Highly Modified Asphalt Binders for An Airfield Paving Project in Alaska: PG 64-40 vs PG 58E-34

Abstract: In recent years, interest in highly modified asphalt binders has grown, especially in airfield paving where grade bumping three or four performance grades is not uncommon to accommodate heavy aircraft and high tire pressures. Leaning on this strategy, a recent airfield project in Alaska specified an ASTM D6373 PG 64-40 binder (PG+ requirement of minimum 75% elastic recovery). Relative to the PG 52-28 base grade typical of that region, the PG 64-40 represented two increases of both high- and low-temperature grades. This binder was used to pave an apron that was then grooved much earlier than 30 days after paving (typical timing requirement). The pavement was opened to traffic shortly thereafter, and a C-17 cargo aircraft was towed across the apron, causing considerable groove collapse and surface scuffing. This damage prompted speculation that the PG 64-40 was inappropriate, and one solution proposed for all subsequent paving, including that of the runway, was to replace the PG 64-40 with a PG 58E-34 D8239 grade. Lack of data and experience with these two binders for airfield paving motivated the effort presented in this paper where the overall objective was to assess both binders and determine a path forward.
To accomplish the objective, the PG 64-40 and PG 58E-34 binders were evaluated through binder and mixture testing, with binder data ultimately proving to be more insightful. Mixture testing included wheel tracking using AASHTO T340 Asphalt Pavement Analyzer (APA) airfield protocols (250 lb wheel load, 250 psi hose pressure) and TP145 Rubber Tire Wheel Tracking (RTrack). Surface degradation was evaluated using the Darmstadt Scuffing Device (DSD) in accordance with European standard prCEN/TS 12697-50:2022 Annex B. All mixture tests were performed on regular ungrooved specimens as well as those that were grooved following airfield grooving geometry and tested at 3, 7, and 28 days after compaction to investigate the effects, if any, of simulated traffic-opening time on groove stability. Binder testing included D6373 true grading, master curves, black space diagrams, and multiple stress creep recovery (MSCR) at extended stress levels up to 25.6 kPa (in contrast to the 0.1 and 3.2 kPa levels specified in D8239).
Wheel tracking showed no meaningful distinction in binder grade or grooving condition, while scuffing results showed slightly higher degradation for the PG 64-40. This aligned reasonably well with the binder true grading results, which were similar. If anything, the PG 64-40 appeared on paper to be the preferred choice based on factors such as a greater useful temperature interval. However, advanced characterization, particularly higher-stress-level MSCR testing, suggested the PG 64-40 polymer network broke down more rapidly as stress increased compared to the PG 58E-34. MSCR testing at the 3.2 kPa level conventionally used for highway loadings did not detect this, posing concerns over the appropriate stress level for airfields and other MSCR implementation considerations, which this paper will explore as one outcome of this study. Absent further understanding of airfield-appropriate MSCR protocols, the PG 58E-34 binder was recommended for the airfield project based largely on the high-stress MSCR characterization.