Field Validation of Performance-Based Polymer-Modified Emulsion Residue Tests the FLH Study

C H A P T E R 4 : PA P E R 2 6

Field Validation of Performance-Based Polymer-Modified Emulsion Residue Tests:

The FLH Study

Gayle KingGHK, Inc., United States

Helen KingGHK, Inc., United States

Larry GalehouseNational Center for Pavement Preservation, United States

Michael VothFederal Lands Highway Division, Federal Highway Administration, United States

Laurand LewandowskiPRI Asphalt Technologies, United States

Chris LubbersKraton Polymers, Inc.

Paul MorrisParagon Technical Services, Inc.

A B S T R A C T

While Superpaves performance-based test methods and specifications revolutionized the characterization ofhot mix asphalts, they are not directly applicable for emulsion-based pavement preservation applications.The Federal Lands Highway (FLH) division of FHWA initiated this study to evaluate polymer emulsion residuerecovery and physical characterization specifications that correlate with field performance. Several laborato-ries tested newly proposed methods on field samples from chip seal and micro surfacing FLH field projectsconstructed in 2008. The labs used low temperature forced draft oven methods to recover emulsion residueto better simulate field curing. To determine resistance to rutting and bleeding, G* and sin were obtainedfrom dynamic shear rheometer (DSR) frequency sweeps on the residues using standard Superpave protocols.

247

P A P E R 2 6

Compendium of Papers from the First International Conference on Pavement Preservation

Creep compliance and percent residue recovery were determined via multiple stress creep recovery (MSCR)testing. Rheological tests were run to measure resistance to low temperature cracking, with Bending BeamRheometry (BBR) as well as DSR frequency sweeps at intermediate temperature with associated low temper-ature modeling. For resistance to aggregate loss on original and PAV-aged residue, participants ran strainsweep tests at 25oC and measured loss in G*. The investigators ran sweep tests (ASTM D 7000) using projectaggregates and emulsions to determine chip seal curing time. To validate the test procedures and determinefailure limits, FLH will track field performance for a minimum of three years. The results presented are thefirst entries into a database needed for development of performance-based specifications for asphalt emulsions.

K E Y W O R D S

Pavement preservation, Polymer Modified Emulsions, chip seal, slurry seal, micro surfacing, performance-based specifications

P R O B L E M S T A T E M E N T

The Federal Lands Highway Division (FLH) of the Federal Highway Administration (FHWA) found that whileFLH has much experience with best practices using conventional asphalt emulsions, there was no definitiveguide for selecting, specifying and using polymer modified asphalt emulsions. FLH and FHWAs Office ofAsset Management initiated a project to develop such a guide. The project included extensive literature search-ing and gathering of best practice information from government, industry and academic experts and practi-tioners, as well as interfacing with several on-going related research projects. Most of the asphalt emulsion testmethods and specifications currently used were developed well before the implementation of performance-based tests now standard in hot mix asphalt (HMA). The experts agreed updated test methods and specifi-cations using newly available tools could provide both more accurate characterization and hopefully bettercorrelation to field performance for all paving asphalt emulsions. While there has been some effort to developemulsion performance-based specifications (Walubita, Lubinda, Epps-Martin, & Glover, 2005), the industryhas not yet accepted a general specification. The Transportation System Preservation (TSP) Research, Devel-opment, and Implementation Roadmap published in 2008 identified a need for performance-based emul-sion specifications. (TSP2, 2008)

The Strategic Highway Research Project (SHRP) developed Superpave, which included performance-based tests and specifications for hot mix asphalt binders. Those are almost universally implemented through-out the U.S. Superpave tests were developed to address rutting at high temperatures, cracking at lowtemperatures and fatigue at intermediate temperatures. Sample preparation protocols were developed to sim-ulate the aging that occurs in a hot mix plant and in relatively thick asphalt pavements on the road. Pavementpreservation emulsion applications do not see the same temperature conditions during manufacture and con-struction, and the distresses are not the same as those for hot mix. While rutting and cracking may be appli-cable for micro surfacing, such phenomena as aggregate shelling and binder bleeding are more of a concernfor all emulsion treatments. (King, Lesueur, Planche & King, 1993) Earlier researchers developing emulsionperformance-based specifications also found resistance from the industry because of the length of time andadded expense of performance-based testing and equipment.

There are several projects by well-respected researchers currently studying asphalt emulsions. Pooling thelearnings and sharing data from these projects can prevent duplication of effort and facilitate test methodand specification development.

248


O B J E C T I V E S

While the primary focus of the FLH study was the field guide of best-practice recommendations and devel-opment of an initial draft, report-only specification for the use of polymer modified asphalt emulsions (PME),the project provided an excellent opportunity for furthering development of emulsion performance-basedspecifications. The ultimate objective is a standard performance-based specification for asphalt emulsions.That was, however, outside the scope of this project. Within this study, the objectives were to bring togetherthe best current knowledge to start a database of laboratory and field results useful for the eventual specifica-tion development. The goals are to tie the test results to the performance of specific emulsion applications,minimize the exposure of emulsion residue to excess heat and agitation (which are not present in the field)during laboratory testing, and maximize the use of the Dynamic Shear Rheometer (DSR) to replace all otheremulsion residue test equipment.

