Click for next page ( 67


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 66
66 ATTACHMENT Manual for Emulsion-Based Chip Seals for Pavement Preservation: Research Report

OCR for page 66
67 CONTENTS 68 Chapter 1 Introduction 68 1.1 Background 68 1.2 Project Objectives and Scope 68 1.3 Organization of the Report 69 Chapter 2 Research Methodology 69 2.1 Introduction 69 2.2 Chip Adhesion to Emulsion and Residue 72 2.3 Time Required Before Brooming or Traffic 73 2.4 Emulsion Consistency in the Field 74 2.5 Pavement Texture Testing 76 2.6 Residue Recovery Methods and Properties 79 2.7 Estimating Chip Embedment Depth During Construction 81 Chapter 3 Results and Analysis 81 3.1 Sweep Test 81 3.2 Field Moisture Tests 81 3.3 Laboratory Sweep Test for Field Materials 85 3.4 Emulsion Consistency in the Field 87 3.5 Pavement Texture Measurement 87 3.6 Residue Recovery Methods and Properties 93 3.7 Estimating Embedment in the Field 95 Chapter 4 Practical Application of the Research 95 4.1 Modified Sweep Test and Critical Moisture Contents 95 4.2 Field Consistency Test 95 4.3 Pavement Texture 95 4.4 Residue Recovery and Desirable Properties 95 4.5 Measuring Aggregate Embedment in the Field 96 Chapter 5 Conclusions and Recommendations 96 5.1 Conclusions 96 5.2 Recommendations 98 References 100 Appendices A through J

OCR for page 66
68 CHAPTER 1 Introduction 1.1 Background chip seals placed on hot mix asphalt pavements. A signifi- cant body of knowledge existed about chip-seal design and Emulsion-based chip seals are the most commonly used construction before this research, much of which is con- type of chip seal in the United States for preserving asphalt tained in this manual. However, other practices in chip-seal pavements. The purpose of these preservation treatments is to technology have been subjective for many years and are seal fine cracks in the underlying pavement surface and pre- considered an art by some. Therefore, the research con- vent water intrusion into the base and subgrade. Chip seals are ducted in this study focused on elements of chip-seal tech- not expected to provide additional structural capacity to the nology that were subjective or not practiced in the United pavement. Benefits are obtained by reducing pavement dete- States. The research presented in this report includes a rec- rioration before significant distress is exhibited. A large body ommended manual that could replace the subjective or of research is available on chip-seal design practices (NCHRP qualitative judgments previously used during chip-seal Synthesis 342: Chip Seal Best Practices), and they were further design and construction with field and laboratory testing, investigated in this project. However, chip-seal design in the and thus can be used to improve the opportunity for success United States has not been developed significantly beyond when building chip seals. early work (McLeod 1960, 1969; Epps 1981). In spite of their apparent benefits, the use of chip seals for pavement preservation in the United States has been hampered 1.3 Organization of the Report by the lack of nationally accepted guidance on their design and construction and appropriate specifications and testing proce- This research report has five chapters. Chapter 1 is the dures for constituent materials. Therefore, research was needed introduction and describes the purpose of the research and to develop a manual that identifies factors that influence chip- the scope of the work. Chapter 2 describes the state of the seal design, construction, and performance and provides guide- practice of chip-seal design and construction. Chapter 3 lines that enable practitioners to improve the opportunity for describes the results and analysis of a series of laboratory and success when building these systems. field tests. Chapter 4 discusses the application of research find- ings. The final chapter presents the study conclusions and recommendations. Further elaborations on the research are 1.2 Project Objectives and Scope provided in Appendices A through J, which are not published This research was conducted to develop a manual describ- herein but are available on the TRB website at http://www. ing the best methods to use for designing and constructing trb.org/Main/Blurbs/164090.aspx.

