Research on creep and thermal stability of asphalt under outdoor aging (NEA) in Tibetan areas

In order to explore the aging performance attenuation mechanism of 90# and SBS (styrene–butadiene–styrene block copolymer) modified asphalt under the large temperature difference and strong ultraviolet radiation in the Tibetan area, this paper adopts the outdoor natural exposure aging (NEA) condition in the Tibetan area. Scanning calorimetry and thermogravimetric analysis techniques were used to test and analyse the permanent deformation resistance and thermal stability of two asphalts with different aging times (0 month (0M), April (4M) and 8 months (8M)) to study the on-site aging mechanism of asphalt in Tibetan areas. The results show that with the prolongation of NEA time, the non-recoverable creep compliance, polymer area and glass transition temperature of the two asphalts decrease, while the average recovery rate and deformation temperature increase, and the aging degree gradually deepens. Under the special outdoor conditions in Tibetan areas, the permanent deformation resistance and thermal stability of SBS modified asphalt are better than 90# asphalt. The research results explored the aging and decay mechanism of the two asphalts from the macro and micro perspectives, which provided theoretical support for the research and development of anti-aging technology in Tibetan areas and guaranteeing the road performance of asphalt pavement in the whole life cycle.


Introduction
The external conditions of high altitude, large temperature difference, strong ultraviolet rays and low temperature in Tibet will seriously affect the service life of asphalt pavement.Due to the large daily temperature difference, strong ultraviolet rays and heavy loads, the aging rate of asphalt is accelerated, resulting in serious cracking of asphalt pavements in Tibetan areas (M.Cheng et al., 2022;Du et al., 2022;Tibet Autonomous Region Department of Transportation, 2020).
At present, domestic and foreign researchers have done a lot of research on the thermo-oxidative aging and ultraviolet aging mechanism of asphalt.Qiu and Ding (2022) used modulated differential scanning calorimetry (DSC) and low-temperature infrared spectroscopy to study the thermally reversible aging mechanism of asphalt during low-temperature storage.Qu et al. (2022) found that the correlation between indoor thermaloxidative aging and long-term service performance of asphalt is insufficient, and the normative and standardization of ultraviolet aging is insufficient.Based on outdoor aging pavement monitoring is the future direction of asphalt aging research, outdoor aging in Tibetan areas will provide data support for the establishment of indoor simulation and unified research during actual service.P. F. Cheng et al. (2022) studied the aging-healing law of asphalt from the microscopic morphology and chemical structure through dynamic shear rheometer, scanning electron microscope and Fourier transform infrared spectroscopy test.Guo and Liang (2022) reviewed the current status and aging laws of asphalt aging research technology at home and abroad, and found that the current indoor aging simulation test method is relatively simple, and the aging test method under the combined action of multiple factors and the correlation of outdoor field aging are lacking.
However, due to the limitations of geography and test conditions, the research on asphalt aging in Tibetan areas under the conditions of high cold, high altitude, and strong ultraviolet rays according to local conditions is still lacking and in-depth.For a long time, the research on asphalt aging has mostly relied on indoor simulation tests, resulting in a certain degree of disconnection between research and practical application, and did not take into account the coupling factors in the actual environment.The research on the performance attenuation of on-site aging asphalt in Tibetan areas from the macro-micro multi-scale method is an important basic research topic that needs to be carried out, which can serve the engineering practice more effectively.
In order to explore the aging performance decay mechanism of asphalt binder under special external conditions in Tibetan areas, this paper adopts multiple stress creep recovery test, fluorescence microscopy (FM) qualitative and semi-quantitative analysis, DSC test and thermogravimetric (TG) analysis technology from a macro and micro perspective.The aging process of 90# and styrene-butadiene-styrene (SBS) modified asphalt under outdoor conditions ( natural exposure aging (NEA)) in Tibetan areas and the evolution of aging characteristics of modifiers were studied.Focus on revealing the decay mechanism of permanent deformation resistance and thermal stability performance before and after aging of outdoor asphalt in Tibetan areas, and provide technical support for road construction and maintenance in Tibetan areas.

