- Email: [email protected]
Using a modified asphalt bond strength test to investigate the properties of asphalt binders with poly ethylene wax-based warm mix asphalt additive
Accepted Manuscript Using a modified asphalt bond strength test to investigate the properties of asphalt binders with poly ethylene wax-based warm mix asphalt additive Taha A. Ahmed, Hosin “David” Lee, R. Christopher Williams PII: DOI: Reference:
S1996-6814(17)30085-8 http://dx.doi.org/10.1016/j.ijprt.2017.08.004 IJPRT 114
To appear in:
International Journal of Pavement Research and Technology
Received Date: Revised Date: Accepted Date:
23 March 2017 16 July 2017 15 August 2017
Please cite this article as: T.A. Ahmed, H.l. Lee, R.C. Williams, Using a modified asphalt bond strength test to investigate the properties of asphalt binders with poly ethylene wax-based warm mix asphalt additive, International Journal of Pavement Research and Technology (2017), doi: http://dx.doi.org/10.1016/j.ijprt.2017.08.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Using a Modified Asphalt Bond Strength Test to Investigate the Properties of Asphalt Binders with Poly Ethylene Waxbased Warm Mix Asphalt Additive. Taha A. Ahmeda*, Hosin “David” Leeb and R. Christopher Williams c a
Lecturer, Australian College of Kuwait, Department Of Civil Engineering, Australian College
of Kuwait, P.O. Box 1411, Safat 13015, Kuwait, Telephone: +965-90986055. b
Professor (Ph.D.),
Environmental
Engineering Fax: +1-319-335-5660. c
Professor (Ph.D.), Iowa State University, Department of Civil, Construction and Environmental
Engineering, 482A Town Engineering Building, Ames, Iowa, 50011, Telephone: +515-294-4419 Fax: (515) 294-8216 *Corresponding author,
(Taha Ahmed) (Hosin Lee)
Using a Modified Asphalt Bond Strength Test to Investigate the Properties of Asphalt Binders with Poly Ethylene Waxbased Warm Mix Asphalt Additive.
Abstract This paper presents a feasibility of a modified Asphalt Bond Strength (ABS) test method for use with a new adhesion testing device with three different pullout stubs and its application for evaluating Polyethylene (PE) wax-based Warm Mix Asphalt (WMA) additive. Four different asphalt binders were used to evaluate the feasibility of applying the existing AASHTO ABS test method using a new ASTM-certified adhesion testing device and three different pullout stubs. The modified ABS test method was used to evaluate the loss of adhesion and cohesion in asphalt bond strength due to moisture-induced damage. The paper also discusses rheological properties of extracted asphalt binders from three different test sections in Minnesota, Ohio and Iowa constructed using PE wax-based WMA additive and different amounts of Reclaimed Asphalt Pavement (RAP) materials. Based on the modified ABS test results using the new adhesion testing device, it was found that the proposed pullout stub with 0.0 mm thickness (no edge) exhibited consistent results. Based on the results from extracted asphalt performance grading and the multiple stress creep recovery (MSCR) test, it was found that the PE wax-based WMA additive, which was specially designed for asphalt mixtures with a high RAP content, significantly improved the rheological properties of aged asphalt binders from RAP materials. Keywords: Warm-mix asphalt, Poly Ethylene (PE) wax-based, Performance, Moisture damage, Adhesion, Cohesion, Asphalt bond strength (ABS), Reclaimed asphalt pavement (RAP), Multiple stress creep recovery (MSCR), Non-recoverable creep compliance, Percent recovery.
1. Introduction and Literature Review The bond strength between asphalt binder and aggregate can be explained in terms of the adhesion and cohesion characteristics of the asphalt mixture’s components. The lake of enough bond strength between the asphalt binder and the aggregate in the asphalt mixture can lead to moisture damage. Moisture damage is the result of moisture interaction with the asphalt binderaggregate adhesion within the asphalt mixture, making it more susceptible to moisture during cyclic loading (1, 2, 3). This weakening, if severe enough, can result in stripping. Previous research showed that there are three main adhesion mechanisms that can describe the adhesion between the asphalt binder and aggregate surface; chemical adhesion, surface energy, and mechanical adhesion. Chemical adhesion occurs as a result of forming water-insoluble components caused by a chemical reaction between the acidic and basic components of asphalt and aggregate surface. Some research studies suggested that the bond formed by chemical sorption might be necessary in order to minimize stripping potential in asphalt–aggregate mixtures. In general, some aggregates with acidic surfaces don’t react as strongly with asphalt binders, which may not be enough to counter other moisture damage causing factors (1). Surface energy can be explained in terms of relative wettability of aggregate surface by water or asphalt. The surface tension between the asphalt binder and the aggregate at the wetting line is less than the surface tension between the water and the aggregate due to its higher viscosity. Thus, if all the three are in contact (water, aggregate, and asphalt binder), water more likely replaces asphalt binder. This will result in less aggregate coating by asphalt binder and eventual striping. The interfacial tension between aggregate and asphalt binder depends on the asphalt type, aggregate type, and the aggregate surface roughness (1, 2).
Mechanical adhesion depends on the physical properties of the aggregate such as surface texture, porosity or absorption, surface coatings, surface area, and particle size (1). Asphalt binder gets into the aggregate surface pores and irregularities, and when it hardens it causes a mechanical lock. The interference of moisture with the asphalt binder penetration reduces the mechanical lock and leads to stripping. Good mechanical lock can improve the nature of the chemical bond between the asphalt binder and aggregate surface even in the presence of water (1, 2). Cohesion is developed in the asphalt mastic, asphalt binder mixed with fine aggregate, and it would depend on the rheological properties of the asphalt binder. The asphalt mastic’s resistance to microcracking is highly influenced by the dispersion of the mineral filler. Thus, the cohesive strength is controlled by the combination and the interaction of asphalt binder and the mineral filler (3). Water can affect the cohesion of asphalt mastic in several ways such as weakening the mastic due to water saturation and void swelling (4). An asphalt mixture could lose up to 50 percent of its resilient modulus upon saturation but, upon drying, the modulus can be completely recovered (5). This is shown graphically in Figure 1. [Insert Figure 1] Figure 1 Effect of Moisture on Resilient Modulus is Reversible (5). Based on the above discussion, it can be said that the moisture damage in asphalt mixture can be due to a cohesion failure within the mastic, or an adhesion failure at the aggregate-asphalt binder interface. Both failure modes can be related to the characteristics of the asphalt mastic and the asphalt film thickness around aggregate particles. Thus, asphalt mixtures with thin asphalt film tend to fail in tension by adhesive bond rupture whereas asphalt mixture with thicker asphalt film thickness tends to fail in a cohesive failure mode due to the damage within mastic. The determination of asphalt film thickness that differentiate the two modes of failure would depend
on the rheological properties of the asphalt, the amount of damage the asphalt or mastic can withstand prior to failure, a rate of loading, and a temperature at the time of loading (6). Therefore, several test methods were developed to evaluate the loss of adhesion or cohesion between the asphalt binder and the aggregate surface due to moisture-induced damage. However, many of these tests use indirect methods to measure the adhesion and cohesion properties, which may not be very consistent (7). Therefore, a new test procedure (AASHTO TP 91-11: Standard Method of Test for Determining Asphalt Binder Bond Strength by Means of the Asphalt Bond Strength (ABS) Test) was developed for measuring moisture damage characterization based on asphalt-aggregate adhesive and cohesive properties (8). The new method measures the pull-off force needed to break the bond strength between the asphalt binder and the aggregate surface using adhesion testing device. This AASHTO procedure is based on test results using only one type of devices. However, the test procedure should be universally applicable using multiple testing devices. Therefore, in this study, a different adhesion testing device is proposed to improve the consistency of the existing AASHTO procedure. Several studies have reported that, Warm Mix Asphalt (WMA) mixtures are more susceptible to moisture than conventional Hot Mix Asphalt (HMA) mixtures (9). Therefore, this research study used the new modified Asphalt Bond Strength (ABS) test method to investigate the adhesive and cohesive characteristics of WMA mixtures prepared using Poly Ethylene (PE) waxbased additive. This PE wax-based additive works by reducing the viscosity of the asphalt at lower temperatures with a melting point of 100 °C and crystallization point of 90 °C. The PE wax-based additive controls the crystallization so that it does not become brittle at low temperature. Additionally, the PE wax-based additive contains anti-stripping agent to enhance the bonding of asphalt binder to the aggregate surface. In this study, three different forms of the
PE wax-based WMA additive are evaluated to measure their effectiveness in improving the rheological properties of asphalt binders.