S C O P E

Based on current literature and information gathered from researchers of other on-going projects (ASTMCommittee D 4.42, in progress) (Shuler and Epps-Martin, in progress) (Bahia and Sebaaly, in progress) (Kim,in progress) (Turner and Harnsberger, in progress) (Moulthrop and Hicks, in progress), several test methodswere proposed for a draft specification. The draft specification was revised after further consultation with in-dustry groups, including the Asphalt Emulsion Manufacturers Association (AEMA), FHWAs binder experttask group, the Transportation Research Board (TRB) asphalt and pavement preservation committees, theInternational Slurry Surfacing Association (ISSA) and the Asphalt Recycling and Reclaiming Association(ARRA). An experimental testing protocol was then put together to run the tests in the draft under a varietyof conditions as well as other tests for verification. Because there is no current national standard for manypolymer-modified asphalt emulsions, and the FLH study recommended polymer modification for all theirapplications, the specification was designed specifically for PMEs; however, the residue test methods shouldbe applicable for all emulsions.

Sample Preparation: Residue Recovery

Field performance depends upon the physical properties of the cured emulsion residue. Emulsions cure at am-bient temperatures in the field. The high temperatures of current residue recovery methods (distillations andoven evaporations) change the binders rheological properties, typically cutting the modulus in half by heat-ing the sample to 350F (177C). (Hanz, Arega & Bahia, 2009) Phase angles from high temperature distilla-tion residues also suggest that heating can cause cross-linking and damage to polymer additives. Therefore, itis generally agreed that recovery should simulate field curing. Such industry groups as AEMA, ASTM andEuropean agencies are evaluating methods of recovering emulsion residue without undue heat or agitation,including the Forced Draft Oven (FDO) procedure, the Stirred Can Test and the Moisture Analyzer. A low tem-perature (140F, 60C) FDO using a silicone mold is preferred, because the residue can be easily removed fromthe mold without reheating; it is run at conditions most closely simulating field conditions; and it has givenacceptable results according to inter-laboratory reliability testing and comparison of residue and base asphaltproperties. The method has been adopted by ASTM, as D7497 - 09 Standard Practice for Recovering Residuefrom Emulsified Asphalt Using Low Temperature Evaporative Technique.

249


Rheology

Researchers have found rheology is the best way to characterize the performance properties of polymeric andhigh float asphalt gel structures, as well as asphaltic materials. Rheology is the study of deformation and flowof matter, specifically quantifying the viscous and elastic components. The dynamic shear rheometer (DSR)and bending beam rheometer (BBR) are now used for specifying hot mix asphalt, and, with some adapta-tions, are recognized as the most potentially valuable tools for specifying emulsion residues.

Superpave Rheology Methods for Hot Mix Asphalt: DSR and BBR

Performance-based properties for hot mix asphalt binders are determined by the DSR for a minimum strengthmodulus (G*/sin ) and a maximum phase angle (), as well as BBR for low temperature brittleness (S, stiff-ness, and m-value, relaxation rate). An asphalt residue should be stiff (it should not deform too much), itshould be elastic (it should be able to return to its original shape after load deformation), and it should notbe brittle at low temperature (it should not crack or lose its bond with aggregates). The higher the strengthmodulus is, the better the asphalt residue is able to resist deformation. The lower the phase angle, the more itis able to recover its original shape after being deformed by a load. A low stiffness and a high relaxation rateare desirable at low temperatures for resisting cracking, raveling and shelling. For hot mix, DSR testing isdone at the high pavement temperature (Th) at the project site, and BBR testing is done at a temperature tocharacterize the rheology at the low pavement temperature.

DSR Strain Sweep

Takamura (2005) suggests that asphaltic binders that lose strength when tire contact moves an embedded chipare a major cause of chip loss and raveling. Polymers create additional tensile strength with elongation, pullingthe aggregate back to its original position when the tire (loading) has passed. This is particularly importantfor problem areas such as intersections or driveway exits where turning tires are most prone to dislodge chips.Rather than adding an expensive tensile test to the specification, he theorized that this property can be meas-ured with a DSR strain sweepincreasing the strain applied to a binder at constant temperature and fre-quency and determining when the modulus (strength) drops a given percentage. Bahia, Hanz and Jenkins(2008) furthered that work.

New DSR Methods: Multi Stress Creep Recovery and Low Temperature Characterization

The recently developed DSR test method in the Multi Stress Creep Recovery (MSCR) mode has been foundto give more accurate characterization of percent recovery (recoverable strain) and Jnr (compliance or in-verse of stiffness) of polymer modified asphalts. (Bahia, Hanson, Zeng, Zhai, Khatri, & Anderson, 2001) (DAn-gelo & Dongre, 2009) Lower Jnr means the residue is more resistant to deformation, flow or bleeding.

