Bond Strength and Non-Destructive Testing to Identify Delamination
Figure 1. Example of pavement distress caused by debonding.
A strong bond between pavement layers is necessary to ensure long-term performance of newly constructed asphalt pavements and overlays. To create a good bond between asphalt layers, tack coats (typically asphalt emulsions or paving grade asphalt binders) are often used.
As a vehicle moves over a pavement, horizontal friction forces at the contact between tires and the pavement surface induce shear stresses through the pavement. If the induced shear stress is greater than the bond strength at a layer interface, debonding may occur, causing pavement distresses such as slippage failures (Figure 1). Slippage failures can be seen in areas where traffic accelerates, decelerates or turns, inducing high shear stresses. In delaminated areas, the pavement no longer acts as a monolithic structure, and the critical tensile strain will be located at the debonded interface rather than the bottom of the hot-mix asphalt (HMA), causing other distresses, such as fatigue cracking. These pavement distresses require extensive and costly repairs, including patching or complete reconstruction.
Structural Analysis to Characterize Shear Stress Distribution
NCAT researchers conducted a structural pavement analysis to better understand the shear stress distribution throughout the pavement structure. The analysis showed that the critical shear stress at the interface underneath the surface layer is primarily affected by the thickness and stiffness of the surface layer, as well as the variation in stiffness caused by seasonal temperature variations. Subgrade stiffness and total asphalt thickness had less of an effect on the critical interface shear stress.
Surface layer thicknesses from 0.5 to 2.0 inches were analyzed, with thinner surface layers exhibiting higher interface shear stresses. A maximum shear stress of 92 psi was determined for the minimum surface layer thickness of 0.5 inches. Interface shear stress decreased as surface layer thickness increased, with an interface shear stress of approximately 40 psi occurring at a depth of 2.0 inches.
Experiment to Establish Preliminary Bond Strength Requirement
The bond strength between pavement layers can be determined using destructive testing, such as the bond strength test developed at NCAT in the first phase of a study funded by the Alabama Department of Transportation (ALDOT). This method, adopted as ALDOT Procedure 430, shares some similarities to bond strength tests used in Europe and some states, including Florida. In this procedure, cores are placed in a bond strength loading frame and the layers are sheared apart using the Marshall load frame apparatus at a loading rate of two inches per minute and a test temperature of 77°F. Interface bond strength is then calculated by dividing the maximum shear load by the cross-sectional area of the core.
In the second phase of the study sponsored by ALDOT, NCAT investigated bond strengths at five sites that were constructed in the first phase. These pavement sections have been in service for more than four years and show no signs of debonding. Surface layers varied in thickness from 0.7 to 2.1 inches. Tack coats, including asphalt emulsions and a paving grade asphalt binder, were used on new and milled asphalt surfaces and on old Portland cement concrete. The average bond strengths for these five sites were all greater than 100 psi. In addition, nine in-service pavement sections exhibiting slippage failures were also investigated. Cores were extracted in the delaminated areas and from intact areas nearby. Average bond strengths for the intact areas all exceeded 87 psi. In the areas with slippage failures, some specimens broke during coring, and the remainder of the cores from the failed areas had bond strengths ranging from 25 to 60 psi.
Based on the results from the structural pavement analysis and field experiments, a preliminary bond strength requirement of 100 psi, tested according to the ALDOT-430 procedure, was recommended for further evaluation of the interface bond between the surface and underlying layers.
Tack Coat Investigation
Parts of the second phase of the ALDOT-sponsored study focused on evaluating bond strength with regard to tack coat types, application rates, surface preparation and curing time. Four types of emulsions (CRS-2, CRS-2L, CQS-1H and NTSS-1HM) and a paving grade asphalt binder (PG 67-22) were compared at three application rates – low, medium and high, based on existing ALDOT specifications. Specimens were also prepared without tack coats for comparison. The experiment used three surface preparation types: milled, micro-milled and new surfaces.
In the laboratory, slabs were prepared and cored to evaluate bond strength. A field study was also conducted at the test track and several sites across the state to validate the effects of the experimental factors examined in the lab. Cores were extracted following construction, as well as at three months and six months after construction to compare bond strength over time.
Findings from the study include the following:
- Bond strength is improved when the lower surface is either milled or micro-milled.
- Tack coats should be used at higher application rates on milled surfaces—approximately twice the rate used on new asphalt surfaces.
- Each type of tack used at ALDOT-specified application rates can provide an interface bond strength higher than the preliminary requirement of 100 psi, although bond strength was higher for PG 67-22 and NTSS-1HM.
Using NDT to Identify Delamination
Figure 2. GPR unit used in field evaluation at the test track
Poor bond between HMA lifts is difficult to determine before surface pavement distresses become visible. Project R06(D) of the second Strategic Highway Research Program (SHRP 2) is investigating nondestructive testing (NDT) methods that potentially could be used to identify delamination between HMA layers before distress appears. If a rapid NDT method could identify and quantify delaminations following construction or as part of a pavement-management system, then repair or rehabilitation could be considered before problems arise. This ongoing investigation is a collaborative effort between NCAT, Infrasense, the U.S. Army Corps of Engineers' Engineering Research and Development Center (ERDC), and the Center for Nondestructive Evaluation (CNDE) at Iowa State University.
Figure 3. Seismic wave technology used in field evaluation at the test track
The original project goal was to find a technology capable of testing an entire lane width in a single pass at safe operating speeds, but the scope broadened to examine more of the available technologies, such as point-load tests. Several nondestructive testing methods were included in a recent field evaluation at NCAT's test track, where sections both with and without delaminated areas were built in the non-trafficked inside lane. These NDT methods include ground-penetrating radar (GPR) and infrared (IR) thermography, as well as seismic wave and deflection measurement methods. Based on the results, two technologies were selected for further evaluation. A GPR unit manufactured by 3-D Radar provided full-lane width coverage at speeds of 20–30 mph. The other technology, developed by Olson Engineering, uses seismic wave technology with a traveling point-load system.