Test Methods, Test Results & Specifications Pavement Testing: · Surface Texture: Circular Texture Meter (CT Meter) ASTM E-2157 The equipment conforms to the Standard Test Method for Measuring Pavement Macro-texture Using the Circular Track Meter. The CTM uses a laser to measure the profile in an 800-mm circumference circle. The mean depth of texture for each 100-mm segment of the arc is computed according to ASTM standard practice. The averages of depths of the two arcs perpendicular to the traveled direction and the two arcs parallel to it are computed. · Friction: Dynamic Friction Tester (DFT) ASTM E-1911 The equipment conforms to Standard Test Method for Measuring Paved Surface Frictional Properties Using the Dynamic Friction Tester. The DFT has three rubber sliders that are spring-mounted on a horizontal rotary disk at a distance of 350mm. The disk is initially suspended above the pavement surface and is driven by a motor until the tangential speed of the sliders is 90 km/h. Water is then applied to the pavement surface by the device, whereupon the motor is disengaged and the disk is lowered to the test surface. The three rubber sliders contact the surface and the friction force is measured by a transducer as the disk spins down. The friction force and the speed during the spin down are saved to a file. This results in a continuous spectrum of dynamic coefficients of friction. The equipment reproduces actual speeds between 0-80 km/hr) and surface bearing loads of vehicles commonly in use. · International Friction Index: as calculated from CTM and DFT measurements · Friction: ASTM E-274, Standard Test Method for Skid Resistance of Paved Surfaces Using a Full-Scale Tire · Permeability: NCAT Field Permeameter · Infiltration: Method Of Test For Determining The Quantity Of Asphalt Rejuvenating Agent Required For An Asphaltic Pavement (Ring Test) Caltrans CT 345 · Change in Modulus: Spectral Wave Analysis with Portable Seismic Pavement Analyzer (PSPA) Laboratory Testing on Cores or Mix Specimens: · Dynamic Creep Test on Rectangular Specimens from Field Cores (DSR) · G* - Dynamic Shear Rheometry (DSR) on binders extracted from core slices · Static Bending Test on rectangular specimens from field cores, Bending Beam Rheometer (BBR) · Laboratory Permeability on field cores Product (Emulsion) Testing: · Saybolt-Furol Viscosity AASHTO T59 · Particle Size · Surface Tension Product (Emulsion Residue) Testing: · Chemical Analysis (Rostler) · Chemical Analysis (Instrumental Methods) Lessons Learned Summary: For detailed reports on each of the test methods, see Appendix 4. For photographs of the test methods, see Appendix 3. o Field friction testing (the International Friction Index from the DFT and CTM) was repeatable, and the results consistent. Initial testing included six individual runs of each test per test section. As repeatability was verified, this was reduced to three runs per section. Additional runs were made only if the original three results were not in close agreement. o The portable devices (International Friction Index - DFT and CTM) gave similar relative rankings as the standard trailer testing (ASTM E-274), as evidenced by comparisons on the tests run in 2006 on MN State Route 251 and County Road 112. CR 112 was a new, very open coarse Superpave mix, and 251 an aging dense-graded surface (except for the chip seal section). The graphs below show the comparison. o It is important to thoroughly sweep off loose sand before running DFT and CTM tests. Loose sand will increase the surface texture, and thus improve calculated IFI measurements, when if fact it is not a stable surface conformation from a skid point of view. Comparison of International Friction
Index and Full-Scale Tire Testing 
Comparison of International Friction
Index and Full-Scale Tire Testing 
o Tire Spin Test (visual observation). When in doubt, improvise. A simple observation of friction can be made by rapidly accelerating a vehicle from a standing stop, and then safely braking to a hard stop within the test section. If tires spin without any traction or if the vehicle slides on braking, one might take relevant precautions for inadequate skid. It is interesting to note, however, that the Reclamite sections seemed very slick in this test, even though all friction test measurements were comparable to other sections. o To be effective, fog seal emulsions must infiltrate into the pavement surface. It was hoped that permeability would be a key measure for both determining application rate of sealers and their effectiveness over time. However, results from the field permeameter (NCAT) were not reliable on highly permeable surfaces like CR 112 because of incomplete sealing. o The Ring Test is a bit subjective, but gives a quick and easy indication of the relative ability of an emulsion to infiltrate into a pavement surface. o Laboratory permeability on field cores may be a better measure for predicting emulsion infiltration, and for evaluating the finished seal’s ability to keep water out of the pavement. The graph shows test results from lab testing on cores from the Arizona project. The tests were run at the North Central Superpave Center. Clearly, the chip seal was most effective at sealing out the water (lowest permeabilities). The fog seal emulsions generally did a better job than the rejuvenator oil of sealing the pavement against moisture. In all cases, the CSS treated sections had better sealing than the controls. Lab Permeability Test on Cores taken in 2006 – Minnesota Projects 
