3.1. The Boiling Results
The results of the boiling test are shown in
Table 9. The stripping section of the asphalt film has been marked with a red frame, and the corresponding adhesion grades are shown in
Table 10.
It can be seen from
Table 9 and
Table 10 that with the increase in bio-oil dosage, the overall peeling area of the asphalt film from the aggregates was slightly increased. The low content of bio-oil could improve the adhesion of COBA and aggregate, while the opposite was true for the high content of bio-oil. For instance, when the content of bio-oil was 5%, there was no obvious asphalt film shedding on the surface of the aggregates. When the content of bio-oil was 10%, the slag surface started to show patchy peeling and a small quantity of the asphalt film peeled off from the surface of the limestone, but this peeling area was still smaller than that of base asphalt. While the content of bio-oil was 15%, the peeling area of asphalt film on the surface of limestone was further increased. This is due to the fact that the added bio-oil increases the acid value of asphalt, and the acid molecules in asphalt will bind to the active site on the surface of the aggregate silicate to form alkyl acids, and then crystallize to form adhesions [
23]. However, this acidic compound is easy to hydrolyze in water, which will weaken the moisture damage resistance of the COBA–aggregate. Therefore, during the boiling process, the more serious the hydrolysis process of the accumulated alkyl acids on the surface of the aggregates coated by COBA with high bio-oil dosage, the larger the area of asphalt-film peeling.
Under the action of short-term thermal oxidation, the adhesion between COBA and the aggregates was basically the same as the virgin status, only the adhesion grade of 15% COBA and the aggregates decreased to 3. However, the peeling degree of the asphalt film was aggravated after long-term aging, which led the peeling area to exceed 10%. This is due to the fact that COBA becomes hard and brittle under thermal oxidation, and 15% COBA with a higher bio-oil dosage is more prone to aging. Secondly, the polarity of water is much larger than COBA, and its diffusion ability on the surface of the aggregates is stronger. Therefore, the water can displace the asphalt film from the surface of aggregates more easily, thus reducing the adhesion grade of COBA–aggregate.
Under the same conditions, the adhesion grade of COBA and limestone was the same as that of steel slag, but the peeling positions of asphalt film were slightly different. The steel slag is more hygroscopic and has a rough surface, so the aged COBA is more likely to be peeled off from the surface of the steel slag. Meanwhile, the main mineral phase of limestone is alkaline CaCO3, which can react with acidic compounds in COBA to produce some water-insoluble sticky substances. Therefore, the asphalt film on the surface of limestone is more likely to peel off at the corners and uneven edges.
Since the results of the boiling test are affected by several factors such as the angle of the aggregate, the surface roughness, and the thickness of the asphalt film, only a rough qualitative evaluation of the adhesivity of the COBA and aggregates can be made. To accurately distinguish the relationship of the moisture damage resistance of COBA and aggregates under the same adhesion grade, it needs to be further determined through other tests.
3.2. The Photoelectric Colorimetry Results
In this study, the peeling ratio of asphalt film from aggregates before and after thermal oxidation was calculated, and the results are shown in
Figure 3.
As shown in
Figure 3, the peeling ratio of the COBA film from the steel slag exceeded 100% before and after thermal oxidation. Meanwhile, the same phenomenon was observed for COBA and limestone after thermal oxidation, which does not match the actual outcome. This phenomenon persisted after three parallel tests, which we considered to have been a result of a strong absorption impact by the test materials on the phenol saffron solution, thus interfering with the test results.
Since biomass heavy oil is an agricultural forestry waste and steel slag is an industrial solid waste, few have applied them in road engineering, and related studies have not yet been found. The feasibility of photoelectric colorimetry for evaluating the adhesivity of common asphalt and aggregates has been previously demonstrated. Therefore, for the phenomenon of peeling ratio greater than 100%, this study is focused on analyzing the adsorption of phenol saffron solution by steel slag, limestone, and castor oil-based bio-oil.
From
Figure 4, it can be seen that the adsorption ratio of limestone to phenol saffron after the water bath was 4–5 times higher than before. However, the adsorption ratio of steel slag to phenol saffron was only 6–7%, essentially indicating no adsorption. In addition, after the water bath, the residual concentration of phenol saffron in the solution containing steel slag mixture was lower than that only containing steel slag, demonstrating the adsorption to the phenol saffron of COBA. At the same time, the adsorption amount of COBA to phenol saffron was comparable to that of limestone when the COBA film was not exfoliated from the aggregate. This further confirmed the adsorption effect of COBA on phenol saffron. The peeling ratio was calculated by dividing the adsorption capacity of the asphalt mixture by the adsorption capacity of aggregates after the peeling test. As a result, the peeling ratio of the COBA film from the steel slag exceeded 100%, as shown in
Figure 3.