Christensen, Anderson, and Marasteanu (Marasteanu & Anderson, 1996) showed that rheological mas-ter curves of modulus (G*) versus temperature and phase angle versus temperature can be mathematicallymodeled using the now well-accepted CAM model. According to this model, rheological measurements at onetemperature range can predict rheological properties at a very different temperature. (Marasteanu, Basu, Hesp& Voller, 2004) Bahia is developing a procedure using the CAM model to test DSR frequency sweep data atintermediate temperatures, then use that master curve to predict low temperature emulsion residue physicalproperties (G* and phase angle). (Pavement Preservation Expert Task Group, 2008) In this frequency sweepmode, the temperature and strains are kept constant as the loading time or frequency is varied.

250


Turner and Harnsberger (in progress) at Western Research Institute are also studying the feasibility ofreplacing the BBR with DSR testing for low temperature characterization; they are using smaller plates at thelow temperatures used in the BBR.

Aged Sample Preparation: Pressure Aging Vessel

AASHTO R 28, Accelerated Aging of Asphalt Binder Using a Pressurized Aging Vessel (PAV), is believed to bethe best alternative for simulating long-term aging because it is run at a reasonable temperature simulatingfield conditions, and because it is a proven AASHTO test method. Rheological tests on PAV residue shouldcharacterize low-temperature behavior after aging (brittleness, cracking, aggregate loss and raveling potential)and characterize the modified binder as it ages.

Optimizing Testing Costs and Time

One of the reasons earlier attempts at emulsion performance-based specifications have failed is the concernthat performance-based testing will be more time-consuming (two or more days) and result in shipping, con-struction and acceptance delays. Suppliers also do not want different specifications and pre-certificationrequirements for different geographic regions or markets. Similar concerns with Superpave technology re-sulted in an Approved Supplier Certification Program for allowing shipping binder from authorized suppli-ers before testing was completed. The FHWA Pavement Preservation Expert Task Group (ETG) has assigneda sub-committee which is in the process of writing a supplier pre-certification or delayed-acceptance pro-gram for emulsions. (Pavement Preservation Expert Task Group, 2008) This will be fully coordinated with theSuperpave binder and mix ETGs, and advanced to the AASHTO Highway Subcommittee on Materials andAEMA/ARRA/ISSA for their consideration.

Replacing the BBR with one of the DSR methods discussed above for low temperature characterization,as well as running the DSR strain sweep for adhesion loss, will also reduce equipment and testing costs as wellas testing time. Work is also in progress to use DSR methods to characterize polymer elasticity and to definethe non-linear rheological behavior typical of high float emulsion residues.

The Draft Specification

Table 1 gives the draft specification for the emulsion residue testing protocol based on the collected informa-tion.

251


Emulsion Application Performance-Related Tests: Sweep Test and Wet Track Abrasion

ASTM D7000 Standard Test Method for Sweep Test of Bituminous Emulsion Surface Treatment Samples usesproject aggregate and emulsion to determine compatibility of the chip seal emulsion and aggregate, indicat-ing how quickly the adhesion and chip retention develops. Following recommendations from Takamura ofBASF, the ASTM procedure was modified slightly to improve reproducibility. Changes include: Felt pad preheated to 35C (95F) in oven prior to use. Aggregate surface dampened with about 4 grams (0.14 oz) of water prior to spreading onto the emulsion

on the sweep test pad. 12-in by 14-in (30.5 by 35.6-cm) rectangular felt pads in place of circular pads. ISSA TB 100 Wet Track Abrasion of Slurry Surfaces is a standard test for performance-related micro surfac-ing design recognized as correlating well with performance.

Project Testing Protocols

The goal is a performance-based specification using a testing protocol that is efficient, reliable and accuratelycharacterizes the field behavior. The ultimate specification, with a target testing cost of $1000 per individualcertification, will only use the test conditions needed for a specific application. For this study, field project sam-ples were tested over a broad range of temperature and loading conditions to give a better understanding ofmaterial properties, the feasibility of the draft specification and potential specification limits. Table 2 gives thefull protocol, listing the selected tests from those described above. In several cases, such as the low tempera-ture analysis, the data was collected for use by researchers in other projects. Field project samples were alsosent to other researchers for their related projects. All of the data collected has been posted on NCPPs web-site. (National Center for Pavement Preservation, 2009) It is expected that the data will begin a database usedby other researchers for the development of emulsion performance-based specifications.

252

Table 1. Draft Performance-Based Specification - PME Residue

C H A P T E R 4 : PA P E R 2 6 253

Table 2. Testing Plan Protocols

% Recovery & Jnr at each stress level

% Recovery & Jnr at each stress level


The Field Projects

This testing was run on samples from three FLH field projects during the summer of 2008. A scheduled 2009project will also be tested.