2. Asphalt Bond Strength Test Method ASTM-certified adhesion testing device, PosiTest AT-A automatic adhesion tester, was used to perform the ABS test following the AASHTO TP 91-11 procedure. In order to prepare the test specimens, the asphalt binder is attached to the aggregate surface by means of adhesion at controlled environmental and moisture conditions. According to AASHTO TP 91-11 procedure, all aggregate substrates were prepared by cutting quarried rocks using a standard rock saw to create parallel faces and then Lapped using a 280-grit silicon carbide material on a standard lapidary wheel to remove saw marks and ensure a consistent surface roughness. After cut and lapped, samples were cleaned for 60 minutes in ultrasonic cleaner containing distilled water at a temperature of 60o C to remove residual particles on the plate surface (8). To measure the pulloff force needed to detach the asphalt binder samples from the aggregate surface, a hydraulic pressure is applied to pull out a stub attached to the asphalt sample. Figure 1 shows a schematic of the pull-off test method and the PosiTest Pull-Off Adhesion Tester. The AASHTO TP 91-11 procedure recommended new geometry and treatment to the stub surface used with PATTI Quantum Gold adhesion tester in order to create a rough texture that would provide a mechanical interlock and larger contact area between the asphalt binder and stub. This AASHTO procedure is based on test results using only one type of device and one pullout stub of 0.8-mm thickness. Therefore, for this study, three different modified pullout stubs with thicknesses of 0.8 mm, 0.4 mm and 0 mm (no edge) are proposed to be used with the PosiTest AT-A automatic adhesion tester. Figure 2 shows one pullout stub used for PATTI Quantum Gold and three modified stubs used for PosiTest Pull-Off adhesion tester device.
[Insert Figure 2] Figure 2 Schematic of the Pull-off Test Method (left) and PosiTest Pull-Off Adhesion Tester (right) (10). [Insert Figure 3] Figure 3 Pull-out Stubs for a) PATTI Quantum Gold Adhesion Tester, and b) PosiTest® Pull-Off Adhesion Tester. Limestone aggregate and four different binder grades were tested using the three pullout stubs in order to find out which pullout stub should be used with the new adhesion testing device. Only un-conditioned samples were tested in this part of the study.
3. Test Sections with Poly Ethylene Wax-based WMA Additive Three different forms of the PE wax-based WMA additive; pellets, liquid, and specially designed liquid form for mixtures with high RAP amount were evaluated in this study to measure their effectiveness in improving the rheological properties of asphalt binders. The three forms were implemented in three different projects with a rate of 1.5% by optimum asphalt content of the mix. Asphalt binders were extracted from field WMA mixtures from following three test sections. First, the PE wax-based pellet (PE-Pellet) form was applied to a 2.0-inch mill and overlay WMA test section on the southbound outside lane of TH 169 state highway in Champlin, Minnesota. Asphalt binder PG64-28 and Granite aggregate along with 25% RAP by total weight of mix were used to prepare the mix. Another HMA test section was constructed for comparison using the same properties. The HMA and WMA mixtures were designed according to Superpave mix design procedure for a medium traffic level of 3 to 10 million ESALs according to Minnesota Department of Transportation (MnROAD) mix design requirements. Second, the PE wax-based liquid (PE-Liquid) form was applied to a 3.0-inch asphalt layer on State Highway 158 in Lancaster, Ohio. The 3.0-inch layer consisted of intermediate layer with a
thickness of 1.75 inch and surface layer with a thickness of 1.25 inch. WMA and HMA mixes were designed according to Marshall mix design procedure following Ohio Department of Transportation (ODOT) mix design specifications for a medium traffic volume. The mix used a blend of Limestone and Natural gravel aggregates and 25% RAP materials by weight for the intermediate layer and 20 % RAP by weight for the surface layer. Asphalt binder PG70-22 was use for the surface layer and PG64-22 was used for the intermediate layer. Only asphalt binder extracted from the surface layer mixtures was evaluated in this study. Third, the specially designed PE wax-based liquid (SDPE-Liquid) form for mixtures with high RAP amount was applied to a surface layer with a thickness of 1.5 inch on state highway 6 in Iowa City, Iowa. Two WMA and HMA test sections were constructed for a 10 million ESALs traffic volume. The mixes were designed according to Superpave mix design procedure per Iowa Department of Transportation (Iowa DOT) mix design requirements. The mixtures used asphalt binder PG64-28 and Limestone aggregate along with 30% fractionated RAP by binder replacement. Table 1 shows a summary of properties and the materials used in the test sections. Table 1 Summary of the Materials Used in the Test Sections. [Insert Table 1] 4. Modified Asphalt Bond Strength Test Results In order to capture the effect of different pullout stub thicknesses and the asphalt binder grade on the asphalt-aggregate bond strength, test specimens using four different asphalt binder grades; PG58-28, PG64-22, PG64-28M (polymer modified) and PG70-22M (polymer modified) were prepared following AASHTO TP 91-11 test procedure. Asphalt samples, limestone aggregate plates and pullout stubs were heated up at 150° C for a minimum of 30 minutes to remove absorbed water on the aggregate surface and provide a better bond between the asphalt binder and the aggregate surface. The aggregate plates were then brought to an application temperature
of 60° C, whereas the application temperature for the pullout stubs still 150° C. After a sufficient heat-up time, the molten asphalt samples were carefully poured in 10.0 mm diameter Dynamic Shear Rheometer (DSR) silicon molds and left for 30 minutes to reach the room temperature. The asphalt samples were attached to the aggregate plates by placing them on the surface of the pullout stub then firmly pressing each stub with the asphalt sample into the aggregate plate surface until the stub is in contact with the aggregate surface and no excess of asphalt flowing is observed. The prepared samples were then left for 24 hours at room temperature before testing. The ABS test results are shown in Figure 3 with the numbers above each bars representing the average value and the whisker representing the standard deviation. Overlapping of the standard deviations implies the similarity in the measured ABS values between the different types of pullout stubs. As can be seen from Figure 3, the 0.0 mm pullout stub produced significantly higher and more consistent ABS values than 0.4 mm and 0.8 mm stubs. It can be seen from Figure 3 that the bond strength was significantly affected by the asphalt type. [Insert Figure 4] Figure 4 ABS Test Results of Modified Pullout Stubs. In order to measure the actual thickness of the applied asphalt layer, the aggregate plates were cut across the centerline of the asphalt samples and investigated using Olympus SZ61 microscope equipped with Olympus DP26 digital camera as shown in Figure 4 a). The average values of the asphalt layer for 0.8 mm, 0.4 mm and 0.0 mm pullout stubs were 998 μm, 539 μm and 106 μm, respectively. Figure 4 b) shows samples of asphalt layer thicknesses created by the pullout stubs after running the test. As can be seen from Figure 4 b), the measured layer thicknesses were slightly higher than the real stubs’ thicknesses due to the irregularity and pores of the aggregate plates and the groves on the stubs. Asphalt layer thicknesses were measured
after the asphalt samples were exposed to a direct tension during the test, which resulted in a plastic deformation in the asphalt samples. According to AASHTO TP 91-11, samples are considered failed in adhesion if more than 50% of the binder is removed after the test is performed. Otherwise, the failure is cohesive. All the tested samples failed in a cohesive mode except for PG58-28 and PG64-22 asphalt samples tested with 0.0 mm pullout stubs. Figure 4 c) shows a picture of a test sample with 75% cohesive failure with the yellow areas indicating an adhesive failure. [Insert Figure 5] Figure 5 a) Olympus SZ61 Microscope, b) Samples of Asphalt Layer Thicknesses Created by the Pullout Stubs, and c) Test Sample with75% Cohesive Failure. By looking at the measured asphalt layers and the corresponding mode of failure, it can be seen that samples with high asphalt layer thickness always fail in cohesive mode. This explains why all the asphalt samples created using 0.8 mm, and 0.4 mm stubs failed in a cohesive mode. On the other hand, the asphalt samples created by 0.0 mm stub exhibited both cohesive and adhesive failure modes. The ABS values obtained from 0.0 mm pullout stubs were significantly higher than the others because there were more direct contact points between the pullout stub and the aggregate plate. Two-way ANOVA analysis, as shown Table 2, was conducted to evaluate the impact of pullout stub type, the asphalt performance grade and their interaction on the measured ABS values between the asphalt binder and the aggregate surface. It can be seen from Table 1 that, the pullout stub type, the asphalt performance grade and their interaction had significant impacts on the measured ABS value between the asphalt binder and the aggregate surface. It can be concluded that using the 0.0 mm (no edge) pullout stubs with the new adhesion tester device
produced more consistent and reasonable ABS values by providing more direct contact points between the pullout stub and the aggregate surface. Table 2 ANOVA Table of the ABS Test Results. [Insert Table 2] 5. Rheological and Adhesive/Cohesive Characteristics of Extracted Asphalt Binders The asphalt binders of the collected mixtures from the test sections were extracted, re-graded and then evaluated using the Multiple Stress Creep Recovery (MSCR) test following ASHTO TP70, "Standard Method of Test for Multiple Stress Creep Recovery (MSCR) Test of Asphalt Binder Using a Dynamic Shear Rheometer (DSR)." The MSCR test uses the creep and recovery method to measure the percent recovery and non-recoverable creep compliance (Jnr). It can be used to predict the rutting susceptibility of the mix (11). Typically, reducing Jnr value by half can reduce rutting by half (12). The modified ABS test using the PosiTest Pull-Off adhesion testing device and 0.0 mm (no edge) pullout stub was used to evaluate the asphalt-aggregate bond strength of the extracted and recovered asphalt binders in both dry (un-conditioning) and wet (moisture conditioning) situations. According to AASHTO TP91-11, the moisture conditioning process includes soaking the prepared samples for 24 hours in a heated water bath at 40°±2° C followed by leaving for 1 hour at room temperature before testing. The specified conditioning temperature by the AASHTO TP 91-11 procedure can be changed to match the conditioning temperature used for AASHTO T 283 (Standard Method of Test for Resistance of Compacted Asphalt Mixtures to Moisture-Induced Damage) or any other reasonable temperature as needed. However, in this study, the conditioning temperature kept unchanged because the study focused mainly on the feasibility of using the existing test procedure with a new adhesion testing device other than the one was originally used.