In summary, the adsorption of phenol saffron by COBA affects the peeling ratio of bitumen from the aggregate surface. Moreover, limestone and steel slag have a large difference in the absorption of phenol safranin. Therefore, the adhesion properties between COBA and different aggregates could not be evaluated directly using photoelectric colorimetry. However, the adhesion of the asphalt and aggregate of the same type can still be determined by observing the color difference between the residual solution and the standard solution, as shown in
Figure 4b,d.
3.3. The Contact Angle Test Results
In this research, the contact angle, polar component, dispersion component, and surface free energy of the COBA and aggregates were measured using the sessile drop method. The results of cohesion work, the adhesion work, the exfoliation work, and the compatibility ratio of COBA–aggregate are shown in
Figure 5.
(1) Surface free energy and cohesion work of COBA
As shown in
Table 8, the surface energy and cohesion work of COBA increased with the increase in bio-oil addition and short-term thermal oxidation. For the virgin samples, the cohesion work of COBA increased by 12.9%, 25.1%, and 31.4%, respectively, compared with the base asphalt when the bio-oil content was increased from 5% to 15%. After short-term aging, the surface energy and cohesion work of COBA increased by 4.0% and 16.4%, respectively, compared with the virgin sample. This indicated that the addition of bio-oil and short-term thermal oxidation could increase the cohesive properties of COBA.
(2) Work of the adhesion of COBA–aggregate
The work of adhesion characterizes the adhesion performance between asphalt and aggregates under anhydrous conditions. The larger the work of adhesion value, the better the adhesion of asphalt to aggregates. According to
Figure 5a,b, for the virgin samples, the work of adhesion between COBA and aggregates had an average increase rate of 5% with the increase in bio-oil content, among which, the adhesion work increased up to 15.7% when the bio-oil content was 15% compared with the base asphalt. This indicated that the bio-oil had a positive effect on the adhesion work between COBA and aggregates. After short-term aging, the effect of bio-oil content on the adhesion of COBA to aggregates was reduced. In addition, the work of the adhesion of base asphalt and 5% COBA to aggregates increased compared with virgin samples. This was because, after short-term aging, the light components in COBA gradually volatilized, and the macromolecular substances began to migrate qualitatively, which increased the polar components. The polar components adhered to aggregates via chemical adsorption, with high strength, which was not easy to be peeled off, so the adhesion was improved. However, the high content of bio-oil made COBA more prone to aging, and more resin were converted into asphaltene under thermal oxidation. Therefore, when the bio-oil content exceeds 10%, the adhesion between COBA and aggregate decreases. This is consistent with the decrease in adhesion grade via the boiling method.
After long-term aging, the adhesion work between COBA and aggregates was lower than that of base asphalt, and decreased by 12.6% on average compared to the virgin samples. This is due to the rapid increase in heavy components such as asphaltene and polar functional groups in COBA caused by long-term thermal oxygen, which makes the asphalt molecules harden. In addition, these components are easily converted into carboxylic acid and other substances under these conditions, making the asphalt brittle, thus leading to a decrease in the adhesion between COBA and aggregates.
In terms of the effect of aggregate type, the adhesion work between different aggregates and COBA increased at a similar rate with the increase in bio-oil content. However, the adhesion work between COBA and steel slag was higher. Before and after short-term thermal oxidation, the adhesion work between COBA and steel slag was 7% and 9% higher than that between COBA and limestone, respectively. The adhesion work between 15% COBA and steel slag was up to 55.16 mJ/m
2, which was 6.9% higher than that between COBA and limestone at the same bio-oil dosage. This was because the content of SiO
2 in steel slag is higher than that in limestone, and the acid molecules in COBA are more likely to react with a high-valence metal salt to form alkyl acids, which accumulate and crystallize on the surface of aggregates to form adhesion [
24]. Moreover, based on the physical adsorption theory, the greater the roughness of the aggregate surface, the more effective the contact area, and subsequently the more favorable to asphalt adsorption. Steel slag is also more conducive to COBA adsorption, forming adhesion because of its porous surface and higher roughness than limestone.
(3) Work of exfoliation of COBA–aggregate
The work of exfoliation refers to the energy released by asphalt during the displacement of water from the surface of aggregates. The smaller the absolute value of exfoliation work, the greater the bonding strength of asphalt and aggregates under water conditions, and the stronger the ability of the mixture to resist water damage. According to
Figure 5c,d, for the virgin samples, the exfoliation work between COBA and aggregates decreases with the increase in bio-oil content. This indicates that bio-oil exhibits the effect of improving the anti-stripping at the interface of asphalt–aggregate. Under the action of thermal oxidation, the work of the exfoliation of COBA increased, probably due to the decrease in internal resin content and increase in asphaltene content in COBA after aging. This accordingly reduces the adhesion of COBA–aggregate, thus weakening its resistance to water damage.