The field projects constructed for this study include numerous project sites, at least three emulsion sup-pliers and multiple contractors. Climates ranged from very hot and dry (21-mile, 34-km, SBR latex modifiedasphalt chip seal project at Death Valley National Park) to cold and wet, as well as extreme temperaturerangesan 11-mile (18-km) chip seal at Dinosaur National Monument which spans the borders of Utah andColorado, and 90 miles (145 km) in the Utah Parks project of SBR latex modified emulsion chip seal andnatural rubber latex (NRL) modified micro surfacing to locations in Arches National Park, CanyonlandsNational Park, Natural Bridge National Monument, and Hovenweep National Monument. A summary of theprojects and testing information is given in Table 3. The participating laboratories included the Central Fed-eral Lands High Division (CFLHD), PRI Asphalt Technologies, BASF Corporation, Paragon Technical Serv-ices, Inc., and SemMaterials, LLC.

254

Table 3. Field Projects and Test Laboratories


T E S T R E S U L T S A N D D I S C U S S I O N

Recovery of Emulsion Residue by Forced Draft Oven

Some of the inter-laboratory residue tests did not give acceptable agreement. Further investigation revealedthe testing labs used slightly different procedures for the Forced Draft Oven. Some labs used a silicone moldfor obtaining residue for all residue tests, while at least one lab used PAV pans for all testing. These discrep-ancies will need to be resolved for the final specification.

Residue Aging by Pressure Aging Vessel (PAV)

This study prepared the samples for PAV by running the 48-hour FDO in the same PAV pans to be placed inthe PAV. The residue from the completed PAV was then scraped and tested in the DSR, with minimal to noreheating or agitation required.

There are still some issues. Sufficient emulsion must be placed in the PAV pan to allow adequate filmthickness of the FDO cured emulsion for the standard PAV test. There is some question if all the water is evap-orated during the FDO run in the PAV pans. It may be necessary to use thinner films which age more quickly.The 100C (212F) standard PAV temperature exceeds high pavement temperatures, which may alter curedpolymer structure and/or cause temperature-induced coalescence of recovered asphalt droplets in the residue.The procedure appears to be viable, but more data needs to be collected to determine the optimal conditionsfor aging time, film thickness and temperature for a given application.

Residue Testing Residue Before and After PAV Aging

Multiple Stress Creep Recovery

Figure 1 show plots of PRIs test results for Jnr (compliance) versus the four tested stress levels at the three highpavement test temperatures. Low Jnr indicates resistance to flowrutting and bleeding. The differences inJnr for the three chip seal emulsion residues were extremely high. For a stress of 3200 kPa applied at 64C, Jnrvalues were 1.2 for Utah Arches (CRS-2L-UT), 5.7 for Death Valley (CRS-2L-DV), and 32.1 for Dinosaur Na-tional Monument (Pass Emulsion). When grading HMA binders, a doubling of the Jnr represents a softeningby approximately one full binder grade. This rule of thumb would suggest that the CRS-2L-DV (Death Val-ley) is more than two grades softer than the CRS-2L-UT (Utah Arches) residue, and the Dinosaur NationalMonument Pass emulsion residue another two or three grades softer yet. This range seems excessive, and thegrades as used have no relation to the high temperatures for the respective climates. These surprising resultsaccentuate the need for urgency in developing performance-based emulsion specifications.

Figure 2 gives the test results for the MSCR percent recovery versus the four tested stress levels at the threetest temperatures, as tested by PRI. Again, however, there were huge differences in performance, particularlyat the higher stress levels and temperatures as recommended by FHWA for hot mix asphalt binders. Using astress level of 3200 Pa at 64C, the recoveries ranged from 0.7% for Pass (Dinosaur), 8.6% for CRS-2L-DV(Death Valley), 10.1% for CRS-2L-UT (Utah Arches), and 17.2% for Ralumac micro surfacing (Utah).

255


The rejuvenator/elastomer polychloroprene (Neoprene) product used for the Dinosaur project (Pass) isnot only very soft, but it also exhibits an almost gel-like tendency to completely lose elasticity as the stressincreases. In fact, the emulsion contains an oil designed to soften (rejuvenate) the underlying oxidized pave-ment surface, and a polymer designed not to be swollen by the rejuvenator oil. At 100 Pa and lower test tem-peratures (58 and 64C), it has the best recovery of the three chip seal emulsions; however, at 3200 Pa, thePass exhibits virtually no elasticity at any test temperature.

256

Figure 1. MSCR - Jnr vs. Stress for FDO Residues at 58, 64 and 70C


Figure 2. MSCR Recovery vs. Stress for FDO Residues at 58, 64 and 70C

It seems probable at this time that no single performance-based specification for emulsion chip sealresidues could possibly cover the breadth of consistency and elasticity as evidenced by the elastomeric styrene-butadiene latex emulsions (CRS-2L) and the rejuvenating elastomeric Neoprene product (Pass). Independ-ent performance-based specifications will be needed to define their respective residues.

For the PAV aged residues, the Jnr (compliance) results were consistent and ranked in the same order astheir unaged counterparts, with the exception of the Pass emulsion, which was unable to be tested at the givenconditions because it was still very soft after aging. Lab work is ongoing to understand testing issues thatresulted in problematic data.

