What is the long-term impact of using or not using joint sealants with jointed concrete pavement? There has long been a lack of data to help guide owners in answering this question. In an effort to determine what conditions influence the performance and life-cycle cost of a sealed pavement joint, research was conducted by Dr. Dan Zollinger at the Texas Transportation Institute (TTI) to understand how the initial conditions surrounding the joint affect the joint and the sealant and what that means to long-term pavement performance.

The newly released outcome of this study will be useful to state departments of transportation, contractors, maintenance agencies, and pavement designers in making critical design-related decisions. These decisions must strike a balance between traffic, climate, and base erodibility and how to minimize costs, perform maintenance, and improve performance and pavement serviceability.

The presence of moisture in a pavement structure contributes to distress that can eventually deteriorate the pavement and decrease its service life. This has led to a lot of interest in the effectiveness of joint sealants. In the past 10 to 15 years, there have been both formal and informal studies on the effect of joint sealing, focused on the question of “to seal or not to seal” joints in concrete pavements.

The original test for damaged sealant infiltration at the Riverside Campus of Texas A&M University.
Texas Transportation Institute The original test for damaged sealant infiltration at the Riverside Campus of Texas A&M University.

Zollinger’s research did not attempt to research sealant effectiveness through traditional approaches such as characterizing sealant performance in terms of joint seal properties. Instead, TTI took a more rigorous and fundamental approach to evaluate performance in terms of the amount of water infiltration through the joint and the consequential impacts on subbase erosion and pavement distress. Subbase erosion was used as an indicator of imminent pavement joint faulting. Faulting is a difference in elevation across a joint; it becomes noticeable when the faulting reaches 2.5 mm (0.1 inches).

A field testing program was carried out at the Riverside Campus of Texas A&M University, on SR-59 near Joliet, Ill., and on an experimental section of I-10 in Goodyear, Ariz. The program studied the effectiveness of different sealant types several years after installation when failure begins to affect surface-water infiltration of the joint. They looked at different degrees of failure as represented by different joint openings and bonding conditions.

Testing Approach
This study characterized joint infiltration as a function of both construction quality and environmental factors. The impact of construction was assessed as a function of joint cleanliness, sealant damage, joint movement, and sealant type. The environmental factors were assessed as the predicted infiltration as a function of rainfall intensity, joint geometry, and cross slope. Three sealant types were evaluated: silicone, hot pour, and compression seals.

The impact of joint infiltration on pavement performance was determined by conducting laboratory tests to quantify subbase erosion. This allowed Zollinger to develop a predictive model and supporting software. With these tools an owner can now determine the impact of sealing joints and joint condition on pavement performance. Another aspect of the study evaluated the use of ground-penetrating radar (GPR) to detect moisture under the slabs in the vicinity of the joint.

One question has always been how to determine when to reseal. Historically this was determined by applying weighting factors to sealant condition such as amount of adhesive and cohesive failure or missing sealant. None of these factors, however, address the fundamental properties related to actual performance. With the use of GPR, we can now detect the existence of moisture under the slabs in the vicinity of the joint (from a water infiltration standpoint) and, more importantly, assess when a sealant is no longer effective.

This novel approach used extendable slabs to vary joint width.
Texas Transportation Institute This novel approach used extendable slabs to vary joint width.

Onsite Testing

Cores drilled at pavement joints allowed direct infiltration testing on in-service pavements.
Texas Transportation Institute Cores drilled at pavement joints allowed direct infiltration testing on in-service pavements.

The TTI study tested joints in both controlled field experiments and on in-service pavements. The controlled field experiments, conducted at TTI’s Riverside Campus, were designed to evaluate the effect of sealant damage and joint cleanliness on infiltration rates. For part of the testing a novel approach was developed that allowed the joint opening width to be varied which then could be taken into account along with the joint seal damage.