5.1. Rheological Properties of the Extracted Asphalt Binder The extraction process was done following the ASTM D2172/D2172M-11 “Standard Methods for Quantitative Extraction of Bitumen from Bituminous Paving Mixtures”, and ASTM D5404/D540M-12 “Standard Practice for Recovery of Asphalt from Solution Using the Rotary Evaporator”. After the extraction and recovery of asphalt binders was done, a full performance grading was done following AASHTO M 320 “Standard Specification for Performance-Graded Asphalt Binder” and AASHTO MP19-10 “ Performance Graded Asphalt Binder Using Multiple Stress Creep Recovery (MSCR) Test” (13). Tables 2 through 4 show summaries of the performance grading results for the different extracted and recovered asphalt binders. The target asphalt performance grade for Minnesota test section with 25% RAP by weight of mix was 70-22. Therefore, due to the high RAP content, virgin asphalt grade was lowered by one level to PG 64-28. The extracted asphalt binders from both HMA and PE-Pellet mixtures of Minnesota test section were tested and graded as PG76-22 for a standard traffic “S” level. Asphalt binder PG70-22 was used for Ohio test section with 20% RAP by weight of mix. The extracted asphalt binders from both HMA and PE-Liquid mixtures of Ohio test sections were tested and graded as PG82-16 for a standard traffic “S” level. The target performance grade for Iowa test section was 70-22 with 30% RAP by binder replacement. According to Iowa DOT, asphalt performance grade should be lowered one level when using more than 20% RAP by binder replacement in the mix and, therefore, PG64-28 was used to produce the mixtures. The extracted asphalt binder from the HMA mixture of Iowa test section was tested and graded as PG82-16 for a standard traffic “S” level whereas the extracted asphalt binder from the SDPE-Liquid mixture was tested and graded as PG70-28 for a heavy traffic “H” level. The SDPE-Liquid additive reduced the impact of aging on the asphalt binder
during construction and improved the high and low temperature grades of the extracted asphalt binder. Further, the SDPE-Liquid additive enhanced the asphalt binder characteristics during the MSCR test by decreasing the non-recoverable creep compliance, which resulted in a higher qualified asphalt binder for a higher traffic level “H” instead of a standard traffic level “S” obtained for the extracted asphalt PG82-16 from the HMA mixture.
Table 3 Rheological Properties of the Extracted Asphalt Binders from Minnesota Test Section. [Insert Table 3] Table 4 Rheological Properties of the Extracted Asphalt Binders from Ohio Test Section. [Insert Table 4] Table 5 Rheological Properties of the Extracted Asphalt Binders from Iowa Test Section. [Insert Table 5] 5.2. Asphalt Bond Strength Test Results The modified Asphalt Bond Strength (ABS) test results are shown in Table 5 and plotted in Figure 5 with the numbers above the bars representing the average values and the whiskers representing the standard deviation. Overlapping of the standard deviation implies the similarity in the measured ABS between the asphalt types. For both HMA and PE-Pellet samples from Minnesota test section, moisture conditioned samples exhibited slightly lower ABS values than the un-conditioned samples. Moisture conditioned PE-Pellet samples exhibited slightly lower ABS values than moisture conditioned HMA samples. Moisture conditioned HMA and PE-Pellet samples from Minnesota test section exhibited higher ABS values than samples from Iowa and Ohio test sections, which indicates a high resistance to moisture damage. Moisture conditioned HMA and PE-Liquid samples from Ohio test section exhibited similar
ABS values and significantly lower than un-conditioned samples. Moisture conditioned HMA and PE-Liquid samples from Ohio exhibited the lowest ABS values, which indicate a low resistance to moisture damage. Moisture conditioned HMA and SDPE-Liquid samples from Iowa test section exhibited significantly lower ABS values than un-conditioned samples. The moisture conditioned SDPELiquid samples exhibited significantly higher ABS values than HMA samples. It can be seen that, both HMA and SDPE-Liquid mixtures are susceptible to moisture damage. However, the SDPE-Liquid showed a better resistance to moisture damage, which can be attributed to the SDPE-Liquid additive’s ability to improve the rheological properties of the asphalt binder. Table 6 ABS Test Results of Extracted Asphalt Binders. [Insert Table 6] [Insert Figure 6] Figure 6 ABS Test Results of Extracted Asphalt Binders. 6. Conclusions The study presented a feasibility of a modified asphalt bond strength test method for use with a new adhesion testing device with three different pullout stubs and its application for evaluating PE wax-based WMA additive. Based on the research findings, it can be concluded that the ABS test can be conducted using a new adhesion testing device with a modified 0.0 mm (no edge) pullout stub. The modified ABS test method was successfully applied to evaluate extracted and recovered asphalt binders from three different test sections. The modified ABS test method showed that the asphalt binder type had an impact on the ABS test results similar to the previous study done using the original ABS test method (7, 8). Additionally, it was found that the SDPE-liquid WMA additive significantly improved both
high and low temperature grades of asphalt. The MSCR test results showed that the SDPE-liquid additive significantly enhanced the non-recoverable creep compliance and the percent recovery of the asphalt binder. Furthermore, the samples prepared using the SDPE-liquid additive exhibited a better moisture resistance than HMA samples. In order to improve the repeatability of ABS test, more tests should be performed on various aggregate surfaces, different asphalt binders, and different conditioning methods and temperatures. Also, the impact of aggregate absorption on the moisture conditioned samples should be included in future studies to see if the aggregate absorption has a significant influence on the measured bond strength using the ABS test method. 7. Acknowledgement The Authors would like to acknowledge the great support of the Korean Institute of Civil Engineering and Building Technology and the Mid-America Transportation Center in funding this research study.
References [1] Little, D. N. Jr.; Jones, D. R. IV., “Chemical and Mechanical Processes of Moisture Damage in Hot-Mix Asphalt Pavements,” National Seminar, TRB Committee on Bituminous– Aggregate Combinations to Meet Surface Requirements Transportation Research Board (TRB), San Diego, CA, February 2003, pp. 37–74. [2] Tarrer, A. R.; Wagh, V., “The Effect of the Physical and Chemical Characteristics of the Aggregate on Bonding,” Strategic Highway Research Program, National Research Council, Washington, D.C, 1991. [3] Kiggundu, B. M.; Roberts, F. L., “The Success/Failure of Methods Used to Predict the Stripping Potential in the Performance of Bituminous Pavement Mixtures,” NCAT Report No. 88-03, National Center for Asphalt Technology, Auburn, January 1988. [4] Terrel, R. L.; Al-Swailmi, S., “Water Sensitivity of Asphalt–Aggregate Mixes: Test Selection,” SHRP Report A-403, Strategic Highway Research Program, National Research Council, Washington, D.C. June 1994. [5] Schmidt, R. J.; Graf, P. E., “The Effect of Water on the Resilient Modulus of Asphalt Treated Mixes,” Proc., Association of Asphalt Paving Technologists, Vol. 41, 1972, pp. 118–162. [6] Hefer, A. and Little, D., “Adhesion in Bitumen-Aggregate Systems and Quantification of the effects of Water on the Adhesive Bond,” Report No. ICAR/505-1, International Center for Aggregates Research, December 2005. [7] Raquel, M; Raul, V.; Bahia, H. U., “Measuring the Effect of Moisture on Asphalt–Aggregate Bond with the Bitumen Bond Strength Test,” Transportation Research Board, Washington, DC, 2009, PP. 70-81. [8] American Association of State Highway and Transportation Officials, “AASHTO TP9-11: Standard Method of Test for Determining Asphalt Binder Bond Strength by Means of the Asphalt Bond Strength (ABS) Test,” 2011. [9] Lee, H.; Glueckert, T.; Ahmed, T.; Kim, Y.; Baek, C.; Hwang, S., “Laboratory Evaluation and Field Implementation of Polyethylene Wax-Based Warm Mix Asphalt Additive in USA”, International Journal of Pavement Research and Technology, Vol.6 No.5 September 2013, pp. 547-553.
32
[10] Defelsko Company, “PosiTest, Pull-off Adhesion Tester Operational Manual,” Obtained from http://www.defelsko.com/adhesion-tester/adhesiontester.htm. Site last accessed July 2014. [11] Federal Highway Administration (FHWA), “The Multiple Stress Creep Recovery (MSCR) Procedure,” FHWA-HIF-11-038, April 2011. Obtained from http://www.fhwa.dot.gov/pavement/materials/pubs/hif11038/tb00.cfm. Site last accessed July 2014. [12] DuBois, E.; Mehta, Y., Nolan, A., “Correlation between multiple stress creep recovery (MSCR) results and polymer modification of binder,” Construction and Building Materials, Vol. 65, 29 August 2014, pp 184–190. [13] American Association of State Highway and Transportation Officials, “Standard Specifications of Transportation Materials and Methods of Sampling and Testing,” 29th Edition [CD-ROM], 2009.