257


Figure 3. Effect of Aging on MSCR Recovery at 64C

Figure 3 shows the MSCR recovery results before and after aging. Percent recoveries usually improve fol-lowing PAV aging. First, harder residues produced through the aging process naturally exhibit better recov-ery at a given temperature. Second, some elastomeric polymers may cross-link to some degree during aging.This cross-linking should strengthen the polymer network and improve elasticity. The CRS-2L-UT productimproved in elasticity relatively more than the other latex-modified products. It should be noted that poly-mers composed of butadiene cross-link (increase in molecular weight) during thermal and oxidative agingwhile isoprene containing polymers (natural rubber latex) will chain scission (decrease in molecular weight)or break-down during thermal and oxidative aging.

Each of the products show significant increases in the percent recovery with aging, but the relative changeis quite different. For example, unaged recoveries for the Ralumac micro surfacing residue are considerablyhigher than those from the CRS-2L-UT. However, after aging, the percent recoveries of the two products arealmost equal under most test conditions.

The evolution in Jnr with PAV aging (20 hr, 100C) was evaluated for three of the four products. For theintermediate test conditions of 64C and 3200 Pa, the Jnr fell with aging as follows:

Ralumac: from 1.92 to 0.25. CRS-2L-DV: from 5.5 to 0.66. CRS-2L-UT: from 1.19 to 0.19.

This would suggest that the CRS-2L-DV residue may have experienced some changes in the polymer net-work structure and/or more severe asphalt oxidative aging during the PAV step. More work is needed to un-derstand how the variables of time and temperature impact aged properties in the PAV oven as compared tofield aging. Based upon previously cited rules of thumb, the Ralumac and CRS-2L-DV hardened by threehigh-temperature grades in the PAV (approximately 18C, 64F change in equi-service temperature), and theCRS-2L-UT hardened by approximately 2 grades (15C, 59F).

Figure 4 shows MSCR test results from three laboratories. The data is not very consistent. There was somequestion on the labeling of samples from the Utah parks project, which included both micro surfacing and chipseal emulsions. While the Ralumac results from SemMaterials and BASF are in agreement, they do not agreewith PRIs results. More work needs to be done to standardize the testing for better inter-laboratory agree-ment.

The percent recovery has a very strong dependence on the compliance (inverse modulus), or Jnr of theresidue. For example, the micro surfacing emulsion (Ralumac) has a recovery of 25.9 percent at 58C, 17.2percent at 64C, and 11.1 percent at 70C. The high susceptibility of the MSCR percent recovery to temper-ature is a disadvantage for specifications; it will always be possible to improve acceptance results somewhat by

258


making the residue harder rather than by adding polymer to improve recovery. Of course temperature-dependence is also a problem for the current methods to define polymer elasticity, such as ASTM D 6084Standard Test Method for Elastic Recovery of Bituminous Materials by Ductilometer Elastic Recovery (ER),which is run at a single temperature regardless of grade. New performance-based specifications could changethe test temperature at some standard increment with climate temperature, whereas the classic ER test is al-ways run at a single temperature regardless of grade. Further research might consider running MSCR at lowertest temperatures where recoveries would be higher and possibly less sensitive to temperature. Figure 5 showsthe changes in Jnr and recovery with temperature at 3200 Pa.

259

Figure 4. MSCR Jnr and Recovery Results From 3 Laboratories


Figure 5. Effect of Temperature on MSCR Jnr and Recovery at 3200 Pa

As expected from newly developed HMA binder grading protocols, Figure 5 confirms that the residue Jnrapproximately doubles with each 6C incremental increase in test temperature. There is every reason tobelieve it will be possible to use the climate maps created in LTPPBind (LTPP Products On-Line, 2007) to de-fine and select appropriate emulsions grades for a given locale. However, the test conditions and specificationlimits must be adjusted to best fit the application.

One important reason for replacing the traditional high temperature PG grading parameter G*/sin deltawith the MSCR parameter Jnr is that the latter enables the product specifier to select test conditions under thehigher stress conditions associated with high traffic volumes on HMA or turning rubber tires on chip seals.Earlier PG specifications based only upon conventional asphalt could assume that asphalt is a linear viscoelasticmaterial, and therefore the asphalt modulus G* should be constant for both all applied strain rates and allapplied stresses. The MSCR test as developed during NCHRP 9-10 (Bahia, Hanson, Zeng, Zhai, Khatri &Anderson, 2001) showed clearly that these fundamental assumptions do not apply to polymer modifiedasphalts. Nonlinearity is particularly evident for the softest materials including PME residues, as shown inFigure 6.