Lab Permeability Test on Cores taken in 2006 – Arizona and California Projects 
o While there is probably an ideal emulsion viscosity to have adequate film thickness and infiltration into a pavement, the surface tension of the emulsion in contact with the pavement surface is a better indicator of the ability of the emulsion to infiltrate the surface. Because this testing was initiated very late in this study, there was insufficient data collected in this study to reach any definitive conclusions about optimal viscosity or surface tension. o The particle size of the emulsion should equate with its ability to enter pores in the pavement surface. Emulsified rejuvenator oil particles should be easier to deform and enter pores than more rigid higher viscosity asphalt particles. Only limited particle size data was collected for this study. o There was much scatter in the data from the Spectral Wave Analysis with Portable Seismic Pavement Analyzer (PSPA). It is difficult to tune the instrument to the very shallow depth (3/8 inch) needed to evaluate in-place surfaces for critical low-temperature mechanical properties. This test does offer promise as it is a rapid non-destructive method to time fog seal applications, but it needs further development before being used as a standard tool. The results indicated that the change in modulus between depths of 1 in. (which is the upper resolution of the device) to 4 in. was either small or insignificant. The rheological tests on the cores run at WRI and MTE indicate that 1 in. is too deep to capture the strongest aging effects. Furthermore, the fog seal emulsions rarely infiltrate into the pavement more than 0.5 in. Therefore, use of this device was abandoned early in the project. o G* - Dynamic Shear Rheometry (DSR) on binders extracted from thin (0.3 in) core slices indicated significant softening of the surface materials by sealers containing rejuvenator oils, with G* rankings in order with expectations given the amounts of rejuvenator oil and polymer in the respective sealer formulations. The graphs below demonstrate the impact of various sealers/rejuvenators. It is important to remember that Reclamite is a rejuvenator oil, the products Pass QB, CRF, ERA-1, and ERA-25 contain various blends of rejuvenator oil with asphalt, and the products GSB-B (modified) and CSS do not contain rejuvenator oil. The MN 251 surface has an extremely low permeability (<1 x10-4 cm/s), and only the top slice shows evidence of softening, essentially in the order of rejuvenator content in the emulsion residues. Sealers with no rejuvenator had little impact on the binder modulus. Binders extracted from the second slice of each sealed section fell within experimental error of the second 0.3 mm layer from the control section that was never sealed. The CA 78 surface was more permeable. Products with higher rejuvenator content substantially softened the surface layer, but there is also some softening in the underlying second layer. Rheology of Extracted Cores for Two Test Sites  
(Core slices 0.3 in thick) The hard residue of the Gilsonite modified GSB actually increased the modulus of the extracted binder, but this product is intended to be a surface sealer. The emulsion residue may not impact the rheology of the asphalt cement in the pavement. It is not known if the rejuvenators have penetrated into the pavement surface and actually blended with the aged binder. Hence, mixture tests are needed to clarify findings from binder extractions. · Dynamic Creep Test on Rectangular Specimens from Field Cores (DSR) uses very thin mixture specimens to measure parameters related to mixture stiffness. Flow time to 5% strain at 58°C and 68kPa was used here, but other useful parameters are also reported in the MTE test reports (Appendix 4). Testing single thin-sliced (3/8 in) specimens results in significant variability, but multiple replicates provided excellent information on the reduction in stiffness of the treated sections and the evolution of hardening over time. It also differentiated those rejuvenators that had an effect on (had penetrated into) the lower layer of the binder. In all cases, the surface layer was stiffer than the second layer, as shown in the graph below. Dynamic Creep Rheology on Mix Core Slices (core slices 0.375 in thick) 