Asphalt
In this paper, the most commonly used 90# base asphalt and SBS modified asphalt 1-C in Tibetan areas were selected for outdoor aging test (NEA).The technical indicators of the asphalt were tested before the experiment, and the test results are shown in Tables 1 and 2.

NEA
NEA can reproduce the coupling effect of light, oxygen, heat and water on asphalt in the natural environment to the greatest extent (Cheraghian & Wistuba, 2020;Sun et al., 2020).The test site is Sangri Town, Qusong County,  Shannan Region, Tibet, at an altitude of 3650 m.The thickness of the bituminous film is 3 mm.In the test, a 35 cm * 25 cm * 1.2 cm rectangular flat-bottomed vessel was used as the aging plate, and the 90# base asphalt and SBS modified asphalt NEA were aged for 0 month (0M), April (4M) and 8 months (8M).Preparation of test device agent specimen is shown in Figure 1.

Dynamic shear rheology test
In this paper, dynamic shear rheology (DSR) is the Bohlin Gemini HR(High-speed Rotonetic) type produced by Malvin Company in the UK, and the asphalt film with a predetermined thickness is rotated with a small floating through the oscillating plate to simulate the actual situation of the load on the pavement (Liu & Zou, 2022;Luo, 2021;Ren & Li, 2022).The experimental parameters are the size of the parallel plate is 25 mm, the spacing of the parallel plate is 1 mm, the stress control mode is adopted, the stress value is 0.12 Pa, the experimental frequency is 10 rad/s, and the temperature range is 46-82 • C. Preparation of test device agent specimen is shown in Figure 2.

Fluorescence microscopic test
In this paper, FM surface morphology of 90# matrix asphalt and SBS modified asphalt aged for 0M , 4M and 8M of NEA in Tibet was observed.The XDY-1 fluorescence microscope (FM) was used to test and photograph  at the magnifications of 100× and 400× .The distribution, phase state and morphological structure of polymer in asphalt can be observed by the auxiliary means of the microscopic morphology of polymer modified asphalt during aging (Bai, 2019;Zhang et al., 2016).Preparation of test device agent specimen is shown in Figure 3.

Thermal analysis test
(1) DSC In this paper, 214 DSC experimental analyser produced by NETZSCH in Germany was used.The sample mass was 10 ± 0.5 mg, the experimental temperature was −50 • C to 50 • C and the heating rate was 10 • C/min.Nitrogen was used as the flow dynamic gas atmosphere protection, with the protector rate of 60 ml/min and the purge gas rate of 40 ml/min.Finally, the glass transition temperature T g was used as the low-temperature evaluation index of aging asphalt (Samieadel et al., 2017).
(2) TG analysis test In this paper, STA449C (DSC/TG) thermal analyser produced by NETZSCH was used.The maximum temperature can reach 1500 • C. The single detection sample is not less than 30 mg.Nitrogen is used as the test gas.The heating rate is 5-10 • C/ min.The test temperature is 23-600 • C. Preparation of test device agent specimen is shown in Figure 4.

Rheological properties
Since the rutting factor G * / sin δ index is derived based on the dissipative energy theory, the total dissipated energy generated during the DSR loading process and the viscous part of the dissipative energy theory, the difference between the base asphalt is very small, it cannot be effectively evaluated.High temperature performance: especially in the intermittent unidirectional load generated by the action of the vehicle during service, the delayed elasticity recovers in an instant, and the viscous part cannot be separated.Therefore, indicators such as G * / sin δ are not suitable for evaluating the high temperature stability of modified asphalt.The multistress creep recovery (MSCR) test can deeply reflect the viscoelastic response of the matrix and modified asphalt with different aging degrees.