260

Figure 6. Effect of Stress on MSCR Jnr for CRS-2L-DV


The Jnr for all products increases with applied stress at all temperatures, but the relative non-linearity asexpressed by the slopes varies dramatically from one PME residue to another. All PME residues get softer asincreasing load is applied, but the amount of load-induced softening is highly dependent upon the amountand type of polymer, as well as the grade of the base asphalt. Because the Pass emulsion residue is very soft,it is particularly sensitive to this stress softening effect at high temperatures. The widening gap in Jnr as tem-perature increases is consistent with the fact that softer materials exhibit more non-linearity. Higher appliedstress results in higher strains, while softer materials or hotter liquids yield more at any given stress. Hence,increasing stress, increasing temperature, or softening the base binder all push the results further into thenon-linear region. This effect, when viewed from a chemists point of view, is really a strain dependent issuerelated to the polymer structure. Very long polymer molecules entangle much like long hair tangles. These en-tanglements enable the polymer network to resist flow to a degree much higher than its molecular weightalone would imply. However, as these tangled chains are stretched and unwound, the additional elasticity pro-vided through chain entanglement (increased entropy) is lost. Hence, the polymer network becomes weakerand less elastic as it is stretched to the point that chains begin to disentangle. These effects are tied to thehigher applied strains, regardless of cause (higher stress, higher temperatures or softer base asphalts). Sincepolymers can vary widely in molecular weight, chain length, branching and molecular structure, the strain atwhich these effects become important can vary dramatically. This is not surprising; the behavior is much thesame as woven fabric being much stronger than the individual threads.

The effects of increasing applied stress on percent recovery are considerably more dramatic than thoseimpacting Jnr. As mentioned above, recovery is always reduced when higher stresses result in higher strainswhich dislodge polymer chain entanglements. However, the percent recovery for the Pass emulsion at 64C fellfrom a relatively high 28.8 percent to less than one (see Figure 2) when the applied stress was increased from100 to 3200 Pa. The SBR latex modified CRS-2L residues were also highly sensitive to stress, but maintainedreasonable elasticity even at the highest stress levels. It is also interesting to note that the percent recovery forCRS-2L-DV at different temperatures is surprisingly insensitive to applied stress up to 3200 Pa. CRS-2L-UTand Ralumac show moderate declines in percent recovery as temperature increases, while percent recovery forthe Pass emulsion is extremely sensitive to both temperature and applied stress. A simpler climate-based grad-ing system for percent recovery could be one possible solution.

Low Temperature Bending Beam Rheometer Testing and Continuous Grading

AASHTO T 313 BBR tests were run at two temperatures on the FDO residue. The tests were then used to pre-dict the temperature at which the samples passed the criteria of 300 MPa Stiffness (S) and 0.300 m-value. Theresults, given in Table 4, show that the low temperature grading of the SBR and natural latex-modifiedemulsions were similar, meeting the limiting requirements at -28.8, -30.6 and -26.3C. The neoprene latex-modified Pass emulsion is much softer, as was indicated in the MSCR testing, with a low temperature of -34.7C.

261


Because PG binders are graded in 6C temperature increments, it is easiest to understand differences inasphalt consistency by comparing the temperatures at which materials have the same consistency as meas-ured by the current PG standard, G*/sin delta. Because those using PG specifications are familiar with the tem-perature as defined for HMA applications using a frequency of 10 radians per second and a specificationlimiting modulus of 1.0 kPa for unaged binders, these test conditions were used to define comparable limit-ing temperatures for the emulsion residues. Although not in this report, it should be emphasized that fullfrequency sweep data is available on the NCPP website for all unaged and aged samples at high and interme-diate temperatures, so rheological master curves can be constructed and/or limiting temperatures can bedetermined at other test conditions ultimately deemed appropriate for chip seal applications. As can be seenfrom the data in Table 4, limiting temperatures for the unaged residue from the three chip seal emulsionsranged from 54.6C (Pass) to 81.8C (CRS-2L-UT), a difference of 27.2C or 4 PG binder grades. It is quitesurprising that the two extreme binders were both applied to Utah National Parks during the late fall of 2008.The emulsion (CRS-2L-DV) applied during the same period in Death Valley, CA, one of the hottest locationsin the US, had a consistency near the mid-point of the range between the two Utah projects. This range ofconsistencies seems illogical and accentuates the need for improved emulsion residue specifications. From aresearch point of view, the broad range of properties might accelerate differences in performance to bet-ter select specification limits in the future.

For reasons discussed earlier, it is the goal of the project to investigate the use of intermediate tempera-ture frequency sweeps as a means of replacing the Bending Beam Rheometer as the preferred method for spec-ifying the low temperature performance properties of emulsion residues. The data has been collected and ison the NCPP website.

262

Table 4. Bending Beam Tests and Rheology Limiting Temperatures


Because low temperature properties are best defined in performance-based specifications after the binderis subjected to laboratory aging protocols, frequency sweeps were run on all PAV aged residues at 10C and20C using procedures as designated for intermediate temperature PG Binder grading (8 mm plates, 2 mm gap,5 percent strain, 0.1 to 100 radians/second). All frequency sweep data tables can be found on the NationalCenter for Pavement Preservations (www.pavementpreservation.org) website.

Figure 7. Strain Sweeps on PME Residues

Dynamic Shear Rheometer Strain Sweep

DSR strain sweeps were run on all PAV residues using the test conditions recommended by Bahia (25C, 8-mm plates, 2-mm gap, 10 radians/second, 0.01 to 50 percent strain). As can be seen on Figure 7, logarithmicplots of modulus (G*) versus % strain indicate that the modulus remains relatively constant as strain increases,and then weakens dramatically as the strain exceeds some critical limit. Full strain sweep data is available onthe NCPP website mentioned earlier, and detailed data tables have been forwarded to other researchers for theirfurther analysis as part of ongoing industry efforts to develop performance-based specifications.