Although binder extractions as exhibited by the above graphs shed some light on depth of penetration, these results can be very misleading with respect to actual performance of the rejuvenator in the pavement. When blending any aged asphalt with softening agents, one must always question if the aged mix acts as a black rock, or if the binders uniformly blend to yield the rheology of the extracted binder. The graph below compares findings from extracted binder rheology to those of the mixture Creep Recovery Test for two projects, MN 251 and AZ 87. First note from the permeability data reported in previous graphs that the MN 251 pavement is essentially impermeable (<1 x 10-4 cm/sec), whereas the AZ 87 HMA mix is quite permeable (90 x 10-4 cm/sec). The MN 251 pavement offers a perfect example for the need to evaluate mixture properties directly rather than trying to extrapolate performance from extracted binder properties. WRI DSR tests of G* on extracted binders from MN 251 indicated significant softening in the thin surface layer when sealers contained rejuvenator oils. However, the Dynamic Creep Test on thin mixture specimens did not confirm this rejuvenating effect. These results seemed puzzling until lab tests reported this pavement to be essentially impermeable. Clearly, rejuvenator oils were applied, but could not infiltrate to soften the oxidized asphalt below the surface. Only direct mixture testing could accurately characterize the mixture in the zone where block crack initiation is likely to occur. Additionally, the sealer products with hard residues did not appear to harden the asphalt near the surface even though the extracted binders properties suggested they might do so. It appears that mechanical strength tests for mixtures are much preferable to extracted binder tests when evaluating the thin surface layers of pavement cores. Key lesson learned: If the emulsion can not infiltrate/penetrate the pavement, excess residue remaining on the surface exacerbates friction loss without providing a definitive improvement in pavement life. Fortunately, results for rejuvenator seals applied to the more permeable Arizona 87 dense mix were much more in line with expectations that rejuvenator seals can penetrate into the pavement surface and soften aged asphalt. As can be seen in the lower portion of the graph below, the Reclamite rejuvenator emulsion did soften the pavement surface layer as would have been predicted by the binder extraction. In fact, the rejuvenated binder was even softer than the second thin pavement layer in both cases. As expected, the harder emulsion residues in the sealer products had much less impact on the rheology of the surface layer, although they do appear to help retard oxidation when applied to the right pavements at the right time. Comparison of Dynamic Shear Rheometry on Binder from Extracted Cores and on Mix (Core Slices) using Torsional Bar Geometry from MN 251 and AZ 87 Projects 
It is also interesting to follow the progression of age hardening with time. Since MN 251 is essentially impermeable to water, two important questions need to be addressed. “Can the rejuvenator oils ever penetrate into the surface?”, and “How deep will oxidation progress into this pavement?” The latter question can be answered by reviewing the graph below. Dynamic Creep Recovery tests were run on 3/8” thick mix specimens cut from the top and second layers of test sections at time periods of one month, two years, and four years after the initial fog seal application in 2002. In each case, the stiffness of the top layer as estimated by the time to 5% strain increased substantially over the four year period. The same finding was true for the second layer, although the stiffness after four years was still substantially less than that of the top layer. The three products containing rejuvenator oil exhibited less stiffness than the control in the top layer, and less difference between the top layer and the second layer. Hence, these products do appear to absorb into the relatively impermeable pavement over time to as least offer some protection to the top layer. The CSS-1h sealer with a harder residue did not appear to help. Under the Gilsonite-modified GSB, the top layer was even stiffer than in the control section four years after treatment. The chip seal did appear to protect even this impermeable pavement from rapid oxidation. After four years, the top and second layers of the underlying HMA had similar stiffnesses as those of the rejuvenator treated sections. Dynamic Creep Recovery (DSR) on Specimens Cut from Cores from MN 251 
The graph below offers a different view of the same data in the graph above. Because this was not a new pavement when sealed in 2002, the surface had already oxidized. As can be seen from the one month data, the impermeable surface was not softened much even by rejuvenators one month after application, and the second layers were all essentially the same. As noted above, the softening effect of the rejuvenator oils was much more evident four years after application. The following conclusions can be drawn from these findings: o Even the extremely low pavement permeability did not stop oxidation in either the top or the second layer. o Applying sealers with harder binders to further seal the pavement still did not slow oxidation from the rate observed in both layers of the control section. Dynamic Creep Recovery (DSR) on Specimens Cut from Cores from MN 251 