Effect of NEA on creep properties of asphalt
Figures 5 and 6 show the time-dependent strain curves of 90# asphalt and SBS asphalt aged 4M and 8M by NEA under the stress levels of 0.1 kPa and 3.2 kPa.The test temperature was 60 • C.
As shown in Figures 5 and 6, the two asphalts were subjected to creep loading for 10 times.After 100 s, the strain of the asphalt gradually increased with the increase of loading and unloading time at the stress level of 0.1 kPa and 3.2 kPa.With the increase of loading and unloading time, it becomes smaller, and the aging degree deepens, 0M > 4M > 8M; the strain under different stress levels changes greatly, and the strain increases with the increase of stress, 3.2 kPa > 0.1 kPa.Comparing 90# asphalt and SBS asphalt, it is found that in each stage of NEA aging, the strain of SBS asphalt is always smaller than that of 90# asphalt, and the strain fluctuation is small, and the temperature stability is good; SBS has good elastic recovery ability and resistance to permanent deformation (Fu et al., 2022;Wang & Wang, 2022).This is because 60 • C has far exceeded the softening point of 90# asphalt, and the viscous component is dominant, resulting in poor deformation recovery ability.As shown in Figure 5(a), 90# asphalt tends to be level in the unloading stage, and as the aging time increases with the extension of time, the amplitude gradually appeared in the unloading stage, and the elastic components of the asphalt increased during aging, and the deformation recovery capacity gradually increased.Therefore, with the increase of NEA aging time, the elastic component of asphalt increases, the strain decreases, the deformation recovery ability is enhanced and its high temperature permanent deformation resistance is improved (Zhang et al., 2022).
The test results of the average recovery rate R and the non-recoverable creep compliance Jnr are shown in Figure 7, which quantitatively analyses the deformation resistance of the two asphalts before and after aging.
As shown in Figure 7, with the prolongation of NEA aging time, the average elastic recovery rates of the two asphalts gradually increased, 0M < 4M < 8M, indicating that aging promotes the transfer of viscous components in asphalt to elastic components.Oxidation reaction occurs under the conditions of heat and oxygen, the molecular weight and the content of oxygen-containing  polar functional groups represented by carbonyl and sulfoxide groups in the structure increase, the stiffness of asphalt increases and the strain recovery rate increases under load.Comparing the two asphalts under different stresses, the recovery rate of 90# asphalt is more affected by aging than that of SBS asphalt.Because the chemical structure and functional group butadiene in SBS decrease with aging, it inhibit aging to a certain extent, and the delayed elasticity is better (Cao & Wang, 2021).The nonrecoverable creep compliance Jnr of asphalt decreases   with the prolongation of NEA aging time, and increases with the increase of stress level, and the anti-cracking performance decreases.
In summary, MSCR can simulate the deformation of asphalt pavement during service more realistically than the rutting factor by considering the delayed elastic deformation of the asphalt external force, especially the modified asphalt is more obvious, and it can effectively evaluate its high stability performance.

Qualitative analysis
The test sample was processed by cold cutting method to obtain the real cross-sectional structure of SBS modified asphalt, which avoided the slippage and segregation of the modifier during the extrusion process of the slide method to a certain extent, as shown in Figure 9(a), imaging with 10× (100×) magnification.
Figure 8 shows the fluorescence mapping of 90# matrix asphalt before and after aging.The asphalt in the fluorescent environment is dark and dark, indicating that there is no polymer that reflects the fluorescence.The observation range of the fluorescence test cannot explain the change of the matrix asphalt (Mu, 2021;Yan, 2021).However, there are a few sparse yellow-green fluorescent spots in the image of outdoor aging for 8M, which may be caused by impurities such as dust and small gravel adhered to the asphalt during the outdoor test.
Figure 9(b-d) shows the fluorescence imaging changes of SBS asphalt after outdoor aging for 0M, 4M and 8M.The yellow-green bright blocks in the image indicate that the SBS modifier swells and absorbs light in the asphalt.The oil content is caused by fluorescent reflection (Li et al., 2016).With the prolongation of aging time, the distribution number and size of modifier particles gradually become smaller; the particle outline gradually blurs; the distribution and size become uniform.In Figure 9(b,c), the small particle modifier in the unaged (0M) asphalt has disappeared, and the large particle modifier is decreasing, which is due to the SBS modification under the sun exposure.The agent has begun to degrade, and the small particle modifier is the first to degrade and disappear; in Figure 9(c,d), the fluorescent imaging distribution and particle size of the asphalt aged for 8M are more uniform, almost no large-sized modifier exists, and the small particle size is more uniform.Particle modifiers also become scarce, because the 4M-8M period is the period from May to September in the Tibetan area, and the period from May to September is when the temperature in the Tibetan area is the highest (the temperature of the asphalt surface can reach over 70 • C), and the time when the ultraviolet radiation is the strongest, SBS modifier undergoes faster and more sufficient degradation reaction under the action of light, heat and oxygen.