List of Figures
Figure 1 Effect of Moisture on Resilient Modulus is Reversible (5). Figure 2 Schematic of the Pull-off Test Method (left) and PosiTest Pull-Off Adhesion Tester (right) (10). Figure 3 Pull-out Stubs for a) PATTI Quantum Gold Adhesion Tester, and b) PosiTest® Pull-Off Adhesion Tester. Figure 4 ABS Test Results of Modified Pullout Stubs. Figure 5 a) Olympus SZ61 Microscope, b) Samples of Asphalt Layer Thicknesses Created by the Pullout Stubs, and c) Test Sample with75% Cohesive Failure. Figure 6 ABS Test Results of Extracted Asphalt Binders.
Figure 1 Effect of Moisture on Resilient Modulus is Reversible (5).
Figure 2 Schematic of the Pull-off Test Method (left) and PosiTest Pull-Off Adhesion Tester (right) (10).
(a)
(b) Figure 3 Pull-out Stubs for a) PATTI Quantum Gold Adhesion Tester, and b) PosiTest® Pull-Off Adhesion Tester.
Bond Strength at 25°C (77° F) , psi
550.0 488.0
500.0
454.0
450.0
412.3
400.0 350.0
329.0
307.3
300.0 229.0
250.0
263.0 233.1
259.0249.7
200.0 150.0
135.0129.0
100.0 PG 58-28
PG 64-22 PG 64-28 PM Asphalt Type
0.8 mm Pullout Stub
0.4 mm Pullout Stub
PG 70-22 PM
0.0 mm Pullout Stub
Figure 4 ABS Test Results of Modified Pullout Stubs.
Figure 5 a) Olympus SZ61 Microscope, b) Samples of Asphalt Layer Thicknesses Created by the Pullout Stubs, and c) Test Sample with75% Cohesive Failure
Figure 6 ABS Test Results of Extracted Asphalt Binders.
List of Tables
Table 1 Summary of Properties and Materials Used in the Test Sections. Table 2 ANOVA Table of the ABS Test Results. Table 3 Rheological Properties of the Extracted Asphalt Binders from Minnesota Test Section. Table 4 Rheological Properties of the Extracted Asphalt Binders from Ohio Test Section. Table 5 Rheological Properties of the Extracted Asphalt Binders from Iowa Test Section. Table 6 ABS Test Results of Extracted Asphalt Binders.
Table 1 Summary of Properties and Materials Used in the Test Sections.
Site Location Application Type Aggregate type RAP Content, % Binder Type
Test Section #1 Highway TH 169, Champlin, Minnesota 2.0-inch Surface Layer Granite 25 % (by total mix weight) PG64-28 (polymer modified)
Test Section #2
Test Section #3
Highway 158, Lancaster, Ohio
Highway 6, Iowa City, Iowa
1.25-inch Surface Layer Limestone & Natural gravel 20 % (by total mix weight) PG70-22 (polymer modified)
1.5-inch Surface Layer Limestone 30 % (by binder replacement) PG64-28 (polymer modified) Specially Designed Poly Ethylene wax-based liquid (SDPE-Liquid) Superpave
WMA Additive Type
Poly Ethylene wax-based pellet (PE-Pellet)
Poly Ethylene wax-based liquid (PE-Liquid)
Mix Design Method
Superpave
Marshall
Table 2 ANOVA Table of the ABS Test Results. Source Degrees of Freedom Type 1 SS Mean Square Asphalt Performance Grade (PG) 3 120048.2222 40016.0741 Pullout Stub (PS) 2 329435.0556 164717.5278 PG*PS 6 16502.2778 2750.3796
F Value 59.96 246.83 4.12
P-Value > F <.0001 <.0001 0.0055
Table 3 Rheological Properties of the Extracted Asphalt Binders from Minnesota Test Section. HMA PE-Pellet Specification PG 76-22 PG 76-22 Tests on Original Binder Dynamic shear, G*/sinδ (10 rad/s), kPa 1.914 at 76°C 1.78 at 76°C 1.00 Min Tests on Residue from RTFO, AASHTO T240 Mass loss, % NT NT 1.00% Max Dynamic shear, G*/sin δ (10 rad/s), kPa 3.26 at 76°C 2.304 at 76°C 2.20 Min Multiple Stress Creep and Recovery (MSCR) of Asphalt Binder Property
-1
Non-recoverable creep compliance, J Percent difference between J
nr3.2
nr3.2
and J
2.75 at 76°C
, kpa
nr0.1
,J
,%
nrdiff
34.65 at 76°C
Test Method AASHTO T315 AASHTO T240 AASHTO T315 AASHTO TP 70-11
-1
3.405 at 76°C Max, 4.0 kpa HMA: Standard Traffic “S” 25.85 at 76°C Max, 75% PE-Pellet: Heavy Traffic “S”
Tests on Residue from Pressure Aging Vessel, AASHTO R28 @ 100°C Dynamic shear, G*sin δ at 28°C (10 rad/s), kPa 3929.5 at 25°C 3548.5 at 28°C 5000 Max AASHTO T315 Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR) S-value, MPa 161 at -12°C 188 at -12°C 300 Max AASHTO T313 m-value 0.317 at -12°C 0.3055 at -12°C 0.300 Min
Table 4 Rheological Properties of the Extracted Asphalt Binders from Ohio Test Section. HMA PE-Liquid Specification PG 82-16 PG 82-16 Tests on Original Binder Dynamic shear, G*/sinδ (10 rad/s), kPa 2.2219 at 82°C 2.437 at 82°C 1.00 Min Tests on Residue from RTFO, AASHTO T240 Mass loss, % NT NT 1.00% Max Dynamic shear, G*/sin δ (10 rad/s), kPa 3.3917 at 82°C 3.066at 82°C 2.20 Min Multiple Stress Creep and Recovery (MSCR) of Asphalt Binder Property
-1
Non-recoverable creep compliance, J Percent difference between J
nr3.2
nr3.2
and J
2.87 at 82°C
, kpa
nr0.1
,J
,%
nrdiff
70.4 at 82°C
Test Method AASHTO T315 AASHTO T240 AASHTO T315 AASHTO TP 70-11
-1
3.745 at 82°C Max, 4.0 kpa HMA: Standard Traffic “S” 62.85 at 82°C Max, 75% PE-Liquid: Heavy Traffic “S”
Tests on Residue from Pressure Aging Vessel, AASHTO R28 @ 100°C Dynamic shear, G*sin δ at 28°C (10 rad/s), kPa 4399.5 at 28°C 4330 at 28°C 5000 Max AASHTO T315 Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR) S-value, MPa 119 at -6°C 109 at -6°C 300 Max AASHTO T313 m-value 0.372 at -6°C 0.3835 at -6°C 0.300 Min
Table 5 Rheological Properties of the Extracted Asphalt Binders from Iowa Test Section. HMA SDPE-Liquid Specification PG 82-16 PG 70-28 Tests on Original Binder Dynamic shear, G*/sinδ (10 rad/s), kPa 1.363 at 82°C 1.12 at 70°C 1.00 Min Tests on Residue from RTFO, AASHTO T240 Mass loss, % NT NT 1.00% Max Dynamic shear, G*/sin δ (10 rad/s), kPa 2.326 at 82°C 2.6875 at 70°C 2.20 Min Multiple Stress Creep and Recovery (MSCR) of Asphalt Binder Property
Test Method AASHTO T315
AASHTO T240 AASHTO T315 AASHTO TP 70-11 -1 -1 3.95 at 82°C 1.815 at 70°C Max, 4.0 kpa HMA: Standard Traffic “S” Non-recoverable creep compliance, J , kpa nr3.2 SDPE-Liquid: Heavy Traffic Percent difference between J and J , J , % 39.7 at 82°C 33.7 at 70°C Max, 75% “H” nr3.2 nr0.1 nrdiff Tests on Residue from Pressure Aging Vessel, AASHTO R28 @ 100°C Dynamic shear, G*sin δ at 28°C (10 rad/s), kPa 3675.5 at 28°C 4726.5 at 22°C 5000 Max AASHTO T315 Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR) S-value, MPa 80.95 at -6°C 226 at -18°C 300 Max AASHTO T313 m-value 0.367 at -6°C 0.3125 at -18°C 0.300 Min
Table 6 ABS Test Results of Extracted Asphalt Binders. Mix Type/Asphalt Grade Minnesota _HMA/PG76-22 Minnesota _PE_Pellet/PG76-22 Ohio_HMA/PG82-16 Ohio_PE-Liquid/PG82-16 Iowa_HMA/PG82-16 Iowa_SDPE-Liquid/PG70-28
Un-conditioned Samples Average, psi
488.7 490.7 567.0 475.3 490.0 519.7
St. Dev.
Failure Mode
52.3 41.0 16.5 14.5 50.3 12.6
53% Cohesion 73% Cohesion 76% Cohesion 76% Cohesion 76% Cohesion 76% Cohesion
Conditioned Samples Average, psi
426.0 391.3 200.7 192.0 269.7 322.3
St. Dev.