As shown above with DSR test results on binders and mixes, oxidative aging is very easy to follow by measuring properties related to stiffness or shear modulus at high pavement temperatures where changes in rheology are dramatic. However, the specific distresses normally attributed to asphalt aging are surface raveling, block cracking, and shortened fatigue life. Conventional wisdom typically cites Kandhal studies that correlate low temperature ductility at 15°C with age-induced pavement distress. Surface raveling and pitting occurs when ductility falls below 10 cm, block cracking starts when ductility falls below 5 cm, and surface cracking becomes severe when ductility falls below 3 cm. Hence, pavement damage related to age-embrittlement is more likely to occur at low to ambient temperatures. Although the mechanism for block cracking is not well understood, recent research suggests that asphalt’s low temperature stiffness does not change much upon oxidative aging, but related relaxation and fracture properties change dramatically. Recent laboratory aging studies conducted at Texas A&M and pavement aging studies led by Western Research Institute (WRI) suggest that changes in the low temperature phase angle or the relaxation parameter “m-value” are more important than stiffness. As one example of the observed evolution in binder rheology with aging, the BBR “m-value” drops, thereby shifting the low temperature rheology to increasing “m-control”. Because extracted binders do not always represent the true rheology of the binder near the surface, a BBR test on thin mixture specimens was selected to evaluate the effectiveness of sealers at protecting and/or restoring low temperature mixture properties. · The Static Bending Test on Rectangular Specimens from Field Cores Using the Bending Beam Rheometer (BBR) This procedure developed by Dr. Marasteanu at UMinn cuts thin mix specimens to standard BBR test geometry, and then applies a 500 g load at typical BBR temperatures (-6°C to – 18°C in this study). The stiffness and m-value of the mix are determined, from which low temperature binder properties can be estimated if needed. This method would seem a priori to be a good way to determine the change in m-control given by the treatment seals. BBR test results for cores taken from MN 251 and AZ 87 are shown in the graph below. As expected, the Reclamite product containing only rejuvenator oil did reduce low temperature stiffness vs the control, and the second application reduced it even more. However, each Reclamite application also appeared to reduce the m-value, which is contrary to expectations that softening the asphalt at low temperature would also improve its relaxation properties. Hence, the asphalt appears to be more “m-controlled” after each application of rejuvenator. Because BBR testing was initiated late in the project, this data set is limited, and any conclusions drawn from this data lack statistical confirmation. However, these findings do open the door to the possibility that other additives or oils might do more to enhance low temperature relaxation/fracture properties in aged binders. Further research is clearly needed in this area. Polymer in the CRS-2P(d) and Pass QB appear to have had a slightly positive effect on the m-value. The best protection from aging as measured by m-value was accomplished by chip sealing the pavement so that less oxygen could enter the mix from the surface. Given the variability of testing such small mixture specimens, each of these conclusions is only marginally significant and needs further verification. Although promising for conventional dense HMA mixes, BBR data from the open, asphalt rubber surface course on the California Salton Sea project had extremely high coefficients of variation and could not be used to draw statistically valid conclusions. Static Bending Test on Rectangular Specimens Cut from Field Cores (BBR) 
o Chemical analysis. The Western Research Institute report details several chemical analyses that were done on the products used for the trial. From the field trial results, it is believed that performance testing is of more value than chemical testing. Their report is in Appendix 4. Specifications Specification development was outside of the scope of this study. However, it is strongly recommended that performance-related specifications be developed, and it is hoped the results of this study will be useful in that endeavor. Future Testing Recommendations: There were several topics beyond the scope of this project that warrant further study. These include: § Develop a simpler, more effective field permeability test applicable to fog seals. § Develop relationships between emulsion properties (surface tension, particle size, viscosity) and pavement permeability which can predict infiltration of the emulsions into the pavement surface. § Define a procedure for determining optimum application rates. § Determine if a pay item test strip can improve performance and safety. § Define sand quality: angularity, maximum moisture content. § Understand whether vehicle control on newly sealed sections is adequately predicted by IFI. § Develop and verify performance-related specifications. o Determine aged binder properties that lead to block cracking. o Define desired physical properties of the surface following application. o Define emulsion residue properties in physical or chemical terms for an emulsion purchase spec. o Develop a model to predict emulsion infiltration into the pavement, using Permeability; Emulsion Surface Tension, Particle Size, Viscosity. o Define construction criteria: § Application Rates; Coverage; Permeability drop. § Release to traffic: minimum friction index, break time (no tracking). Full test reports can be found in the links on the right pane of this page and in the The Project Library. |