Minnesota Department of Transportation

511 Travel Info

Road Research

NRRA Flexible Team

Impact of Polymer Modification on IDEAL-CT and I-FIT for Balanced Mix Design

Status: Contract development

Project objectives

The overall objective of this study is to determine the impact of polymer modification on the IDEAL-CT and I-FIT results for BMD. Specifically, the study seeks to evaluate two hypotheses for the unexpected IDEAL-CT and I-FIT results indicating better intermediate-temperature cracking resistance for unmodified mixtures than PMA mixtures. The first hypothesis is related to lack of optimization of asphalt binder content in the mixtures tested, while the second hypothesis is related to the limitation that the current IDEAL-CT and I-FIT procedures cannot appropriately demonstrate the benefits of polymer modification. More detailed discussions of the two hypotheses are provided in the Questions and/or Hypotheses section (Section 7).

This proposed study is highly implementable and beneficial to SHAs that are considering or are in the process of implementing IDEAL-CT or I-FIT as an intermediate-temperature cracking test for BMD. Anticipated research outcomes will: (1) provide these agencies with guidance on more strategic use of PMA for improving the intermediate-temperature cracking resistance of asphalt mixtures, (2) contribute to further enhancement of IDEAL-CT and I-FIT in case the current test procedures cannot appropriately demonstrate the performance benefits of polymer modification, and (3) provide mix designers with guidance on the selection of asphalt binders for improving the IDEAL-CT and I-FIT results for BMD.

Project tasks

Task 1: Project kick-off meeting

  • Deliverable: Meeting notes & revised workplan

Task 2: Material selection and acquisition

Three mix designs will be tested, and each will be evaluated with three virgin binder grades and three asphalt binder contents. A total of 27 asphalt mixtures will be tested with the DSR torsion bar modulus test and IDEAL-CT or I-FIT. Furthermore, 15 selected mixtures will be characterized using the DSR torsion bar fatigue test and a battery of rheological tests on extracted asphalt binders. The proposed rheological tests include PG grading, Delta Tc, MSCR, Glover-Rowe parameter, and LAS.

  • Deliverable: Proposed material selection

Task 3: Work plan execution

The proposed work plan consists of four laboratory experiments as discussed below.

Experiment 1 – Determination of T=G* using the DSR Torsion Bar Modulus Test

The objective of this experiment is to determine the T=G* of asphalt mixtures with different virgin binder grades and asphalt binder contents using the DSR torsion bar modulus test. A total of 27 asphalt mixtures will be tested, which correspond to a combination of three mix designs, three virgin binder grades, and three asphalt binder contents. The test will be conducted in accordance with ASTM D7552, which measures the |G*| and phase angle (δ) values of a mixture torsion bar specimen at isothermal temperatures ranging from -30 to 50°C and various discrete frequencies covering 0.1 to 100 rad/s. The shear strain will be adjusted throughout the test to ensure that the specimen is subjected to sufficient displacement while remaining in the linear viscoelastic range. Three replicates will be tested. For data analysis, master curves will be generated by fitting the |G*| and δ results to discrete relaxation and retardation time spectra using the RHEATM software. For each mix design, the torsion bar |G*| of the unmodified mixture at the volumetric OBC that corresponds to a binder complex modulus between 12 and 60 MPa will be used to select the T=G* of asphalt mixtures with different PG grades and asphalt binder contents. Because of the increased mixture stiffness, using a PMA binder or lowering the asphalt binder content is expected to increase the T=G* of a mixture.

Experiment 2 – IDEAL-CT and I-FIT Testing at 25°C and T=G*

In this experiment, IDEAL-CT will be conducted on asphalt mixtures prepared with the 20% RAP mix design from Alabama and the virgin mix design from Wisconsin, while I-FIT will be conducted on asphalt mixtures prepared with the 14% RAP plus 2% RAS mix design from Wisconsin. Both tests will be conducted at two test temperatures: 25°C and T=G* determined in Experiment 1. A minimum of four replicate samples with 7.0 ± 0.5% air voids will be tested at each temperature. To avoid the introduction of asphalt aging as a compounding factor, both IDEAL-CT and I-FIT tests will be conducted on short-term aged specimens per AASHTO R 30. In addition to the final cracking test parameters CTindex and FI, interim test parameters determined from the load-displacement curve, such as the peak strength, fracture energy, and post-peak slope, will also be analyzed. The analysis of variance (ANOVA) general linear model will first be conducted to determine the impacts of polymer modification, asphalt binder content, test temperature, and their interactions on each of the IDEAL-CT and I-FIT parameters.

Then, the IDEAL-CT and I-FIT results at 25°C will be analyzed to evaluate the first hypothesis discussed previously. If the results consistently show a similar trend, then the first hypothesis is approved. This indicates that a sufficient amount of asphalt binder is needed to demonstrate the positive impact of polymer modification on the intermediate-temperature cracking resistance of asphalt mixtures. In this case, no modifications to the current IDEAL-CT and I-FIT procedures are needed. To improve the cracking resistance of a “dry” mix for BMD, a two-step performance optimization process that consists of using polymer modification and increasing the asphalt binder content would be needed.