Failure Mode
32.2 22.0 12.5 12.2 38.7 5.5
85% Cohesion 73% Cohesion 77% Adhesion 78% Adhesion 80% Adhesion 65% Cohesion
S1996-6814(17)30085-8 http://dx.doi.org/10.1016/j.ijprt.2017.08.004 IJPRT 114
To appear in:
International Journal of Pavement Research and Technology
Received Date: Revised Date: Accepted Date:
23 March 2017 16 July 2017 15 August 2017
Please cite this article as: T.A. Ahmed, H.l. Lee, R.C. Williams, Using a modified asphalt bond strength test to investigate the properties of asphalt binders with poly ethylene wax-based warm mix asphalt additive, International Journal of Pavement Research and Technology (2017), doi: http://dx.doi.org/10.1016/j.ijprt.2017.08.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Using a Modified Asphalt Bond Strength Test to Investigate the Properties of Asphalt Binders with Poly Ethylene Waxbased Warm Mix Asphalt Additive. Taha A. Ahmeda*, Hosin “David” Leeb and R. Christopher Williams c a
Lecturer, Australian College of Kuwait, Department Of Civil Engineering, Australian College
of Kuwait, P.O. Box 1411, Safat 13015, Kuwait, Telephone: +965-90986055. b
Professor (Ph.D.),
Environmental
Engineering Fax: +1-319-335-5660. c
Professor (Ph.D.), Iowa State University, Department of Civil, Construction and Environmental
Engineering, 482A Town Engineering Building, Ames, Iowa, 50011, Telephone: +515-294-4419 Fax: (515) 294-8216 *Corresponding author,
(Taha Ahmed) (Hosin Lee)
Using a Modified Asphalt Bond Strength Test to Investigate the Properties of Asphalt Binders with Poly Ethylene Waxbased Warm Mix Asphalt Additive.
Abstract This paper presents a feasibility of a modified Asphalt Bond Strength (ABS) test method for use with a new adhesion testing device with three different pullout stubs and its application for evaluating Polyethylene (PE) wax-based Warm Mix Asphalt (WMA) additive. Four different asphalt binders were used to evaluate the feasibility of applying the existing AASHTO ABS test method using a new ASTM-certified adhesion testing device and three different pullout stubs. The modified ABS test method was used to evaluate the loss of adhesion and cohesion in asphalt bond strength due to moisture-induced damage. The paper also discusses rheological properties of extracted asphalt binders from three different test sections in Minnesota, Ohio and Iowa constructed using PE wax-based WMA additive and different amounts of Reclaimed Asphalt Pavement (RAP) materials. Based on the modified ABS test results using the new adhesion testing device, it was found that the proposed pullout stub with 0.0 mm thickness (no edge) exhibited consistent results. Based on the results from extracted asphalt performance grading and the multiple stress creep recovery (MSCR) test, it was found that the PE wax-based WMA additive, which was specially designed for asphalt mixtures with a high RAP content, significantly improved the rheological properties of aged asphalt binders from RAP materials. Keywords: Warm-mix asphalt, Poly Ethylene (PE) wax-based, Performance, Moisture damage, Adhesion, Cohesion, Asphalt bond strength (ABS), Reclaimed asphalt pavement (RAP), Multiple stress creep recovery (MSCR), Non-recoverable creep compliance, Percent recovery.
1. Introduction and Literature Review The bond strength between asphalt binder and aggregate can be explained in terms of the adhesion and cohesion characteristics of the asphalt mixture’s components. The lake of enough bond strength between the asphalt binder and the aggregate in the asphalt mixture can lead to moisture damage. Moisture damage is the result of moisture interaction with the asphalt binderaggregate adhesion within the asphalt mixture, making it more susceptible to moisture during cyclic loading (1, 2, 3). This weakening, if severe enough, can result in stripping. Previous research showed that there are three main adhesion mechanisms that can describe the adhesion between the asphalt binder and aggregate surface; chemical adhesion, surface energy, and mechanical adhesion. Chemical adhesion occurs as a result of forming water-insoluble components caused by a chemical reaction between the acidic and basic components of asphalt and aggregate surface. Some research studies suggested that the bond formed by chemical sorption might be necessary in order to minimize stripping potential in asphalt–aggregate mixtures. In general, some aggregates with acidic surfaces don’t react as strongly with asphalt binders, which may not be enough to counter other moisture damage causing factors (1). Surface energy can be explained in terms of relative wettability of aggregate surface by water or asphalt. The surface tension between the asphalt binder and the aggregate at the wetting line is less than the surface tension between the water and the aggregate due to its higher viscosity. Thus, if all the three are in contact (water, aggregate, and asphalt binder), water more likely replaces asphalt binder. This will result in less aggregate coating by asphalt binder and eventual striping. The interfacial tension between aggregate and asphalt binder depends on the asphalt type, aggregate type, and the aggregate surface roughness (1, 2).
Mechanical adhesion depends on the physical properties of the aggregate such as surface texture, porosity or absorption, surface coatings, surface area, and particle size (1). Asphalt binder gets into the aggregate surface pores and irregularities, and when it hardens it causes a mechanical lock. The interference of moisture with the asphalt binder penetration reduces the mechanical lock and leads to stripping. Good mechanical lock can improve the nature of the chemical bond between the asphalt binder and aggregate surface even in the presence of water (1, 2). Cohesion is developed in the asphalt mastic, asphalt binder mixed with fine aggregate, and it would depend on the rheological properties of the asphalt binder. The asphalt mastic’s resistance to microcracking is highly influenced by the dispersion of the mineral filler. Thus, the cohesive strength is controlled by the combination and the interaction of asphalt binder and the mineral filler (3). Water can affect the cohesion of asphalt mastic in several ways such as weakening the mastic due to water saturation and void swelling (4). An asphalt mixture could lose up to 50 percent of its resilient modulus upon saturation but, upon drying, the modulus can be completely recovered (5). This is shown graphically in Figure 1. [Insert Figure 1] Figure 1 Effect of Moisture on Resilient Modulus is Reversible (5). Based on the above discussion, it can be said that the moisture damage in asphalt mixture can be due to a cohesion failure within the mastic, or an adhesion failure at the aggregate-asphalt binder interface. Both failure modes can be related to the characteristics of the asphalt mastic and the asphalt film thickness around aggregate particles. Thus, asphalt mixtures with thin asphalt film tend to fail in tension by adhesive bond rupture whereas asphalt mixture with thicker asphalt film thickness tends to fail in a cohesive failure mode due to the damage within mastic. The determination of asphalt film thickness that differentiate the two modes of failure would depend
on the rheological properties of the asphalt, the amount of damage the asphalt or mastic can withstand prior to failure, a rate of loading, and a temperature at the time of loading (6). Therefore, several test methods were developed to evaluate the loss of adhesion or cohesion between the asphalt binder and the aggregate surface due to moisture-induced damage. However, many of these tests use indirect methods to measure the adhesion and cohesion properties, which may not be very consistent (7). Therefore, a new test procedure (AASHTO TP 91-11: Standard Method of Test for Determining Asphalt Binder Bond Strength by Means of the Asphalt Bond Strength (ABS) Test) was developed for measuring moisture damage characterization based on asphalt-aggregate adhesive and cohesive properties (8). The new method measures the pull-off force needed to break the bond strength between the asphalt binder and the aggregate surface using adhesion testing device. This AASHTO procedure is based on test results using only one type of devices. However, the test procedure should be universally applicable using multiple testing devices. Therefore, in this study, a different adhesion testing device is proposed to improve the consistency of the existing AASHTO procedure. Several studies have reported that, Warm Mix Asphalt (WMA) mixtures are more susceptible to moisture than conventional Hot Mix Asphalt (HMA) mixtures (9). Therefore, this research study used the new modified Asphalt Bond Strength (ABS) test method to investigate the adhesive and cohesive characteristics of WMA mixtures prepared using Poly Ethylene (PE) waxbased additive. This PE wax-based additive works by reducing the viscosity of the asphalt at lower temperatures with a melting point of 100 °C and crystallization point of 90 °C. The PE wax-based additive controls the crystallization so that it does not become brittle at low temperature. Additionally, the PE wax-based additive contains anti-stripping agent to enhance the bonding of asphalt binder to the aggregate surface. In this study, three different forms of the
PE wax-based WMA additive are evaluated to measure their effectiveness in improving the rheological properties of asphalt binders.