Sweep Test

Split samples of emulsion and aggregate from the Utah Arches, Death Valley, and Dinosaur Monument chipseal projects were sent to three participating laboratories for sweep testing. Five single-lab replicates were runusing a two-hour curing period for every trial. The results are given in Table 5.

263


Results from inter-laboratory sweep tests were encouraging, but some questions remain. As can be seenfrom Table 5, intra-laboratory results for the CRS-2L-UT and the CRS-2L-DV were very consistent, with 5-replicate standard deviations ranging from 0.4 to 2.0 percent mass loss. This precision should satisfy needsfor an acceptable specification test. The inter-lab precision is less encouraging. Average results for the ArchesCRS-2L-UT ranged from 11.1 to 16.5 percent mass loss, suggesting there is still room to improve the defini-tion of procedural details in the ASTM draft. For the Death Valley CRS-2L-DV, two labs reported results thatwere within intra-lab variability, but the third lab reported results that were unacceptably high.

Finally, the Pass emulsion did not cure sufficiently in two hours to hold chips, so mass loss was essentially100% and testing was abandoned. It should be understood that the residue from Pass emulsion containsrejuvenator oils, and is therefore very soft. Furthermore, the emulsifier is designed to break more slowly thantypical CRS-2L emulsions. This kind of product has found an important niche in the marketplace, particu-larly when applied to low ADT, highly aged bituminous surfaces that need rejuvenation to prevent furthersurface-initiated cracking. There were no reported problems with this emulsion when it was applied on thefield project. On the other hand, Pass may not be an appropriate emulsion for chip sealing roads with highvolume traffic or for projects that need early cures to minimize traffic control issues. Hence, such a productwould need independent performance-based specifications written for the applications where it is found tobe successful.

C O N C L U S I O N S

Literature searches, information gathering from industry, academic and government experts and a sur-vey confirmed there is a need for emulsion performance-based specifications. A draft specification wasdeveloped using newly developed techniques for setting time (sweep test), emulsion recovery (Forced DraftOven test) and rheological characterization (Dynamic Shear Rheometry compliance and recovery in the Multi-

264

Table 5. Sweep Test Results


Stress Creep Recovery Mode, in Frequency Sweeps and in Strain Sweeps). Samples from field trials were testedusing the new protocols. The Frequency Sweeps were run as a proposed method of using intermediate tem-perature rheology testing with master curve analysis to characterize low temperature behavior, eliminating theneed for expensive and time-consuming Bending Beam Rheometry testing. Preliminary results are mixed, andthe data collected is being shared with other researchers to characterize and specify the performance of themodified residue. There is still work to be done; this initial testing showed inconsistencies that need to be rec-onciled. Of particular note is the discrepancy between softer material used on a field project in a high tem-perature zone and harder material in a colder zone. It is hoped that other researchers, suppliers and usersshould benefit from the results obtained by this testing plan, and it is envisioned that performance-basedspecifications for polymer modified asphalt emulsion surface treatments will be the norm in the not too dis-tant future.

This project has begun leveraging available knowledge and pooling information (test methods, data, andpavement performance) with suppliers and other researchers and agencies (Federal, State, City and County).Because of the high interest by several entities in developing improved emulsion test methods and specifica-tions, an expert task force (ETF) of the Pavement Preservation ETG was been formed by FHWA in 2008. Bycooperating on testing procedures and round robin testing, researchers from several projects will be moreeffective in developing standard procedures. It is recommended that governmental agencies support the cre-ation of a pooled fund study to continue the performance-based testing begun here on future AASHTO agencyfield projects.

Specific areas identified as currently needing more investigation include: Provide clearer differentiation of material performance given variability in climate (temperature, humid-

ity) and traffic. Update asphalt emulsion test methods in ASTM D-244, including measures for laboratory and field vis-

cosity and low-temperature residue recovery, as well as the performance-based testing. Continue the development of rheological methods to insure the presence of optimum levels of polymer

modification or gel (high float) formation in the residue. Improve materials selection, including aggregate specifications and mix-design procedures. Create Delayed Acceptance or Certified Supplier Programs for asphalt emulsions.

Although problems with curing might be visible shortly after construction, ultimate performance cannotbe analyzed until many years later. FLH collects video pavement management data every three years. Morefrequent field inspection may be needed as the draft performance-based tests and ranges are compared tofield performance with time. Tying the field performance information over time to the test results should bean on-going process. A Materials Library of the tested materials should also be maintained, so that materialsmay be retested as the test methods are perfected and pavement performance is known.

In conclusion, current activities of this and other projects are now being fully coordinated with the FHWAPavement Preservation ETG and with the FHWA Superpave ETGs to advance recommendations to theAASHTO Highway Subcommittee on Materials, with the ultimate goal of asphalt emulsion performance-based specifications for more consistent, high quality emulsion applications.

