198 American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) ISSN (Print) 2313-4410, ISSN (Online) 2313-4402 © Global Society of Scientific Research and Researchers http://asrjetsjournal.org/ Analysis of Activation Energy and MSCR for Oxidized Asphalts and Their Impact on the Rutting of Pavements Ana-Sofia Figueroa-Infante. IC. Ph.D.* Universidad de La Salle, Carrera 2#10-70, Bogotá Post Code 111711, Colombia Email: afigueroa@unisalle.edu.co Abstract There have been numerous studies focused on the effect of water on asphalt mixtures, the mineralogy of the aggregates, mastic, and voids. In this research, the effect of water on binder is evaluated in the context of chemical changes, and how such changes impact the material rheological response of asphalt mixtures. This phenomenon has impact on the durability of pavement insomuch as water affects our roads due to prolonged periods of rain and high relative humidity; a typical characteristic of tropical countries. The procedure applied in this research was to submerge 2 mm thick sheets of asphalt in water (of constant controlled properties) for twenty-one (21) months. During that time, the progressive changes in chemical properties, physical properties, and performance of the asphalt were tracked. The results presented in this article are focused on the analysis of the asphalts’ activation energy (and the generated change during the period of submersion) in relation to its response to the Multiple Stress Creep and Recovery (MSCR) test: one of the rheological tests that effectively represents the asphalts’ behavior under different stress and strain levels. Keywords: Flexible Pavement; Oxidized Asphalt; Activation Energy; MSCR; Moisture Damage; Rutting. 1. Introduction Damage in pavements caused by moisture adversely impacts their functionality, decreasing serviceability and increasing maintenance costs significantly. Evidence of this damage is stripping that is related to cohesion and adhesion [1] . ------------------------------------------------------------------------ * Corresponding author. http://asrjetsjournal.org/ American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 199 Adhesion failure is the perfect separation between two surfaces at the interface. Cohesion failure occurs due to internal damage in one of the materials at the interface, usually generating detachment, [2]. The difference between adhesion and cohesion failures is shown in Figure 1. Figure 1: Schematics Showing Adhesion and Cohesion Failure Past research has shown that asphalt exposed to water for an extended period of time changes its chemical composition [3]. For example, after only three months of exposure, an increase in Carboxyl and Sulfoxide functional groups was observed, generating a structure susceptible to fatigue and rutting. To determine the rate of increase in Carbonyls and Sulfoxides, it is necessary to find the activation energy, which is the minimum energy necessary for a change or chemical reaction to occur in a material. Pavements exposed to traffic loads and environmental effects, such as rain or humidity, gradually deteriorate. The intensity of such deterioration depends on: pavement structure, material properties, the difference between adhesion and cohesion failures, and the response of each of the components of the asphalt mixture. In other words, mastic, voids, aggregates and asphalt within the asphalt mixture all contribute to deterioration, however, each one does so in different ways and by differing amounts. Therefore, it is the goal of this research to extend the useful life of pavements through studying these effects. One of the benefits of modern pavement engineering is the possibility to analyze asphalts and (according to their response at different temperatures, loading frequencies, and deformations) predict their performance in the asphalt mixture. Although the role of aggregates is essential, the analysis of binders through some tests such as MSCR, as in [4,5,6], indicates the susceptibility that pavement can have with respect to effects generated by traffic, such as rutting 2. Materials and methods 2.1. Test Plan This experiment included not only the analysis of the properties of eight samples of asphalt with different levels of oxidation, but also wheel tracking tests in order to measure the performance of each mixture manufactured with oxidized asphalt. See Table 1. 