2. Asphalt Bond Strength Test Method ASTM-certified adhesion testing device, PosiTest AT-A automatic adhesion tester, was used to perform the ABS test following the AASHTO TP 91-11 procedure. In order to prepare the test specimens, the asphalt binder is attached to the aggregate surface by means of adhesion at controlled environmental and moisture conditions. According to AASHTO TP 91-11 procedure, all aggregate substrates were prepared by cutting quarried rocks using a standard rock saw to create parallel faces and then Lapped using a 280-grit silicon carbide material on a standard lapidary wheel to remove saw marks and ensure a consistent surface roughness. After cut and lapped, samples were cleaned for 60 minutes in ultrasonic cleaner containing distilled water at a temperature of 60o C to remove residual particles on the plate surface (8). To measure the pulloff force needed to detach the asphalt binder samples from the aggregate surface, a hydraulic pressure is applied to pull out a stub attached to the asphalt sample. Figure 1 shows a schematic of the pull-off test method and the PosiTest Pull-Off Adhesion Tester. The AASHTO TP 91-11 procedure recommended new geometry and treatment to the stub surface used with PATTI Quantum Gold adhesion tester in order to create a rough texture that would provide a mechanical interlock and larger contact area between the asphalt binder and stub. This AASHTO procedure is based on test results using only one type of device and one pullout stub of 0.8-mm thickness. Therefore, for this study, three different modified pullout stubs with thicknesses of 0.8 mm, 0.4 mm and 0 mm (no edge) are proposed to be used with the PosiTest AT-A automatic adhesion tester. Figure 2 shows one pullout stub used for PATTI Quantum Gold and three modified stubs used for PosiTest Pull-Off adhesion tester device.
[Insert Figure 2] Figure 2 Schematic of the Pull-off Test Method (left) and PosiTest Pull-Off Adhesion Tester (right) (10). [Insert Figure 3] Figure 3 Pull-out Stubs for a) PATTI Quantum Gold Adhesion Tester, and b) PosiTest® Pull-Off Adhesion Tester. Limestone aggregate and four different binder grades were tested using the three pullout stubs in order to find out which pullout stub should be used with the new adhesion testing device. Only un-conditioned samples were tested in this part of the study.
3. Test Sections with Poly Ethylene Wax-based WMA Additive Three different forms of the PE wax-based WMA additive; pellets, liquid, and specially designed liquid form for mixtures with high RAP amount were evaluated in this study to measure their effectiveness in improving the rheological properties of asphalt binders. The three forms were implemented in three different projects with a rate of 1.5% by optimum asphalt content of the mix. Asphalt binders were extracted from field WMA mixtures from following three test sections. First, the PE wax-based pellet (PE-Pellet) form was applied to a 2.0-inch mill and overlay WMA test section on the southbound outside lane of TH 169 state highway in Champlin, Minnesota. Asphalt binder PG64-28 and Granite aggregate along with 25% RAP by total weight of mix were used to prepare the mix. Another HMA test section was constructed for comparison using the same properties. The HMA and WMA mixtures were designed according to Superpave mix design procedure for a medium traffic level of 3 to 10 million ESALs according to Minnesota Department of Transportation (MnROAD) mix design requirements. Second, the PE wax-based liquid (PE-Liquid) form was applied to a 3.0-inch asphalt layer on State Highway 158 in Lancaster, Ohio. The 3.0-inch layer consisted of intermediate layer with a
thickness of 1.75 inch and surface layer with a thickness of 1.25 inch. WMA and HMA mixes were designed according to Marshall mix design procedure following Ohio Department of Transportation (ODOT) mix design specifications for a medium traffic volume. The mix used a blend of Limestone and Natural gravel aggregates and 25% RAP materials by weight for the intermediate layer and 20 % RAP by weight for the surface layer. Asphalt binder PG70-22 was use for the surface layer and PG64-22 was used for the intermediate layer. Only asphalt binder extracted from the surface layer mixtures was evaluated in this study. Third, the specially designed PE wax-based liquid (SDPE-Liquid) form for mixtures with high RAP amount was applied to a surface layer with a thickness of 1.5 inch on state highway 6 in Iowa City, Iowa. Two WMA and HMA test sections were constructed for a 10 million ESALs traffic volume. The mixes were designed according to Superpave mix design procedure per Iowa Department of Transportation (Iowa DOT) mix design requirements. The mixtures used asphalt binder PG64-28 and Limestone aggregate along with 30% fractionated RAP by binder replacement. Table 1 shows a summary of properties and the materials used in the test sections. Table 1 Summary of the Materials Used in the Test Sections. [Insert Table 1] 4. Modified Asphalt Bond Strength Test Results In order to capture the effect of different pullout stub thicknesses and the asphalt binder grade on the asphalt-aggregate bond strength, test specimens using four different asphalt binder grades; PG58-28, PG64-22, PG64-28M (polymer modified) and PG70-22M (polymer modified) were prepared following AASHTO TP 91-11 test procedure. Asphalt samples, limestone aggregate plates and pullout stubs were heated up at 150° C for a minimum of 30 minutes to remove absorbed water on the aggregate surface and provide a better bond between the asphalt binder and the aggregate surface. The aggregate plates were then brought to an application temperature
of 60° C, whereas the application temperature for the pullout stubs still 150° C. After a sufficient heat-up time, the molten asphalt samples were carefully poured in 10.0 mm diameter Dynamic Shear Rheometer (DSR) silicon molds and left for 30 minutes to reach the room temperature. The asphalt samples were attached to the aggregate plates by placing them on the surface of the pullout stub then firmly pressing each stub with the asphalt sample into the aggregate plate surface until the stub is in contact with the aggregate surface and no excess of asphalt flowing is observed. The prepared samples were then left for 24 hours at room temperature before testing. The ABS test results are shown in Figure 3 with the numbers above each bars representing the average value and the whisker representing the standard deviation. Overlapping of the standard deviations implies the similarity in the measured ABS values between the different types of pullout stubs. As can be seen from Figure 3, the 0.0 mm pullout stub produced significantly higher and more consistent ABS values than 0.4 mm and 0.8 mm stubs. It can be seen from Figure 3 that the bond strength was significantly affected by the asphalt type. [Insert Figure 4] Figure 4 ABS Test Results of Modified Pullout Stubs. In order to measure the actual thickness of the applied asphalt layer, the aggregate plates were cut across the centerline of the asphalt samples and investigated using Olympus SZ61 microscope equipped with Olympus DP26 digital camera as shown in Figure 4 a). The average values of the asphalt layer for 0.8 mm, 0.4 mm and 0.0 mm pullout stubs were 998 μm, 539 μm and 106 μm, respectively. Figure 4 b) shows samples of asphalt layer thicknesses created by the pullout stubs after running the test. As can be seen from Figure 4 b), the measured layer thicknesses were slightly higher than the real stubs’ thicknesses due to the irregularity and pores of the aggregate plates and the groves on the stubs. Asphalt layer thicknesses were measured
after the asphalt samples were exposed to a direct tension during the test, which resulted in a plastic deformation in the asphalt samples. According to AASHTO TP 91-11, samples are considered failed in adhesion if more than 50% of the binder is removed after the test is performed. Otherwise, the failure is cohesive. All the tested samples failed in a cohesive mode except for PG58-28 and PG64-22 asphalt samples tested with 0.0 mm pullout stubs. Figure 4 c) shows a picture of a test sample with 75% cohesive failure with the yellow areas indicating an adhesive failure. [Insert Figure 5] Figure 5 a) Olympus SZ61 Microscope, b) Samples of Asphalt Layer Thicknesses Created by the Pullout Stubs, and c) Test Sample with75% Cohesive Failure. By looking at the measured asphalt layers and the corresponding mode of failure, it can be seen that samples with high asphalt layer thickness always fail in cohesive mode. This explains why all the asphalt samples created using 0.8 mm, and 0.4 mm stubs failed in a cohesive mode. On the other hand, the asphalt samples created by 0.0 mm stub exhibited both cohesive and adhesive failure modes. The ABS values obtained from 0.0 mm pullout stubs were significantly higher than the others because there were more direct contact points between the pullout stub and the aggregate plate. Two-way ANOVA analysis, as shown Table 2, was conducted to evaluate the impact of pullout stub type, the asphalt performance grade and their interaction on the measured ABS values between the asphalt binder and the aggregate surface. It can be seen from Table 1 that, the pullout stub type, the asphalt performance grade and their interaction had significant impacts on the measured ABS value between the asphalt binder and the aggregate surface. It can be concluded that using the 0.0 mm (no edge) pullout stubs with the new adhesion tester device
produced more consistent and reasonable ABS values by providing more direct contact points between the pullout stub and the aggregate surface. Table 2 ANOVA Table of the ABS Test Results. [Insert Table 2] 5. Rheological and Adhesive/Cohesive Characteristics of Extracted Asphalt Binders The asphalt binders of the collected mixtures from the test sections were extracted, re-graded and then evaluated using the Multiple Stress Creep Recovery (MSCR) test following ASHTO TP70, "Standard Method of Test for Multiple Stress Creep Recovery (MSCR) Test of Asphalt Binder Using a Dynamic Shear Rheometer (DSR)." The MSCR test uses the creep and recovery method to measure the percent recovery and non-recoverable creep compliance (Jnr). It can be used to predict the rutting susceptibility of the mix (11). Typically, reducing Jnr value by half can reduce rutting by half (12). The modified ABS test using the PosiTest Pull-Off adhesion testing device and 0.0 mm (no edge) pullout stub was used to evaluate the asphalt-aggregate bond strength of the extracted and recovered asphalt binders in both dry (un-conditioning) and wet (moisture conditioning) situations. According to AASHTO TP91-11, the moisture conditioning process includes soaking the prepared samples for 24 hours in a heated water bath at 40°±2° C followed by leaving for 1 hour at room temperature before testing. The specified conditioning temperature by the AASHTO TP 91-11 procedure can be changed to match the conditioning temperature used for AASHTO T 283 (Standard Method of Test for Resistance of Compacted Asphalt Mixtures to Moisture-Induced Damage) or any other reasonable temperature as needed. However, in this study, the conditioning temperature kept unchanged because the study focused mainly on the feasibility of using the existing test procedure with a new adhesion testing device other than the one was originally used.
