Bitumen is a commonly used material for construction of flexible and semi-rigid pavements. Crude oil, from which bitumen is refined, is one of the natural resources. According to the sustainability policy, it is important to replace limited raw material with the one from renewable resources, e.g., vegetable-based material. Vegetable-based materials can be applied independently, fully or partially replacing traditional materials or supporting recycling of currently used materials.
Plants are valuable source of materials that can be used to produce binders, liqueurs, or binder solvents. Among the plants that can be used for this are the trees and shrubs from which natural resins are derived or oilseeds to obtain oils.
One of the bio-oil applications can be replacement of organic origin fluxes containing volatile substances. Fluxed asphalts are used in the technology of asphalt mixtures for the production of cold-mixed asphalt and for the repair and maintenance of pavements [1
], or they can be used as asphalt substitutes or rejuvenators [2
]. The experience of using rapeseed oils and animal-origin oils shows that it is possible to apply them to pavement repairs as bitumen fluxes, but bleeding and binder drain down in the pavement can potentially occur [4
]. As the aforementioned implies, it is important to select the right amount of bio-oils for the bitumen and to investigate their properties in particular with regard to the climate in which this technology is used.
Another application of vegetable additives is the rejuvenation of reclaimed asphalt pavement (RAP), which is constantly striving to increase RAP’s content while maintaining pavement durability. When using high RAP content, it is necessary to renew the properties of the RAP-contained-bitumen by using refreshing additives. As previously shown [5
], bio-oil can successfully play the role of such rejuvenators. Vegetable oils added to aged bitumen allow the reduction of the asphaltene content in the binder, but do not change its colloidal structure. It has been found that the addition of too high an amount of vegetable oil can lead to a reduction in the thermostability of the bitumen binder and to the increase of the rutting potential of the mixture [13
]. However, by optimizing the composition of binders containing bio-oils it is possible to improve the fatigue life of bitumen and asphalt mixtures, as well as to reduce the bitumen brittle fracture temperature (in single-edged notched bending (SENB) test) and thermal stress temperature (in the thermal stress restrained specimen test (TSRST)) [6
]; such rheological characterization is important in bitumen modification with bioadditives, polymers, and rubber from recycling or acids [15
Using vegetable origin additives, in particular oils, it should be remembered that most of the current applications are intended to soften or flux the bitumen. This is a convenient solution due to the good mixability of vegetable oils with bitumen, a homogeneous product with reduced stiffness is obtained. Especially in the works [4
] there are indications of restrictions in too high a content of these fluxes due to the potential problem of pavement properties destruction in a form of a permanent deformation. The problem of too high liquefaction of the bitumen binder can be reduced by polymerizing oils in the bitumen binder. It has been found that methyl esters of fatty acids, obtained by transesterification from vegetable oils, may be used as bitumen flux for this purpose [18
]. In this case, the hardening of the binder is obtained not by evaporation but by crosslinking of the flux in the presence of oxygen.
Oxidative polymerization of vegetable oils is a free radical chain reaction, includes initiation, propagation, and termination steps [19
]: Initiation RH → R• + H•; Propagation R• + O2
→ ROO•, ROO• + RH → ROOH + R•; Termination ROO• + R• → ROOR, R• + R• → RR
Oxypolymerization leads to the crosslinking of their structural units [20
]. The efficiency of crosslinking depends on the number of double bonds and their position in the aliphatic chain of fatty acid. Vegetable oils comprise a variable number of double bonds depending on the composition. The composition of the oil and, hence, its reactivity to oxypolymerization is genotype-specific and, therefore, variable and dependent on the climatic conditions [25
Previous studies have shown that the liquid products obtained from oxidation of rapeseed oil methyl esters can be used as a bitumen flux [27
]. Fluxed road bitumen containing this oxidation product increased its consistency over time, had a high ignition temperature, and displayed satisfactory adhesion to the aggregate. The limited amount of double bonds in rapeseed oil esters does not over-stiffen the asphalt mixture. It allows for a two-stage production process and full use of bio-oil potential. In the first production stage the bitumen is liquefied and refreshed, and in the second stage (service life) polymerization of the oil in the bitumen occurs. This type of two-stage technology can be used in sustainable road surfaces without compromising its performance [29
The aim of the work presented in this paper was to examine the structural changes occurring during hardening bitumen with a bio-flux additive.
2. Materials and Methods
Two types of road bitumens (35/50 and 70/100 grades) produced from Uralic crude oil and vegetable origin flux were used for the study. This bio-flux was prepared in a special oxidation process. The material for oxidation was rapeseed oil methyl ester (RME) obtained by transesterification of rapeseed oil. Some of the physicochemical properties are summarized in Table 1
Cobalt acetate (Co(CH3COO)2•4H2O) was applied as a catalyst of oxidation while cumene hydroperoxide (C9H12O2) was used as a promoter of the reaction.
2.2. Production Procedure
Methyl esters of rapeseed oil were oxidized at 20 °C or 90 °C in a 500 mL reactor, depending on the composition (Version A or Version B, respectively). The composition and production parameters are shown in Table 2
and the reactor view is shown in Figure 1
The glass column reactor (1000 mL) was filled with ca. 300 g of rapeseed oil methyl ester. The air inlet was at the bottom of the reactor and air was dispersed through sintered glass. The air flow rate used in the experiment (500 L kg−1 h−1) increased the air-oil phase volume by about 50%. The catalyst was added to the raw material in 0.1 wt % amounts (in terms of cobalt content) before it was subjected to oxidation. Catalytic oxidation was also performed in the presence of cumene hydroperoxide (1%/wt) used as a promoter of oxidation reactions. The liquid oxidation products were collected for analysis every 0.5 h. The volatile products were trapped in a scrubber with trichloroethylene which was placed in a dewar vessel filled with a cooling mixture.