2.1.1. Asphalt The changes in the mechanical properties of the binder after water conditioning for extended periods of time (i.e., S0, S3, S6, S9, S12, S15, S18 and S21 months) were studied in this research. Sheets of 80-100 (1/10 mm) penetration grade asphalt (of 2mm thickness) were conditioned resting on 4mm thick sheets of glass. The length and width of each sample was 45x45 cm (as shown in Figure 2). These samples were submerged in potable water at 16°C during 21 months in different containers made of natural glass fiber to avoid any water American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 200 contamination. The test plan for this experiment is indicated in Table 1. The rheological analysis was replicated 10 times with 200 measurements for each test. The chemical and physical measurements were replicated 6 times for each test. Table 1: Test Plan (Asphalt Properties Analysis) Figure 2: Placement of the asphalt submerged in containers with water: (a) samples of 2mm thick sheets of asphalt on glass (b) placement of the samples (c) submerged samples This article presents the changes that occurred in the asphalt in regard to Activation Energy, the MSCR and its performance. The analysis charted of the asphalt was trimestral during the complete period of immersion (21 months). 2.1.2. Water In such an experiment, where potable water is used, an important consideration is water quality and the magnitude of suspended and dissolved particles. To decide on water quality, a comparison was made between laboratory test results and data from the rainwater database of the Hydrology Institute of Meteorology and Environmental Studies-IDEAM from Colombia between 1999 and 2010. The results of this comparison indicated that the acidity, conductivity, dissolved oxygen, and sulfate content of rainwater was not significantly different from potable water. The properties of water used for asphalt submersion are shown in Table 2. 2.1.3. Asphalt Mixture The asphalt mixture was made based on the Superpave® process, level 1 and the maximum nominal size of the Water Control Physical Properties Rheological Properties Chemical Properties Dynamic Properties Constant Variable Variable Variable Variable Dissolved Oxygen Penetration at 25°C(1/100mm) Complex Modulus|G*|(Pa) Asphaltenes (%) Wheel Tracking Sulphates Softening Point (T°C) PG max (°C) Resins(%) Dynamic Modulus Chlorides Penetration Index PG (PAV) (°C) Saturated(%) Trapezoidal Fatigue Hydrogen Potential-PH Ductility (100+cm) PG (water) (°C) Aromatic(%) Conductivity Flaming Point (T°C) Jnr,3.2 (1/kPa) Carboxyls Turbidity Ignition Point (T°C) Elastic Recovery Sulfoxides Total Alkalinity Brookfield Viscosity at 135°C(Pa.s) Traffic level,(Heavy) Ic-Colloidal Instability Index Acidity Specific Gravity Is-Colloidal Clasification Hardness Binder from S0* to S21** *Original condition of asphalt **Submerged aspahlt (S). The number indicates how many months the aspahlt was exposed to the water oxidation a b c American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 201 aggregate was 19 mm. The results are shown in Table 3. Table 2: Properties in the water at 16°C Table 3: Asphalt Mixture Design (maximum size 19mm) Asphalt Content (%) Air Voids (%) VMA (%) VFA (%) 5.7 4.1 14.2 71.6 2.1.4. Wheel Tracking Test In this experiment, 13 tons of force was applied to the asphalt mixtures through a wheel and a resultant pressure of 0.662MPa was calculated at the point of contact. The rotation of the wheel was set to 42 r.p.m above the briquette (of dimensions: 300mm x 300mm x 50mm) at 60 ° C. See Figure 3. Figure 3: Wheel Tracking Test at 60°C Plastic deformation increased dramatically as the oxidation time of the asphalt increased, see Table 4. Figure 4 shows the total disintegration of the asphalt mixture for each period of oxidation. 1 2 3 #DIV/0! 27 9,111 1 0,08 #DIV/0! #DIV/0! 27 1,369 0 0,18 #DIV/0! #DIV/0! 27 12,000 2 0,15 #DIV/0! #DIV/0! 27 7,070 0 0,03 #DIV/0! #DIV/0! 27 50,733 7 0,13 #DIV/0! #DIV/0! 27 0,911 0 0,17 #DIV/0! #DIV/0! 27 14,000 2 0,12 #DIV/0! #DIV/0! 