5.1. Rheological Properties of the Extracted Asphalt Binder The extraction process was done following the ASTM D2172/D2172M-11 “Standard Methods for Quantitative Extraction of Bitumen from Bituminous Paving Mixtures”, and ASTM D5404/D540M-12 “Standard Practice for Recovery of Asphalt from Solution Using the Rotary Evaporator”. After the extraction and recovery of asphalt binders was done, a full performance grading was done following AASHTO M 320 “Standard Specification for Performance-Graded Asphalt Binder” and AASHTO MP19-10 “ Performance Graded Asphalt Binder Using Multiple Stress Creep Recovery (MSCR) Test” (13). Tables 2 through 4 show summaries of the performance grading results for the different extracted and recovered asphalt binders. The target asphalt performance grade for Minnesota test section with 25% RAP by weight of mix was 70-22. Therefore, due to the high RAP content, virgin asphalt grade was lowered by one level to PG 64-28. The extracted asphalt binders from both HMA and PE-Pellet mixtures of Minnesota test section were tested and graded as PG76-22 for a standard traffic “S” level. Asphalt binder PG70-22 was used for Ohio test section with 20% RAP by weight of mix. The extracted asphalt binders from both HMA and PE-Liquid mixtures of Ohio test sections were tested and graded as PG82-16 for a standard traffic “S” level. The target performance grade for Iowa test section was 70-22 with 30% RAP by binder replacement. According to Iowa DOT, asphalt performance grade should be lowered one level when using more than 20% RAP by binder replacement in the mix and, therefore, PG64-28 was used to produce the mixtures. The extracted asphalt binder from the HMA mixture of Iowa test section was tested and graded as PG82-16 for a standard traffic “S” level whereas the extracted asphalt binder from the SDPE-Liquid mixture was tested and graded as PG70-28 for a heavy traffic “H” level. The SDPE-Liquid additive reduced the impact of aging on the asphalt binder
during construction and improved the high and low temperature grades of the extracted asphalt binder. Further, the SDPE-Liquid additive enhanced the asphalt binder characteristics during the MSCR test by decreasing the non-recoverable creep compliance, which resulted in a higher qualified asphalt binder for a higher traffic level “H” instead of a standard traffic level “S” obtained for the extracted asphalt PG82-16 from the HMA mixture.
Table 3 Rheological Properties of the Extracted Asphalt Binders from Minnesota Test Section. [Insert Table 3] Table 4 Rheological Properties of the Extracted Asphalt Binders from Ohio Test Section. [Insert Table 4] Table 5 Rheological Properties of the Extracted Asphalt Binders from Iowa Test Section. [Insert Table 5] 5.2. Asphalt Bond Strength Test Results The modified Asphalt Bond Strength (ABS) test results are shown in Table 5 and plotted in Figure 5 with the numbers above the bars representing the average values and the whiskers representing the standard deviation. Overlapping of the standard deviation implies the similarity in the measured ABS between the asphalt types. For both HMA and PE-Pellet samples from Minnesota test section, moisture conditioned samples exhibited slightly lower ABS values than the un-conditioned samples. Moisture conditioned PE-Pellet samples exhibited slightly lower ABS values than moisture conditioned HMA samples. Moisture conditioned HMA and PE-Pellet samples from Minnesota test section exhibited higher ABS values than samples from Iowa and Ohio test sections, which indicates a high resistance to moisture damage. Moisture conditioned HMA and PE-Liquid samples from Ohio test section exhibited similar
ABS values and significantly lower than un-conditioned samples. Moisture conditioned HMA and PE-Liquid samples from Ohio exhibited the lowest ABS values, which indicate a low resistance to moisture damage. Moisture conditioned HMA and SDPE-Liquid samples from Iowa test section exhibited significantly lower ABS values than un-conditioned samples. The moisture conditioned SDPELiquid samples exhibited significantly higher ABS values than HMA samples. It can be seen that, both HMA and SDPE-Liquid mixtures are susceptible to moisture damage. However, the SDPE-Liquid showed a better resistance to moisture damage, which can be attributed to the SDPE-Liquid additive’s ability to improve the rheological properties of the asphalt binder. Table 6 ABS Test Results of Extracted Asphalt Binders. [Insert Table 6] [Insert Figure 6] Figure 6 ABS Test Results of Extracted Asphalt Binders. 6. Conclusions The study presented a feasibility of a modified asphalt bond strength test method for use with a new adhesion testing device with three different pullout stubs and its application for evaluating PE wax-based WMA additive. Based on the research findings, it can be concluded that the ABS test can be conducted using a new adhesion testing device with a modified 0.0 mm (no edge) pullout stub. The modified ABS test method was successfully applied to evaluate extracted and recovered asphalt binders from three different test sections. The modified ABS test method showed that the asphalt binder type had an impact on the ABS test results similar to the previous study done using the original ABS test method (7, 8). Additionally, it was found that the SDPE-liquid WMA additive significantly improved both
high and low temperature grades of asphalt. The MSCR test results showed that the SDPE-liquid additive significantly enhanced the non-recoverable creep compliance and the percent recovery of the asphalt binder. Furthermore, the samples prepared using the SDPE-liquid additive exhibited a better moisture resistance than HMA samples. In order to improve the repeatability of ABS test, more tests should be performed on various aggregate surfaces, different asphalt binders, and different conditioning methods and temperatures. Also, the impact of aggregate absorption on the moisture conditioned samples should be included in future studies to see if the aggregate absorption has a significant influence on the measured bond strength using the ABS test method. 7. Acknowledgement The Authors would like to acknowledge the great support of the Korean Institute of Civil Engineering and Building Technology and the Mid-America Transportation Center in funding this research study.
References [1] Little, D. N. Jr.; Jones, D. R. IV., “Chemical and Mechanical Processes of Moisture Damage in Hot-Mix Asphalt Pavements,” National Seminar, TRB Committee on Bituminous– Aggregate Combinations to Meet Surface Requirements Transportation Research Board (TRB), San Diego, CA, February 2003, pp. 37–74. [2] Tarrer, A. R.; Wagh, V., “The Effect of the Physical and Chemical Characteristics of the Aggregate on Bonding,” Strategic Highway Research Program, National Research Council, Washington, D.C, 1991. [3] Kiggundu, B. M.; Roberts, F. L., “The Success/Failure of Methods Used to Predict the Stripping Potential in the Performance of Bituminous Pavement Mixtures,” NCAT Report No. 88-03, National Center for Asphalt Technology, Auburn, January 1988. [4] Terrel, R. L.; Al-Swailmi, S., “Water Sensitivity of Asphalt–Aggregate Mixes: Test Selection,” SHRP Report A-403, Strategic Highway Research Program, National Research Council, Washington, D.C. June 1994. [5] Schmidt, R. J.; Graf, P. E., “The Effect of Water on the Resilient Modulus of Asphalt Treated Mixes,” Proc., Association of Asphalt Paving Technologists, Vol. 41, 1972, pp. 118–162. [6] Hefer, A. and Little, D., “Adhesion in Bitumen-Aggregate Systems and Quantification of the effects of Water on the Adhesive Bond,” Report No. ICAR/505-1, International Center for Aggregates Research, December 2005. [7] Raquel, M; Raul, V.; Bahia, H. U., “Measuring the Effect of Moisture on Asphalt–Aggregate Bond with the Bitumen Bond Strength Test,” Transportation Research Board, Washington, DC, 2009, PP. 70-81. [8] American Association of State Highway and Transportation Officials, “AASHTO TP9-11: Standard Method of Test for Determining Asphalt Binder Bond Strength by Means of the Asphalt Bond Strength (ABS) Test,” 2011. [9] Lee, H.; Glueckert, T.; Ahmed, T.; Kim, Y.; Baek, C.; Hwang, S., “Laboratory Evaluation and Field Implementation of Polyethylene Wax-Based Warm Mix Asphalt Additive in USA”, International Journal of Pavement Research and Technology, Vol.6 No.5 September 2013, pp. 547-553.
32
[10] Defelsko Company, “PosiTest, Pull-off Adhesion Tester Operational Manual,” Obtained from http://www.defelsko.com/adhesion-tester/adhesiontester.htm. Site last accessed July 2014. [11] Federal Highway Administration (FHWA), “The Multiple Stress Creep Recovery (MSCR) Procedure,” FHWA-HIF-11-038, April 2011. Obtained from http://www.fhwa.dot.gov/pavement/materials/pubs/hif11038/tb00.cfm. Site last accessed July 2014. [12] DuBois, E.; Mehta, Y., Nolan, A., “Correlation between multiple stress creep recovery (MSCR) results and polymer modification of binder,” Construction and Building Materials, Vol. 65, 29 August 2014, pp 184–190. [13] American Association of State Highway and Transportation Officials, “Standard Specifications of Transportation Materials and Methods of Sampling and Testing,” 29th Edition [CD-ROM], 2009.