OCR for page 66
69 CHAPTER 2 Research Methodology 2.1 Introduction based on how easily chips can be dislodged from the emul- sion. This experience is often gained by trial and error, some- Research conducted in this project focused on aspects of times leading to vehicle damage when residues that have not chip-seal technology that have been qualitative in the past or gained sufficient strength release chips under traffic loads were based on material properties that did not necessarily (Gransberg and James 2005; Shuler 1998). Several tests such relate to chip-seal performance. Quantitative methods were as Vialit (Vialit plate shock test), frosted marble (Howard developed to help replace past subjective practices and allow et al. 2009), and the sweep test (Cornet 1999; Barnat 2001; improved prediction of chip-seal behavior in the field. The ASTM D7000) attempt to quantify this adhesive behavior following issues were addressed in the research through lab- and identify when chip seals are ready to accept uncontrolled oratory and field experiments: traffic. However, these tests have shown high variability and have therefore not been widely adopted. One method uses a Chip adhesion to emulsion and residue, hand broom to sweep the chips, and the chip seal is judged Time required before sweeping and uncontrolled traffic, ready for traffic when the amount of chips dislodged during Emulsion consistency in the field, this procedure is less than 10%. This test is attractive since it Surface texture measurement, and uses actual construction materials and with practice could Residue recovery and properties. be a means to evaluate adhesion. The sweep test described by ASTM D7000, Standard Test Method for Sweep Test of 2.1.1 Chip Seal Definition Bituminous Surface Treatment Samples, appeared to be a reasonable approach to simulating the forces that dislodge Chip seals considered in this research are based on emulsi- aggregate chips from chip seals. This procedure is relatively fied asphalt binders and natural mineral aggregate chips. The effective at evaluating differences in adhesive abilities of dif- chip seal is constructed by spraying the asphalt emulsion onto ferent emulsions with a single aggregate. This test utilizes a the existing asphalt pavement, dropping the aggregate chips template for specific aggregate gradations to establish the into the asphalt emulsion, and embedding the chips in the emulsion application rate. While a single emulsion applica- emulsion using pneumatic-tired rollers. The purposes of the tion rate is suitable for relative comparison between emul- chip seal are to preserve existing asphalt pavement by sealing sions, when aggregate sizes differ, the embedment percentage the surface before cracking occurs or after minor cracks have changes, which affects chip retention. In addition, the test emerged and to provide additional surface friction. describes a procedure of "hand casting" the aggregates onto the emulsion prior to testing. Attempts to repeatedly place precise amounts of aggregate on test samples during this research 2.2 Chip Adhesion to Emulsion proved difficult to replicate. Therefore, the test apparatus was and Residue modified so that the exact amount of chips was placed on the The required adhesive and cohesive strength of the emul- test pad each time. To determine if the modified test proce- sion residue used as the binder in a chip seal directly influ- dure would be useful to evaluate the adhesive ability of dif- ences when the chip seal can be opened to traffic after ferent emulsions and different aggregate chips under varying construction. This strength is usually judged subjectively moisture conditions, a controlled laboratory experiment was during construction by experienced personnel who decide conducted.

OCR for page 66
70 2.2.1 Experiment Design niques (ANOVA). The experiment was repeated for each emul- sion to eliminate potential variability that could be associated Because of variability associated with the manner with with differences in emulsion behavior due to aging. which aggregate chips are prepared for testing according to ASTM D7000, a modification to the procedure was made to precisely control how chips are placed on the test pad prior 2.2.2 Materials to sweeping. To determine if the modified procedure was an A variety of emulsions were selected to represent the range improvement over the ASTM procedure, an experiment was available for construction. These included conventional conducted to measure the ability of the modified sweep test to and polymer modified anionic (RS-2 and RS-2P), high float discriminate between four independent variables believed to (HFRS-2P), and cationic types (CRS-2 and CRS-2P). Pro- affect early chip-seal performance. These variables were aggre- duction of these emulsions using a laboratory emulsion mill gate source, emulsion type, emulsion cure level, and aggregate in close proximity to the research laboratory was desirable chip moisture content. since emulsions have limited shelf life. These factors helped to reduce variability of the emulsion materials. Properties of 2.2.1.1 Independent Variables the emulsions are shown in Table 1. A variety of aggregates were used to determine if the mod- Independent variables in this experiment are the following: ified sweep test could discriminate between different miner- Aggregate source: basalt, granite, limestone, alluvial alogy, shape, and texture. These were a limestone (LSTN) Emulsion type: RS-2, RS-2P, CRS-2, CRS-2P, HFRS-2P aggregate from Colorado Springs, CO, granite (GRNT) from Emulsion cure level: 40%, 80% Pueblo, CO, basalt (BSLT) from Golden, CO, and an alluvial Aggregate chip source (ALLV) from Silverthorne, CO. The properties of moisture content: dry, saturated surface dry (SSD) these materials are presented in Table 2. A full-factorial, randomized experiment was designed for each emulsion according to the model shown below 2.2.3 Sweep Test Procedure (Anderson 1993): The test procedure is described in detail in Appendix B. Differences between the procedure conducted in the research Yikl = + Ai + Wk + M l + AWik + AM il + WM kl + AWM ikl + ikl and that described by ASTM D7000 include the following: Where 40% initial embedment of the aggregate chips, Yikl = chip loss, %; 40% and 80% emulsion moisture loss, and = mean loss, %; Consistent, uniform application of the aggregates to the Ai = effect of aggregate i on mean loss; Wk = effect of water removed k on mean loss; test pad. Ml = effect of aggregate moisture l on mean loss; In this procedure, asphalt emulsion is applied to a 15-pound AWik, etc. = effect of interactions on mean loss; and per square yard roofing felt substrate in a circle by means of ikl = random error for the ith aggregate, kth water a steel template with an 11-in. diameter cutout. Emulsified removed, and lth replicate. asphalt is screeded level with the template by means of a strike- This experiment design was chosen because results can be off rod as shown in Figure 1. Aggregate is then placed mechan- easily evaluated using conventional analysis of variance tech- ically using a dropping apparatus as shown in Figure 2. The Table 1. Emulsion properties. Emulsion Tests RS-2P RS-2 CRS-2 CRS-2P HFRS-2P Viscosity, SFS 122F 108 96 78 119 132 Storage Stability, 1 day, % 0.1 0.1 0.2 0.1 0.2 Sieve Test, % 0.0 0.0 0.0 0.0 0.0 Demulsibility, 35 ml 65 72 76 76 42 Residue, by evaporation, % 65.1 68.0 67.9 67.7 65.3 Residue Tests Penetration, 77F, 100g, 5s 115 112 125 121 115 Ductility, 77F, 5cm/min 100+ 100+ 55 65 60 Float, 140F, s na na na na 1290