27 8,667 1 0,16 #DIV/0! #DIV/0! 27 31,556 6 0,18 #DIV/0! Parameter DISSOLVED OXYGEN SM 4500 -0 C 10,00 8,00 10,00 8,80 9,20 8,40 9,60 8,80 9,20 SULPHATES SM 4500-SO4 E 1,22 1,25 1,46 1,50 1,31 1,52 1,15 1,87 1,04 CHLORIDES SM 4500-CL-B 13,00 10,00 11,00 13,00 9,00 15,00 12,00 12,00 13,00 PH (potential of hydrogen) SM 4500-H+B 7,33 7,15 6,95 7,23 7,05 7,15 7,24 6,70 6,83 CONDUCTIVITY SM 2510 B 54,80 53,50 54,80 55,80 53,20 52,40 TURBIDITY SM 2130 B 0,80 0,80 0,80 0,80 1,20 SM 2320 B 12,00 14,00 12,00 12,00 52,10 45,80 34,20 ACIDITY SM 2310 B 10,00 8,00 8,00 1,10 0,80 0,90 1,00 TOTAL ALKALINITY 34,00 14,00 16,00 14,00 16,00 16,00 10,00 8,00 10,00 10,00 8,00 6,00 HARDNESS SM 2340 C 40,00 36,00 30,00 30,00 24,00 22,00 34,00 34,00 Name Measured points Overage Desvest Covtest American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 202 Figure 4: Plastic deformation for asphalt S0 to S21 Table 4: Plastic deformation 2.1.5. Activation Energy Activation energy, AE, is defined as the minimum energy needed for a change or chemical reaction to occur in a material. For the submerged asphalt samples, energy increased with respect to time due to the rise of polar molecules within the asphalt. From the analysis of background AE it was discovered that, for different asphalts, the AE varies between 44 kJ/mol and 90 kJ/mol, [3]. The following example presents calculations of AE for asphalt sample S0. The activation energy, see Figure 5, was calculated based on the Arrhenius equation, equation (1). (1) Where: 𝐾𝐾: Kinetic constant 𝐴𝐴: Specific constant for each reaction. 𝐸𝐸𝑎𝑎: Activation Energy J/mol 𝑅𝑅: Universal constant of gases. 8,3143 J*K-1* mol-1 Asphalt S0 S3 S6 S9 S12 S15 S18 S21 Plastic deformation (mm) at 105 minutes 19,7 10,6 8,5 7,4 9,42 11,7 11,97 17,11 American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 203 𝑇𝑇: Absolute Temperature (ºK) Applying the laws of logarithms: ln 𝐾𝐾 𝐴𝐴 = − 𝐸𝐸𝑎𝑎 𝑅𝑅𝑇𝑇 ln𝐾𝐾 − ln𝐴𝐴 = − 𝐸𝐸𝑎𝑎 𝑅𝑅𝑇𝑇 ln𝐾𝐾 = ln𝐴𝐴 − 𝐸𝐸𝑎𝑎 𝑅𝑅𝑇𝑇 (2) ln𝐾𝐾 = Temperature Dependent Variable, 1 𝑇𝑇 = Independent variable 𝐸𝐸𝑎𝑎 𝑅𝑅𝑇𝑇 = Slope of the linear graph ln(ƞ) vs. 1/K ln𝐴𝐴 =Independent term or intercept of the linear graph, Example of Activation Energy calculation for the original asphalt, S0. Figure 5: Analysis and calculation of the activation energy from the Arrhenius Equation, asphalt S0. Based on the data in Figure 5 , the graph of ln(ƞ) vs. 1/T, in Figure 6 was constructed. Arrhenius equation R= 8,314 J/(mol*Kelvin) 1/R= 0,120 Asphalt Type n (Pa.s) ln(n) Temperature (°C) Temperature (°k) Temperature (1/°k) 15,85 2,76 80,10 353,26 2,83 7,37 2,00 90,20 363,36 2,75 3,16 1,15 100,20 373,36 2,68 1,41 0,35 110,10 383,26 2,61 0,73 -0,31 120,20 393,36 2,54 0,42 -0,86 130,00 403,16 2,48 0,26 -1,34 140,20 413,36 2,42 0,17 -1,76 150,10 423,26 2,36 0,12 -2,14 160,30 433,46 2,31 0,08 -2,48 170,00 443,16 2,26 0,06 -2,79 180,00 453,16 2,21 m= 8,9566 Activation Enerfy m * R = Ea 74,47 KJoule/mol Intercept -22,839 A (Amplitud factor)= EXP(-22,839) 1,205E-10 A = 1,205E-10 80 /1 00 o rig in al S 0 ln η = 𝐸𝐸𝑎𝑎 1 𝑅𝑅 ∗ 1 𝑇𝑇 + ln𝑨 𝐾𝐾 = 𝐴𝐴 ∗ 𝑆𝑆𝑥𝑥𝑆𝑆 −𝐸𝐸𝑎𝑎𝑅𝑅𝑇𝑇 American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 204 Figure 6: Activation Energy for all periods of oxidation 𝐸𝐸𝑎𝑎 = (𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑆𝑆𝑜𝑜 𝑡𝑡ℎ𝑆𝑆 𝑆𝑆𝑙𝑙𝑙𝑙𝑆𝑆𝑙𝑙𝑙𝑙 𝐺𝐺𝑙𝑙𝑙𝑙𝑆𝑆ℎ)𝑥𝑥𝐾𝐾 𝐸𝐸𝑎𝑎 = 8,9566 ∗ 8,3143 J/(mol ∗ K) 𝐸𝐸𝑎𝑎 = 74,465 𝑘𝑘𝑘𝑘/mol In this way, the calculation of 𝐸𝐸𝑎𝑎 was interpolated for all months of submersion. In this experiment, the maximum AE was found in asphalt sample S21 at 84,16 KJ/mol. This value is very close to the maximum reported so far [3], which indicates that after approximately 21 months, asphalt experiences a maximum level of oxidation. See Figure 7. Figure 7: Activation Energy (EA) for each oxidation period 2.1.6. Multi Stress Creep and Recovery-MSCR The MSCR and the percentage of recovery are two of the setting that are calculated using measured tension in different cycles of strength, [7], in order to analyze the response of asphalt under traffic loads. The measurement of the Multiple Stress Creep Recovery–MSCR test of the submerged asphalt samples was American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 205 performed for the maximum Performance Grade-PG of