Oxidized RME in the presence of reaction promoters was stored at 5 °C in a sealed container without light and oxygen. Modifications of bitumen were carried out by heating the binder to 150 °C in a laboratory oven. Then a bio-flux (at room temperature) was added to the hot bitumen and stirred with a glass baguette for about half a minute—methyl esters of rapeseed oil were well homogenized with the bitumen. No component separation was observed during the modification process. Test samples were taken immediately after modification, and the remaining binder was poured onto a flat tray in a thickness of approximately 1 mm for further conditioning over time.
Samples were conditioned under room conditions. After 28 days, the binder was collected cold with a heated spatula and test specimens were formed.
2.3. Analysis Methods
Concentration of organic peroxide in liquid oxidation products was assessed on the basis of the peroxide value determined according to the ISO 3960 standard. The acid value was tested according to the ISO 660 standard. Iodine amount was determined according to the ISO 3961 standard.
Viscosity of the flux was measured using a Brookfield DV-II+ Pro apparatus (Brookfield Engineering Laboratories, Middleborough, MA, USA). Flash point of the flux was determined using a Marcusson flash-point apparatus (Labor Muszeripari Muvek, Esztergom, Hungary).
Road bitumen modified with bio-additive was subjected to a dynamic viscosity test at 60 °C using a Brookfield apparatus (according to EN 13302:2010) and tests with a dynamic shear rheometer MCR 101 (Anton Paar GmbH, Graz, Austria) were conducted according to AASHTO T 315 (Standard Method of Test for Determining the Rheological Properties of Asphalt Binder Using a Dynamic Shear Rheometer (DSR)). Tests in DSR apparatus were conducted in a set of two parallel 25-mm diameter plates, gap 1 mm, in a temperature range between 0 °C and 100 °C, with strain between 0.1% and 12%, depending on the test temperature.
The main test method to evaluate to chemical changes during bitumen hardening was Fourier transform infrared spectroscopy (FTIR), combined with attenuated total reflection (ATR). ALPHA FTIR device was used with the diamond ATR crystal (Bruker Corporation, Billerica, MA, USA). Infrared spectroscopy was applied to the analysis of original bitumen, as well as the bitumen with bio-flux. The fractional composition of bitumens (asphaltenes, resins, aromatics, and saturates) was determined by the thin-layer chromatography with a flame-ionization detection (TLC-FID). For the analysis an IATROSCAN MK-5 (IATRON Laboratories, INC., Tokyo, Japan) apparatus equipped with Chromarod was used.
Catalytic oxidation of the rapeseed oil methyl ester results in a consistency increase, which should be attributed to the disappearance of the most reactive double bonds after two hours of oxidation. This finding is confirmed by the decrease in the iodine value. The observed fall in the content of double bonds in rapeseed oil methyl ester upon oxidation gives evidence that some part of these bonds was engaged in oligomerization and/or was oxidatively decomposed. The viscosity weight is about 1.3 and 1.8 times that of the raw material, which substantiates only a partial oligomerization of the molecules of unsaturated fatty acid methyl esters. Oligomers are formed mainly at the stage of peroxide generation. It can be assumed that peroxides are intermediate structures in the oxyoligomerization of unsaturated fatty acid methyl esters. A higher viscosity of the liquid oxidation products is attained when the reaction is carried out in the presence of cumene hyperoxide. When the flux containing this hydroperoxide is heated during flash point determination, cumene hydroperoxide decomposes (its half-life temperature is decreased by the cobalt catalyst), and the hydroxyl radicals formed initiate scission of the double bonds. This results in the formation of low molecular compounds with a flash point lower than those of the esters.
Oxidized rapeseed oil methyl ester is a reactive flux, and hardening is possible once mixed with bitumen. After 28 days of conditioning, structural changes are confirmed by effective formation of oxygen organic compounds in fluxed binders. The absorption bands of hydroxyl, aldehydes, ketones, ether, and carboxylic group are evidenced in the FTIR spectra. The hardening of the binder after application is obtained by further polymerization of the fluxes’ vegetable-based components during consumption of molecular oxygen from the air. The reaction schemes for the formation of higher molecular and decomposition products are presented in Figure 7
The kinetic reaction of oxidative polymerization in the bituminous medium depends on the diffusion of oxygen and it is slower in binders with higher consistency. The observed intensity bands of the oxygen functional group are more noticeable in the spectrum of fluxed 70/100 bitumen than for harder 35/50 binder.
Similar to the FTIR observation are the results of the fractional analysis of the fluxed binders. After 28 days of conditioning the increased concentration of resins suggests higher molecular polarity in fluxed binder. The changes in the fractional composition with reduction of the asphaltenes content improves the colloidal stability of the fluxed bitumens after stabilization.
Based on the rheological curve shown in Figure 6
(dashed lines), changes in the complex modulus curve (G*) and phase angle (δ) can be observed for the binders exposed to air for 28 days. Exposure to the air of the thin layer of bitumen causes the oxypolymerization reaction. It can be seen that both asphalt 70/100 and asphalt 35/50 are partially reconstructed for the original properties of the bitumen. This can be determined by analyzing the position of the binder curves after 28 days (dashed line), which runs between clean bitumen and bitumen with added bio-flux tested immediately after mixing. The shift of G* and δ curves for bio-fluxed bituminous binders towards the clean bitumen curve demonstrates overlapping polymerization in the modified bituminous binder under the influence of oxygen from the air.