggregates is proposed. To improve the sustainability of the material, an experimental mixture was produced with the partial substitution of natural aggregates with artificial ones obtained through the alkali-activation of waste basalt powder. The research program provided for a physical and mechanical laboratory characterization.

On the basis of the presented results, the following conclusions can be drawn:

• The adopted particles size distribution and polymeric binder amount allow the achievement of good workability and a higher porosity if compared to traditional semi-porous asphalt concretes. The presence of synthetic aggregates did not affect the mixing procedure and the workability properties of the mixture. The visual analysis of the samples and their inner structure highlighted a correct distribution of the artificial aggregates within the mixture.


In the light of the above, the use of transparent polymeric binder seems to be a viable solution for the production of low impact semi-porous layers for use in urban areas. Future studies will investigate the possible substitution of higher quantities of natural aggregate with synthetic material aiming to the production of 100% synthetic mixtures. Furthermore, the use of a centrifugal granulator can improve the quality of the synthetic aggregates and convert the production from the laboratory to the industrial scale.

**Author Contributions:** Conceptualization, P.T. and C.S.; investigation, P.T.; data curation, P.T. and C.S.; writing—original draft preparation, P.T. and C.S.; writing—review and editing, P.T. and C.S.; supervision, C.S.

**Funding:** This research received no external funding.

**Acknowledgments:** Authors are grateful to CORECOM s.r.l. that supplied the polymeric transparent binder during the whole research.

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