In terms of aggregate type, the exfoliation work of COBA–limestone was much smaller than that of COBA–steel slag, which was only half of it. According to
Table 4, the SiO
2 dosage in steel slag was 8.4 times that in limestone, so COBA interacted with steel slag to form more acidic compounds, which could provide better adhesion. However, carboxylic acid compounds can ionize readily underwater and increase the moisture susceptibility of asphalt concrete, which results in a rapid decrease in the interfacial adhesion [
25]. Also, the silicate in the aggregates and the H in the water produce a coulomb interaction and combine in the form of hydrogen bonds, causing steel slag to demonstrate a stronger hygroscopicity than limestone. In addition, the steel slag itself will have a hydration reaction with water, which can replace water during the hydration process, so the exfoliation work increases and the resistance to water damage stripping decreases.
(4) Compatibility ratio of COBA
The compatibility ratio is used to characterize the compatibility of asphalt and aggregates. The greater the compatibility ratio, the better the adhesion stability of the asphalt mixture. From
Figure 5e,f, the compatibility ratio of COBA and limestone was twice higher than that of COBA and steel slag. The steel slag had a higher hygroscopicity and limestone was less affected by water, with a much lower exfoliation work than that of steel slag. Furthermore, with CaCO
3 the as main component, limestone is more alkaline than steel slag, so it has a stronger acid-base reaction with COBA and better adhesion stability. Of course, the large adhesion work between steel slag and COBA is accompanied by large exfoliation work. This indicated that its adhesion with asphalt cannot work stably in the presence of water, which needs to be further explored.
3.4. The Contact Angle Moisture Susceptibly Results
In this study, the moisture susceptibility of COBA was evaluated by observing the change in the contact angle of COBA before and after water conditioning on the slide surface. Typically, the smaller the contact angle, the better the adhesion of the asphalt to the solid surface [
26]. The contact angle of COBA is shown in
Table 11.
As observed in
Table 11, for the virgin samples, the contact angle between COBA and the glass slide decreased with the addition of bio-oil before water conditioning. It demonstrated the increase in adhesion between asphalt and siliceous aggregates by bio-oil. With the process of water conditioning, the contact angle between COBA and the glass slide increased, indicating the decrease in adhesivity. This is because the affinity of COBA and water for SiO
2 is different under chemically driven moisture damage [
27]. Initially, COBA covers the surface of the dry slide with a low contact angle. But water gradually penetrates into the interface of COBA–slide and replaces COBA on the aggregates, increasing the contact angle between COBA and the glass slide.
After aging, the contact angle between COBA and the slide increased gradually. Under thermal oxidation, the light components in COBA decreased, accompanied by the increase in colloidal components, asphaltene, and other heavy components. As a result, the diffusion of COBA on the surface of the slide decreased and the contact angle increased. Meanwhile, the contact angle of COBA changed more after water conditioning, and the moisture damage resistance also changed significantly. Therefore, the moisture damage resistance of COBA was evaluated by calculating the CAMSI of COBA before and after thermal oxidation. The results are shown in
Figure 6.
The increase in CAMSI indicates a decrease in the adhesivity of asphalt with silica when it is exposed to water [
12]. As shown in
Figure 6, for the virgin samples, the CAMSI of mixtures with 5% and 10% COBA decreased by 7.5% and 4.4%, respectively, compared to base asphalt, while the CAMSI of mixtures with 15% COBA increased by 58.5%. This is consistent with the water-boiling method, demonstrating that low bio-oil dosage can improve the moisture damage resistance of COBA. This is due to the fact that low bio-oil dosage can better interact with asphalt and prevent moisture sub-penetration into the binder, reducing its moisture susceptibly [
28]. In fact, while bio-oil changes the rheological properties of asphalt, it also weakens the durability of its mixture. The presence of a large amount of bio-oil in 15% COBA will cause the acidic compounds inside to diffuse rapidly and interact with the siliceous aggregate to produce carboxylic acid, etc. They will accumulate and crystallize at the interface of the aggregates, and then hydrolyze upon contact with water, which is also the reason for the weakening adhesivity between the aggregates and COBA with a high dose of bio-oil [
29,
30].