each asphalt sample with the purpose of evaluating the permanent and recovered strain, in accordance with [7] which says, “With the new MSCR specification, the binder testing is done at the high environmental temperature that the pavement is expected to experience. If the climate grade is a PG 64 or PG58, you would do all high temperature testing at 64°C or 58°C. If heavy traffic is expected the specification requirement is changed, i.e. a lower Jnr value is required to reflect the increased stress the pavement will experience, but testing is still done at say 64°C for a PG 64 climate.” Table 4 indicates the values of the complex modulus [G*] of each sample for the maximum PG used in the MSCR test. See Table 5. Table 5: Value of module |𝐺𝐺∗| a 64°C for all asphalts S0 S3 S6 S9 S12 S15 S18 S21 933 1261 1343 1375 1689 2158 2147 1382 As for the variation of accumulated deformation, after each period of submersion it was observed that asphalt sample S3 became more and more rigid and, therefore, decreased its accumulated deformation over time. For submersion periods S6 to S18, the accumulated deformation gradually increased, however, asphalt sample S21 broke this trend by exhibiting a decrease. S0 and S3 were tested at 58°C. S6 through S21 were tested at 64°C. See Figure 8. Figure 8: Multiple Stress Creep and Recovery (MSCR) for asphalts from S0 to S21. Table 6 shows the change in creep compliance unrecoverable Jnr and the elastic recovery of each asphalt sample. American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 206 Table 6: Results of MSCR trial at maximum PG temperature for each asphalt Asphalt Level of Stress 0,1 kPa-1 Level of Stress 3,2kPa-1 Rdiff Jnr,diff Jnr,0,1 (1/kPa) Elastic recovery Jnr 3,2(1/kPa) Elastic recovery S0 1,574 0,36 1,703 0,000 894 8,16 S3 0,333 15,97 0,359 11,410 29 7,54 S6 4,176 0,00 4,450 0,000 -64 6,55 S9 3,929 0,00 4,220 0,000 -77 7,42 S12 2,606 0,68 2,870 0,000 654 10,16 S15 2,629 0,00 2,862 0,000 -585 8,87 S18 2,820 0,00 3,061 0,000 -335 8,55 S21 2,570 0,07 2,838 0,000 5368 10,46 The Jnr values obtained for each asphalt sample are related to the traffic levels for which the asphalt would have the best response in terms of permanent deformation. See Table 7. Table 7: Correlations between Jnr and adequate levels of traffic Level of traffic Jnr (kPa-1), maximum ESALs and speed standard (S) 4 < 10 million and standard speed (VM > 70 km/h)a heavy (H) 2 Between 10 and 30 million of slow traffic (20 < VM < 70 km/h) Very heavy (V) 1 > 30 million of stopped traffic (VM < 20 km/h) Extremely heavy (E) 0,5 > 30 million of stopped traffic (VM < 20 km/h) areas of heavy traffic stopped or ports Figure 9: Comparison of Jnr and activation energy, AE, for all samples with respect to time. American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 207 The activation energy for all submersion periods showed uniform changes ranging from 74.47 to 84.16 KJ / mol. There were abrupt decreases to the Jnr of the asphalt during the first 3 months of submersion, however, after a sharp increase until month 6, from month 6 to 12 it gradually decreased until reaching a level at which it stayed relatively constant during the remaining months. Figure 1 shows the trends of Jnr and AE with respect to time. 3. Conclusions and Recommendations During the period of submersion, physical, rheological and chemical monitoring was done every three months to study changes to the asphalt caused by water. Additionally, some tests were done to compare the performance of common asphalt mixtures with each sample of asphalt (for S0 through S21). However, in this article only the results from the MSCR test and the AE test are analyzed and discussed. In total, eight periods of submersion were analyzed, taking into account the initial conditions of the material. The properties of the conditioning binders in water: S0, S3, S9, S12, S15, S18 and S21 are presented in Table 8. Table 8: Asphalt properties from S0 to S21 Activation energy is the minimum energy required to generate a chemical reaction or change in a material. In this experiment, the activation energy of the asphalt samples increased by 13% after 21 months of submersion. According to international literature, AE values vary