Semi-quantitative analysis
In order to more accurately explain the outdoor aging characteristics of asphalt at the microscopic level, semiquantitative analysis was carried out on the fluorescence imaging of SBS modified asphalt before and after NEA aging by using the functions of segmentation, counting and measurement of Image Pro Plus software.
The steps for quantifying fluorescent images with Image Pro Plus are as follows: adjust the image brightness uniformity -contrast adjustment (the best display effect of fluorescent substances) -segmentation (fluorescent substances are identified with special colours, colour change) -measure ( area, per area) -export calculation results (sum per area), the calculated data are shown in Figure 10 and Table 3.
As shown in Figure 10 and Table 3, the area ratio of the SBS polymer gradually decreased with the aging time, and the 8M time decreased from 6.36% to 2.95%, indicating severe degradation; the degradation rate at 4M was 20.75%, and the degradation rate at 8M was as high as 41.47%, the degradation rate nearly doubled.The accelerated degradation rate of SBS at 8M is because from May to September, the temperature and ultraviolet rays are the strongest in the Tibetan area, and the aging is serious, which is consistent with the reason in the qualitative analysis.The degradation of SBS particles is due to the aging of SBS modified asphalt under the action of light, heat, oxygen and other factors.The light components in asphalt, SBS particles degrade with the reduction of light components, and the particles gradually develop from large to small to disappear.SBS mainly plays the role of tackifying and toughening in asphalt.This section explains the degradation of low-temperature performance of SBS modified asphalt after outdoor aging (the degradation of SBS leads to the decrease of toughness), and the macroscopic manifestation is the decrease of ductility.

DSC
DSC is a common thermal analysis method.Its principle is to measure the relationship between the energy difference between the measured substance and the control substance in the process of temperature change and temperature.Then the phase change of the internal composition and structure of the material and the temperature sensitivity of the asphalt are analysed.DSC can measure the characteristic point of phase change temperature.The area of exothermic peak and absorption peak on the curve corresponds to the heat released/absorbed by phase change, and the abscissa and ordinate correspond to the temperature and heat difference, respectively.Take DSC as an example.When the temperature rises gradually and passes the glass transition temperature of the polymer, the baseline on the DSC curve moves towards the endothermic direction (as shown in Figure 11).Point A in Figure 11 is the point where the baseline starts to deviate.Extend the baseline before and after the transition, and the vertical distance between the two lines is the order difference J .At , point C can be found at J/2 , and the tangent from point C intersects with the previous baseline at point B. The temperature value corresponding to point B is the glass transition temperature T g .DSC evaluates its low-temperature performance by analysing the degree of change of asphalt with temperature at the microscopic level.As the temperature increases, the aggregation state of the internal components of the asphalt changes, from glassy state (small deformation) to high elastic state (stable deformation) to viscous flow state (irrecoverable deformation) transition (Jin et al., 2022;Wang et al., 2022).The glass transition temperature aging index T g was introduced to jointly evaluate the low-temperature performance decay degree of different NEA aging times.The test results are shown in Table 4 and Figure 12.

High temperature analysis (TG analysis)
In this paper, the asphalt before and after NEA aging is heated to 600 • C at room temperature, and the initial decomposition temperature Onset, end temperature End, deformation temperature Inflection and other indicators are automatically identified and analysed by proteus analysis, and the high stability performance after aging is evaluated.The test results are shown in Table 5 and Figure 13.
As shown in Figure 13, the TG curve of asphalt can be divided into three stages with the increase of temperature.The first stage: from room temperature to 250 • C, the mass loss rate is almost zero, which is relatively stable; asphalt begins to decompose, and mass loss occurs, depending on the volatilization of light components such as saturates and aromatics; the third stage: 390-530 • C, the main stage of mass loss has almost ended, and the components have changed significantly.
It can be concluded from Table 5 that as the aging degree of NEA deepens, the initial decomposition temperature and deformation temperature of asphalt gradually increase, and the mass loss rate gradually decreases.The initial temperature and deformation temperature of asphalt are higher than 90# asphalt in all stages of aging, and the high temperature performance is significantly better than 90# asphalt.Due to the change of the internal components of the asphalt due to aging, the cracked components are transformed into a stable structure, the molecular weight increases, the force increases, and the high temperature stability improves, which is also consistent with the characterization of more macro indicators such as the increase of complex modulus and the increase of softening point.