265


A C K N O W L E D G M E N T S

This work is being accomplished with the help of many entities. The authors would especially like to thankthe support, guidance and patience from the staff of the Central Federal Lands Highway division and the Fed-eral Highway Administrations Office of Construction and System Preservation who greatly encouraged andfacilitated the information sharing among researchers, suppliers and specifiers. We also thank the many, manymaterial suppliers, agency users, academic and State Highway researchers and independent testing labs whogave their support, expertise, testing and materials (much of which was donated), to further the goals of thisproject. We would especially like to thank Jim Sorenson, Koichi Takamura, Arlis Kadrmas, Hussein Bahia,Patte Hahn, John Johnson, PRI Asphalt Technologies, BASF Corporation, Paragon Technical Services Inc.,Kraton Polymers U.S. LLC, and SemMaterials, LLC. Members of the Binder Expert Task Group, various Trans-portation Research Board committees, and the Asphalt Emulsion Manufacturers Association InternationalTechnical Committee have also reviewed proposed methods and shared their knowledge. The authors areexcited about the prospects for the future as many researchers and field practitioners are now sharing infor-mation and working together to develop polymer modified emulsion specifications that will provide higherquality, longer lasting and lower cost surfaces for Public Lands and highway users.

R E F E R E N C E S

Bahia, H., Hanson, D.I., Zeng, M, Zhai, H., Khatri, M.A., and Anderson, M., Characterization of Modified Asphalt Binders in

Superpave Mix Design, NCHRP REPORT 459, Transportation Research Board National Research Council, National Academy

Press, Washington, D.C., 2001

Bahia, H., Hanz, A., and Jenkins, K., Performance Grading of Bitumen Emulsions for Sprayed Seals, Mid-Continent Transportation

Research Forum, Madison, Wisconsin, August 14-15, 2008.

Bahia, H. and Sebaaly, P., Emulsion Cold Mix (Asphalt Research Consortium), in progress

DAngelo, J. and Dongre, R. N, Practical Use of Multiple Stress Creep Recovery Test: Characterization of Styrene-Butadiene-Styrene

Dispersion and Other Additives in PMA Binders, TRB 88th Annual Meeting Compendium of Papers DVD, 2009

Hanz, A., Arega, Z., and Bahia, H., Rheological Evaluation of Emulsion Residues Recovered Using Newly Proposed Evaporative Tech-

niques, TRB 88th Annual Meeting Compendium of Papers DVD, 2009

Kim, R., Chip Seal Design and Performance, North Carolina DOT Project HWY 2004-04, in progress

King, G.N., Lesueur, D, Planche, J.P., and King, H. Using SHRP Test Procedures To Evaluate High Float And Polymer Modified

Bitumen Emulsion Residues, Proceedings, First World Emulsion Congress, Paris, 1993

LTPP Products On-Line, LTTPBIND software download 2007, Retrieved June 15, 2009 from http://ltpp-products.com/index.asp

Marasteanu, M., Basu, A., Hesp, S., and Voller, V., TimeTemperature Superposition and AASHTO MP1a Critical Temperature for

Low-temperature Cracking, International Journal of Pavement Engineering, 1477-268X, Volume 5, Issue 1, Pages 31 38, 2004

Marasteanu, M. and Anderson, D., Time-Temperature Dependence of Asphalt Binders An Improved Model, AAPT, vol. 65, p. 408,

1996

266


Moulthrop, J. and Hicks, G., Slurry/Micro-Surface Mix Design Procedure, Caltrans Contract 65A0151, in progress

National Center for Pavement Preservation, Polymer Modified Emulsion Study, Retrieved June 15, 2009 from

http://www.pavementpreservation.org/fhwa/pme09.php

Pavement Preservation Emulsion Task Force, Meeting Minutes, December 15, 2008, Lexington, Kentucky, Retrieved June 15, 2009,

from http://www.pavementpreservation.org/expert/12-15-08-ETG-ETF-Final.pdf

Shuler, S. and Epps-Martin, A., Manual for Emulsion-Based Chip Seals for Pavement Preservation (NCHRP 14-17), in progress

Takamura, Koichi, Pavement Preservation Using the SBR Latex Modified Asphalt Emulsion, BASF Corporation, Presented at the

International Latex Conference, Charlotte, NC, July, 2005

TSP2 (Transportation System Preservation Technical Services Program Transportation System Preservation), TSP Research Road

Map, 2008, Retrieved June 14, 2009 from http://www.tsp2.org/roadmap/

Turner, F. and Harnsberger, M., Using DSR and Rheological Modeling to Characterize Binders at Low Temp, Western Research

Institute, in progress

Walubita, L. F., Epps Martin, A. and Glover, C. J., A Surface Performance-Graded Specification for Surface Treatment Binders:

Development and Initial Evaluation, Texas Transportation Institute, FHWA Report No. FHWA/TX-05/0-1710-2, June 2005

267

Field Validation of Performance-Based Polymer-Modified Emulsion Residue Tests the FLH Study

Documents

Transcript of Field Validation of Performance-Based Polymer-Modified Emulsion Residue Tests the FLH Study