from 44 kJ / mol to 90 kJ / mol (Coe & Cui, 2003) for different asphalts. It is therefore concluded that maximum oxidation of the asphalt samples was realized after 21 months. However, in the recommendations for this research it is suggested to perform experimental water diffusion tests on the asphalt. The maximum non-recoverable plastic strain of the MSCR test was obtained after three months of submersion. This result was consistent with the physical and chemical results of asphalts sample S3 and is further supported by the fact that the strongest changes of the submerged asphalt were observed in the first three months. For a deformation of 10% in the MSCR trial, the highest accumulated deformation was obtained after three months of submersion. Given the observation of the tendencies of penetration and dynamic trials, an abrupt decrease in asphalt quality is realized during this time. All submerged asphalt samples didn’t exhibit elastic recovery, making them more susceptible to rutting. The only exception was asphalt sample S3, which Flash Point Cleveland cup Softening Point Brookfield Viscosity Penetration Ductility Standard ASTM D 92 ASTM D36 ASTM D4402 06 ASTM D5 ASTM D113 S0 50,6 50,6 0,381 83,2 144,5 S3 52,0 52,0 0,452 58,7 106,8 S6 53,0 53,0 0,453 48,4 101,7 S9 53,0 53,0 0,394 43,0 93,7 S12 52,2 52,2 0,527 47,1 95,1 S15 52,7 52,7 0,564 46,6 94,2 S18 53,0 53,0 0,584 46,5 93,2 S21 53,3 53,3 0,619 46,5 92,5 TEST (average) Asphalt American Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS) (2018) Volume 44, No 1, pp 198-208 208 demonstrated the worst values of quality and performance for all tests. The variation of the complex modulus of all asphalt samples with respect to temperature exhibits an oscillatory behavior. All asphalt samples became more susceptible to rutting, and stripping as the submersion time increased. This research evidences the physiochemical and rheological changes that asphalt is subjected to because of water, either in liquid (rain or stagnant water) or vapor (moisture) form. This is a typical condition in tropical countries, where frequent rain and high temperatures are common. In accordance with the maximum activation energy and the MSCR test results for changes to the asphalt samples, it is necessary to highlight the variation of performance with respect to traffic levels. Initial testing of the asphalt (S0) yielded performance levels acceptable for heavy traffic but after three months, when the deterioration from oxidation was evident, performance levels were only acceptable for standard traffic. This implies that it is necessary to consider oxidation in pavement design since, currently, only the load function of growth rate and the type of pavement are taken into consideration. References [1] A. Figueroa, R. Velasquez, F. Reyes, and H. Bahia, “Effect of Water Conditioning for Extended Periods on the Properties of Asphalt Binders,” Transp. Res. Rec. J. Transp. Res. Board, no. 2372, pp. 34–45, 2013. [2] K. Kanitpong and H. Bahia, “Relating adhesion and cohesion of asphalts to the effect of moisture on laboratory performance of asphalt mixtures,” Transp. Res. Rec. J. Transp. Res. Board, no. 1901, pp. 33–43, 2005. [3] D. S. Coe and Z. H. Cui, “Determinación de la energ{\’\i}a de activación de flujo en ligantes asfálticos,” in XII Congreso Ibero-Latinoamericano del Asfalto CILA, Quito (Ecuador), Disponible en www. technopave. com/publications/CILA-2003, 2003. [4] H. Bahia, N. Tabatabaee, C. Clopotel, and A. Golalipour, “Evaluation of the MSCR test for modified binder specification,” in The Fifty-sixth Annual Conference of the Canadian Technical Asphalt AssociationCanadian Technical Asphalt Association, 2011. [5] M. D. I. Domingos and A. L. Faxina, “Susceptibility of asphalt binders to rutting: literature review,” J. Mater. Civ. Eng., vol. 28, no. 2, p. 4015134, 2015. [6] R. Stevens, J. Stempihar, B. S. Underwood, and P. Pal, “Evaluation of Multiple Stress Creep and Recovery (MSCR) Data for Arizona,” Int. J. Pavement Res. Technol., vol. 8, no. 5, pp. 337–345, 2015. [7] J. Bukowski, J. Youtcheff, and T. Harman, “THE Multiple Stress Creep Recovery (MSCR) Procedure,” Off. Pavement Technol. - FHWA‐HIF‐11‐038, no. April, pp. 1–9, 2011. References