If the first hypothesis is disapproved, then asphalt binder content will have no impact on the IDEAL-CT and I-FIT results of comparing PMA versus unmodified mixtures. In this case, modifications to the current test procedures are needed to capture the cracking performance benefits of polymer modification. Test results obtained at T=G* will then be analyzed to evaluate the second hypothesis. Two key questions that need to be answered are, “Can the test results at T=G* demonstrate better intermediate-temperature cracking resistance for PMA versus unmodified mixtures?” and “Can the test results at T=G* show an intuitive sensitivity to changes in asphalt binder content?” If positive answers to both questions can be obtained, then the proposed modification of conducting IDEAL-CT and I-FIT at T=G* instead of 25°C is preliminarily justified.

Experiment 3 – Validation of IDEAL-CT and I-FIT with Two Cyclic Fatigue Tests

The overall objective of this experiment is to further validate IDEAL-CT and I-FIT with two asphalt binder and mixture cyclic fatigue tests: the DSR torsion bar fatigue test and the LAS test. The DSR torsion bar fatigue test will be conducted in accordance with a standard operating procedure developed at MTE as a research tool for comparative testing of binder formulations and forensic studies. The test is a time sweep test at a prescribed stress level and temperature. Stress levels ranging from 250 to 1,000 kPa will be evaluated through preliminary testing. Test data will be modeled using the VECD theory to generate C-versus-S curves, which provide a fundamental relationship between stiffness and material integrity and allow the prediction of cycles to failure (Nf) at a given strain level.

The LAS test will be conducted on extracted asphalt binders without additional Rolling Thin Film Oven (RTFO) or Pressure Aging Vessel (PAV) aging so that they will be subjected to the same aging condition as the IDEAL-CT and I-FIT specimens. Testing will be conducted in accordance with AASHTO TP 101 using two replicates. Test results will be analyzed using the continuum damage approach. The major outcome of the test is a relationship between the fatigue parameter (Nf, normalized to 1 million ESALs) versus the applied shear strain as a pavement structure indicator. At a given strain level, asphalt binders with a higher Nf value are expected to have better resistance to fatigue damage than those with a lower Nf value.

Both the DSR torsion bar fatigue and LAS tests will be conducted at 25°C and T=G* (determined in Experiment 1), but only using 15 selected asphalt mixtures that are prepared with unmodified binders at three asphalt binder contents and PMA binders at the volumetric OBC only. For each test, comparing the Nf results at the volumetric OBC will indicate the impact of polymer modification on the fatigue resistance of asphalt binders and mixtures, while comparing those at different asphalt binder contents will indicate the impact of asphalt film thickness. The foremost analysis of this experiment is to compare the DSR torsion bar fatigue and LAS results against the IDEAL-CT and I-FIT results to determine whether they can provide consistent conclusions regarding the intermediate-temperature cracking resistance of PMA versus unmodified binders and mixtures within each mix design.

Experiment 4 – Supplementary Binder Rheological Testing

In addition to the LAS test, extracted asphalt binders in Experiment 3 will be characterized with a battery of rheological tests to determine their PG grades, Delta Tc, MSCR Jnr and percent recovery, and Glover-Rowe parameters. Low-temperature PG grade, Delta Tc, and the G-R parameter will be determined using the 4mm DSR setup. Each test will be conducted with two replicates. Data analysis will be conducted in two aspects. First, the binder TIT-PG will be determined from the PG grade per Asphalt Institute recommendations and compared against the mixture T=G* determined from the DSR torsion bar modulus test in Experiment 1. Second, the correlation between the various binder rheological test parameters and the IDEAL-CT and I-FIT results will be evaluated. The rheological test parameters showing a strong correlation with the mixture test results will be identified. This correlation analysis will provide mix designers with guidance on the selection of asphalt binders for improving the IDEAL-CT and I-FIT results for BMD.

  • Deliverable: Determination of T=G*; IDEAL-CT and I-FIT Testing; Validation of IDEAL-CT and I-FIT; Supplementary Binder Rheological Testing

Task 4: Project close-out meeting

  • Deliverables: Quarterly progress reports for duration of project

Task 5: Final report

  • Deliverable: Final report

Project team

Principal Investigators: Fan Yin, f-yin@auburn.edu; Fan Gu, fzg0014@auburn.edu; and Raquel Moraes, moraes@auburn.edu, National Center for Asphalt Technology
Co-PI: Andrew Hanz, MTE Services, Inc. (fka Mathy Construction), Andrew.Hanz@mteservices.com
Technical Liaison: Michael Vrtis, MnDOT, michael.vrtis@state.mn.us
Project Technical Advisory Panel (TAP) – Email the TAP
Contact us to join the TAP

  • Brandon Brever, MAPA
  • Andy Cascione, Flint Hills Resources
  • Jerry Geib, MnDOT
  • Stacy Glidden, Walbec Group
  • Brian Hill, Illinois DOT
  • Hassan Tabatabaee, Cargill