Under the effect of thermal oxidation, the CAMSI after short-term aging was the smallest, while the CAMSI after long-term aging was even smaller than the original status. It demonstrated that COBA still has good adhesion property to silica when exposed to water and the short-term thermal oxidation effect has good advantage to improve the moisture damage resistance of COBA. Hung et al. found that the improved moisture damage resistance of bio-asphalt can be attributed to the ability of bio-oil to delay the formation of acidic compounds [
31].
Rajib et al. [
32] proposed that aged asphalt molecules have a higher interaction energy with SiO
2 and thus can form more hydrogen bonds between the interface of the binder and the surface of SiO
2. In addition, the increased colloids and asphaltenes in the asphalt after thermal oxidation also play an important role in increasing the adhesion between COBA and aggregates. In general, the light components of COBA will evaporate after thermal oxygen, and the adhesion will also be slightly reduced. Ābele A et al. [
33] found that using elastomers (SBS, rubber, etc.) to modify the bio-asphalt makes the polymer or rubber absorb the bio-oil to expand, thus improving the high- and low-temperature performance of the bio-asphalt. Therefore, the follow-up research group is also conducting research on the adhesion properties between composite modified bio-asphalt and aggregates.
3.5. FTIR Results
In this study, FTIR was used to compare and analyze the functional groups of COBA, as shown in
Figure 7.
As shown in
Figure 7, no new absorption peaks appeared in asphalt after the addition of bio-oil, but only the intensity of the peaks was different. Therefore, bio-oil was mainly physically dispersed in asphalt. As the thermal oxidation occurred, the area of characteristic peak for COBA at 1705 cm
−1 and 1030 cm
−1 increased gradually. It was found that asphalt is susceptible to thermal oxidation, resulting in the breaking of C=C to form C=O. At the same tine, the H atoms are oxidized to hydroxyl groups, and the sulfur-containing functional groups react with oxygen to form the polar functional group of S=O. These increase the number of oxygen-sensitive functional groups inside asphalt. For example, the bio-oil showed a clear absorption peak of C=O at 1705 cm
−1, and the larger the content of bio-oil, the more obvious the absorption peaks. Therefore, in order to quantify the variation of the oxygen-sensitive groups of COBA,
IC=O and
IS=O were calculated as shown in
Figure 8.
According to
Figure 8, the
IC=O increased 20, 144, and 340 times compared to that of base asphalt when the bio-oil content was increased from 5% to 15% in the virgin sample. However, the
IC=O of base asphalt was only 1.7 × 10
−4, which was extremely low, indicating that bio-oil brought a large amount of C=O to COBA. On the contrary,
IS=O decreased with increasing bio-oil, which is due to the very low content of S in bio-oil and the dilution of S content after blending with base asphalt.
Under thermal oxidation, the growth of IC=O in COBA was similar in both short-term aging and long-term aging, while IS=O grew rapidly in long-term aging. When the content of bio-oil increased from 5% to 15%, IC=O increased by 2.4, 0.32, and 0.19 times compared to the virgin sample, respectively, after short-term aging; and 5.69, 0.85, and 0.31 times, respectively, after long-term aging. IS=O increased by 0.03~0.36 and 2.45~3 times, respectively, in short-term and long-term aging, indicating that the production of S=O in COBA occurs mainly in the long-term aging. By considering the change of various groups in COBA and base asphalt before and after thermal oxidation, it can be assumed that the generation of C=O in COBA is mainly contributed by bio-oil, while S=O mainly originates from the base asphalt.
3.6. Correlation Analysis
As the
IC=O exhibited to have been the most sensitive to oxygen, the relationship of
IC=O to the adhesion of COBA–aggregate was analyzed before and after thermal oxygen aging, as shown in
Figure 9.
As illustrated in
Figure 9, there is a high linear relationship between
IC=O of COBA and adhesion work between COBA and aggregates, and R
2 is mostly around 0.9. This indicates that the change of COBA oxygen-sensitive functional groups before and after thermal oxidation is well correlated with the adhesion property of COBA–aggregate.
IC=O can be used as the evaluation index of asphalt–aggregate adhesion under thermal oxidation. After short-term aging, the R
2 of
IC=O and adhesion work is low. This is likely because after short-term thermal oxidation,
IC=O increased with bio-oil content, and C=O with a higher concentration is easily converted to form alcohols and carboxylic acids, which makes the asphalt brittle. At the same time, C=O is prone to pyrolysis reaction at high temperatures, which makes the COBA molecular chain break, thus leading to asphalt aging and peeling. Notably, the increase in adhesion work between the COBA and aggregates is abrupt at 10% of bio-oil content, and the two follow an inconsistent trend change, so the correlation is low.