Conclusions
In this paper, through dynamic shear rheological test, fluorescence microscopic test, DSC test and TG analysis, the rheological properties of 90# and SBS modified asphalt before and after the aging of outdoor (NEA) asphalt in Tibetan areas are explored from a macro and micro perspective, and the evolution of microscopic properties.The main conclusions are as follows: (1) Under the conditions of special high temperature difference and ultraviolet rays in Tibetan areas, with the prolongation of NEA aging time, the elastic recovery rates of 90# and SBS modified asphalts gradually increased, and aging promoted the transformation of viscoelastic components in asphalt.And the recovery rate of 90# asphalt is more affected by aging than SBS.
(2) The qualitative and semi-quantitative results of FM showed that with the prolongation of NEA aging time, the area of SBS polymer decreased from 6.36% to 2.95%, and the degradation rate reached 41.47%.From the microscopic level, the low-temperature performance attenuation mechanism of SBS modified asphalt after special outdoor aging in Tibetan areas was analysed.(3) DSC test results show that with the increase of NEA aging time ( 0-8M), the T g aging index of 90# and SBS modified asphalt decreased to 0.83 and 0.87, respectively.Under special light and thermaloxidative aging in Tibetan areas, the light components (hydrogen) of the asphalt are detached and transferred to the heavy components (asphaltene), the molecular weight becomes larger, the arrangement becomes tidy, the low-temperature performance of SBS asphalt is better than 90# and the decay rate is also slower than 90#.

Figure 5 .
Figure 5. Creep recovery curve of 90# asphalt with different NEA aging time.(a) Creep recovery curves of 90# asphalt at different aging times under 0.1 kPa stress and (b) creep recovery curves of 90# asphalt at different aging times under 3.2 kPa stress.

Figure 6 .
Figure 6.Creep recovery curve of SBS asphalt with different NEA aging time.(a) Creep recovery curves of SBS asphalt at different aging times under 0.1 kPa stress and (b) creep recovery curves of SBS asphalt at different aging times under 3.2 kPa stress.

Figure 7 .
Figure 7. R and Jn of two kinds of asphalt NEA after aging.(a) Average recovery rate R of two kinds of asphalt and (b) non-recoverable creep compliance Jnr of two kinds of asphalt.

Figure 8 .
Figure 8. Schematic of fluorescence imaging before and after aging of 90# matrix asphalt.(a) Schematic of fluorescence imaging before aging of 90# matrix asphalt and (b) schematic of fluorescence imaging after aging of 90# matrix asphalt.

Figure 9 .
Figure 9. Schematic diagram of fluorescence before and after aging of SBS modified asphalt NEA.(a) Fluorescence image of SBS modified asphalt (glass slide), (b) 0M fluorescence image of SBS asphalt aging, (c) 4M fluorescence image of SBS asphalt aging and (d) 8M fluorescence image of SBS asphalt aging.

Figure 10 .
Figure 10.Schematic diagram of fluorescence before and after aging of SBS modified asphalt NEA.(a) Image Pro Plus handles area images (0M NEA); (b) Image Pro Plus handles area images (4M NEA) and (c) Image Pro Plus handles area images (8M NEA).

Figure 12 .
Figure 12.Schematic diagram of T g (DSC) variation with NEA aging time.(a) Schematic diagram of 90# asphalt DSC and (b) schematic diagram of SBS asphalt DSC.

Figure 13 .
Figure 13.Schematic diagram of TG curves of asphalt before and after NEA aging.(a) TG curve of 90# asphalt and (b) TG curve of SBS asphalt.

Table 2 .
Parameters of SBS asphalt.

Table 3 .
Polymer distribution and area ratio of SBS asphalt before and after NEA aging.

Table 4 .
Changes of asphalt T g , endothermic heat and T g aging index before and after NEA aging.

Table 5 .
Changes of asphalt temperature Onset, end temperature End and deformation temperature Inflection before and after NEA aging.