Testing of in-service pavements occurred on Route 59 in Plainfield, Ill. and on I-10 in Arizona, just west of Phoenix. Testing of in-service pavements allowed the authors to relate sealant effectiveness (measured as infiltration rate) to actual pavement performance. The Illinois highway was four years old and included sealed and unsealed joints. The Arizona project was 20 years old and consisted of 20 test sections including four different base types. The field testing consisted of conducting infiltration and Falling Weight Deflectometer.

Limited GPR testing was also conducted at each of the sites to evaluate the potential for using GPR to detect when the sealant was allowing water to infiltrate the joint. A handheld portable GPR unit was used for this testing. Subbase samples were also retrieved through core holes in the pavement to enable laboratory erosion testing using a Hamburg Wheel Tracking Device (HWTD).

A hand held portable ground-penetrating radar unit.
Texas Transportation Institute A hand held portable ground-penetrating radar unit.

The HWTD test consisted of two component layers. One was a concrete cap on top and the other was the material of interest, which was placed directly under the concrete cap. As a wheel passed on top of both layers, sensors recorded the deflection for each pass. The testing was conducted under wet conditions, which normally results in erosion due to mechanical and hydraulic shearing on the subbase layer generated by slab movement under an applied load. Erosion resistance (ER) is defined as the amount of erosion at 1 million load applications under HWTD erosion testing. The greater the ER number, the lower the resistance a subbase or subgrade material has against erosion.

Research Results
A focus of this study was to link the effectiveness of older joint sealants to when erosion of the subbase would occur and thus potential pavement faulting. A spreadsheet was developed to consider the three main elements of subbase erosion:
(1) the rate of erosion of the base/subbase,
(2) existence of moisture under the slab (as reflected by the number of wet days), and
(3) traffic.

The model was calibrated with lab and field data and is useful for design and maintenance purposes. The findings are:

  • If joint seals are properly installed, they can be very effective in preventing moisture infiltration. Unsealed joints have significantly higher inflow rates than joints with damaged sealants.
  • The water infiltration rates for dirty joints, such as sealants installed in an unclean reservoir or dirt-filled unsealed joints, were as high as those for clean joints with 50% debonding.

Such a tool allows a pavement owner to determine the most effective combination of key pavement design features. From the experimental results, it is clear that the management of a sustainable concrete pavement system requires greater emphasis on performance monitoring rather than on performance repair. This concept is not widely practiced and challenges traditional repair and rehabilitation philosophies. The insights gained from this study support an examination of the commercial viability of such an approach.

Mechanistic-Empirical Fault Prediction Model
A mechanistic-empirical fault prediction model previously developed under National Ready Mixed Concrete Association (NRMCA) funding was improved upon as part of this research. The impact of joint seal effectiveness was directly employed within the fault prediction model.

The erosion resistance of joint sealant materials, the number of wet days, and the traffic load were defined and coupled in this model to effectively analyze the potential for faulting and erosion in jointed concrete pavements. The model uses faulting/erosion as a function of the number of load repetitions in respect to wet days and the erosion resistance of the subbase.

The model can be calibrated for local conditions as a function of distinct characteristics of the subbase or subgrade, which is an important capability in life cycle analysis. The model has been successfully implemented into a spreadsheet format. Results show that the model fits well with the field data and can be implemented for design and maintenance management purposes.

This study confirms that properly installed joint seals can be very effective in preventing moisture infiltration, which can lead to performance issues related to erosion damage. By using the mechanistic-empirical fault prediction model the effectiveness of sealant in pavement sustainability can be determined. With this model, DOTs and contractors can determine how to efficiently and effectively seal joints so the life of the pavement can be prolonged. To learn more about this research and results, a complete version of Zollinger’s report is available at www.sealnoseal.org.

Scott Eilken is a co-chair of the Seal/No Seal Group and owner of Quality Saw & Seal. He is active in the International Grooving & Grinding Association. The Seal/No Seal Group was formed to take responsibility for determining the long-term effectiveness of sealants in concrete pavements. More information on the Seal/No Seal Group can be found at www.sealnoseal.org.