List of Figures
Figure 1 Effect of Moisture on Resilient Modulus is Reversible (5). Figure 2 Schematic of the Pull-off Test Method (left) and PosiTest Pull-Off Adhesion Tester (right) (10). Figure 3 Pull-out Stubs for a) PATTI Quantum Gold Adhesion Tester, and b) PosiTest® Pull-Off Adhesion Tester. Figure 4 ABS Test Results of Modified Pullout Stubs. Figure 5 a) Olympus SZ61 Microscope, b) Samples of Asphalt Layer Thicknesses Created by the Pullout Stubs, and c) Test Sample with75% Cohesive Failure. Figure 6 ABS Test Results of Extracted Asphalt Binders.
Figure 1 Effect of Moisture on Resilient Modulus is Reversible (5).
Figure 2 Schematic of the Pull-off Test Method (left) and PosiTest Pull-Off Adhesion Tester (right) (10).
(a)
(b) Figure 3 Pull-out Stubs for a) PATTI Quantum Gold Adhesion Tester, and b) PosiTest® Pull-Off Adhesion Tester.
Bond Strength at 25°C (77° F) , psi
550.0 488.0
500.0
454.0
450.0
412.3
400.0 350.0
329.0
307.3
300.0 229.0
250.0
263.0 233.1
259.0249.7
200.0 150.0
135.0129.0
100.0 PG 58-28
PG 64-22 PG 64-28 PM Asphalt Type
0.8 mm Pullout Stub
0.4 mm Pullout Stub
PG 70-22 PM
0.0 mm Pullout Stub
Figure 4 ABS Test Results of Modified Pullout Stubs.
Figure 5 a) Olympus SZ61 Microscope, b) Samples of Asphalt Layer Thicknesses Created by the Pullout Stubs, and c) Test Sample with75% Cohesive Failure
Figure 6 ABS Test Results of Extracted Asphalt Binders.
List of Tables
Table 1 Summary of Properties and Materials Used in the Test Sections. Table 2 ANOVA Table of the ABS Test Results. Table 3 Rheological Properties of the Extracted Asphalt Binders from Minnesota Test Section. Table 4 Rheological Properties of the Extracted Asphalt Binders from Ohio Test Section. Table 5 Rheological Properties of the Extracted Asphalt Binders from Iowa Test Section. Table 6 ABS Test Results of Extracted Asphalt Binders.
Table 1 Summary of Properties and Materials Used in the Test Sections.
Site Location Application Type Aggregate type RAP Content, % Binder Type
Test Section #1 Highway TH 169, Champlin, Minnesota 2.0-inch Surface Layer Granite 25 % (by total mix weight) PG64-28 (polymer modified)
Test Section #2
Test Section #3
Highway 158, Lancaster, Ohio
Highway 6, Iowa City, Iowa
1.25-inch Surface Layer Limestone & Natural gravel 20 % (by total mix weight) PG70-22 (polymer modified)
1.5-inch Surface Layer Limestone 30 % (by binder replacement) PG64-28 (polymer modified) Specially Designed Poly Ethylene wax-based liquid (SDPE-Liquid) Superpave
WMA Additive Type
Poly Ethylene wax-based pellet (PE-Pellet)
Poly Ethylene wax-based liquid (PE-Liquid)
Mix Design Method
Superpave
Marshall
Table 2 ANOVA Table of the ABS Test Results. Source Degrees of Freedom Type 1 SS Mean Square Asphalt Performance Grade (PG) 3 120048.2222 40016.0741 Pullout Stub (PS) 2 329435.0556 164717.5278 PG*PS 6 16502.2778 2750.3796
F Value 59.96 246.83 4.12
P-Value > F <.0001 <.0001 0.0055
Table 3 Rheological Properties of the Extracted Asphalt Binders from Minnesota Test Section. HMA PE-Pellet Specification PG 76-22 PG 76-22 Tests on Original Binder Dynamic shear, G*/sinδ (10 rad/s), kPa 1.914 at 76°C 1.78 at 76°C 1.00 Min Tests on Residue from RTFO, AASHTO T240 Mass loss, % NT NT 1.00% Max Dynamic shear, G*/sin δ (10 rad/s), kPa 3.26 at 76°C 2.304 at 76°C 2.20 Min Multiple Stress Creep and Recovery (MSCR) of Asphalt Binder Property
-1
Non-recoverable creep compliance, J Percent difference between J
nr3.2
nr3.2
and J
2.75 at 76°C
, kpa
nr0.1
,J
,%
nrdiff
34.65 at 76°C
Test Method AASHTO T315 AASHTO T240 AASHTO T315 AASHTO TP 70-11
-1
3.405 at 76°C Max, 4.0 kpa HMA: Standard Traffic “S” 25.85 at 76°C Max, 75% PE-Pellet: Heavy Traffic “S”
Tests on Residue from Pressure Aging Vessel, AASHTO R28 @ 100°C Dynamic shear, G*sin δ at 28°C (10 rad/s), kPa 3929.5 at 25°C 3548.5 at 28°C 5000 Max AASHTO T315 Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR) S-value, MPa 161 at -12°C 188 at -12°C 300 Max AASHTO T313 m-value 0.317 at -12°C 0.3055 at -12°C 0.300 Min
Table 4 Rheological Properties of the Extracted Asphalt Binders from Ohio Test Section. HMA PE-Liquid Specification PG 82-16 PG 82-16 Tests on Original Binder Dynamic shear, G*/sinδ (10 rad/s), kPa 2.2219 at 82°C 2.437 at 82°C 1.00 Min Tests on Residue from RTFO, AASHTO T240 Mass loss, % NT NT 1.00% Max Dynamic shear, G*/sin δ (10 rad/s), kPa 3.3917 at 82°C 3.066at 82°C 2.20 Min Multiple Stress Creep and Recovery (MSCR) of Asphalt Binder Property
-1
Non-recoverable creep compliance, J Percent difference between J
nr3.2
nr3.2
and J
2.87 at 82°C
, kpa
nr0.1
,J
,%
nrdiff
70.4 at 82°C
Test Method AASHTO T315 AASHTO T240 AASHTO T315 AASHTO TP 70-11
-1
3.745 at 82°C Max, 4.0 kpa HMA: Standard Traffic “S” 62.85 at 82°C Max, 75% PE-Liquid: Heavy Traffic “S”
Tests on Residue from Pressure Aging Vessel, AASHTO R28 @ 100°C Dynamic shear, G*sin δ at 28°C (10 rad/s), kPa 4399.5 at 28°C 4330 at 28°C 5000 Max AASHTO T315 Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR) S-value, MPa 119 at -6°C 109 at -6°C 300 Max AASHTO T313 m-value 0.372 at -6°C 0.3835 at -6°C 0.300 Min
Table 5 Rheological Properties of the Extracted Asphalt Binders from Iowa Test Section. HMA SDPE-Liquid Specification PG 82-16 PG 70-28 Tests on Original Binder Dynamic shear, G*/sinδ (10 rad/s), kPa 1.363 at 82°C 1.12 at 70°C 1.00 Min Tests on Residue from RTFO, AASHTO T240 Mass loss, % NT NT 1.00% Max Dynamic shear, G*/sin δ (10 rad/s), kPa 2.326 at 82°C 2.6875 at 70°C 2.20 Min Multiple Stress Creep and Recovery (MSCR) of Asphalt Binder Property
Test Method AASHTO T315
AASHTO T240 AASHTO T315 AASHTO TP 70-11 -1 -1 3.95 at 82°C 1.815 at 70°C Max, 4.0 kpa HMA: Standard Traffic “S” Non-recoverable creep compliance, J , kpa nr3.2 SDPE-Liquid: Heavy Traffic Percent difference between J and J , J , % 39.7 at 82°C 33.7 at 70°C Max, 75% “H” nr3.2 nr0.1 nrdiff Tests on Residue from Pressure Aging Vessel, AASHTO R28 @ 100°C Dynamic shear, G*sin δ at 28°C (10 rad/s), kPa 3675.5 at 28°C 4726.5 at 22°C 5000 Max AASHTO T315 Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR) S-value, MPa 80.95 at -6°C 226 at -18°C 300 Max AASHTO T313 m-value 0.367 at -6°C 0.3125 at -18°C 0.300 Min
Table 6 ABS Test Results of Extracted Asphalt Binders. Mix Type/Asphalt Grade Minnesota _HMA/PG76-22 Minnesota _PE_Pellet/PG76-22 Ohio_HMA/PG82-16 Ohio_PE-Liquid/PG82-16 Iowa_HMA/PG82-16 Iowa_SDPE-Liquid/PG70-28
Un-conditioned Samples Average, psi
488.7 490.7 567.0 475.3 490.0 519.7
St. Dev.
Failure Mode
52.3 41.0 16.5 14.5 50.3 12.6
53% Cohesion 73% Cohesion 76% Cohesion 76% Cohesion 76% Cohesion 76% Cohesion
Conditioned Samples Average, psi
426.0 391.3 200.7 192.0 269.7 322.3
St. Dev.
Failure Mode
32.2 22.0 12.5 12.2 38.7 5.5
85% Cohesion 73% Cohesion 77% Adhesion 78% Adhesion 80% Adhesion 65% Cohesion