OCR for page 66
71 Table 2. Aggregate properties. Passing, % Sieve No. Sieve Size (in.) (mm) LSTN GRNT BSLT ALLV 3/4 19.0 100 100 100 100 1/2 12.5 100 100 100 100 3/8 9.5 100 99 100 99 5/16 8.0 100 50 79 73 1/4 6.3 48 9 30 33 4 4.75 1 1 1 2 8 2.36 1 1 1 2 16 1.18 1 1 1 2 30 0.60 1 1 1 2 50 0.30 1 1 1 2 100 0.15 1 1 1 2 200 0.075 1 1 1 2 Bulk specific gravity 2.615 2.612 2.773 2.566 Loose unit weight, lbs/cf 78.3 84.0 92.2 86.1 Los Angeles Abrasion Loss, % 26.3 27.8 20.1 22.0 Flakiness Index 33.8 5.8 13.1 10.5 aggregate is then set in place, one stone thick, by means of a of a weighted brush which is spun by a planetary motion mixer compactor as shown in Figure 3. The specimen is then placed for 1 min as shown in Figure 4. The specimen is then removed in a 160F oven to allow the emulsified asphalt to cure to 40% from the machine and brushed by hand to remove all particles moisture loss or 80% moisture loss after which the specimen is that were mechanically dislodged from the specimen surface. removed from the oven. It is then cooled, and any loose parti- The mass loss is then determined, which is expressed as percent cles are removed. The specimen is then swept under the action loss of the original aggregate mass. Figure 1. Emulsion strike-off apparatus.

OCR for page 66
72 Figure 2. Dropping apparatus placing aggregate on test pad. 2.3 Time Required Before Brooming lead to delays and congestion. Also, if light brooming results in or Traffic damage to the chip seal, the chip seal is often left unbroomed until binder strength increases. However, allowing traffic on the Determining when the first brooming can be accomplished fresh, unswept chip seal can lead to flying chips and potential to remove excess chips or when to open a fresh chip seal to damage. traffic is one of the most subjective decisions that must be The modified sweep test, which measures the relative made. Releasing traffic too soon can lead to vehicle damage adhesive strength of emulsions and emulsion residues in the due to flying aggregate particles. Releasing traffic too late can laboratory, was used to evaluate materials from full-scale Figure 3. Compactor setting aggregates on test pad.

OCR for page 66
73 Figure 4. Modified sweep test mixer. chip seals. The objective of this experiment was to deter- residue on a scale of 1 (no strength) to 10 (ready for traffic), mine if the moisture content of the chip seal in the field judged by pulling three chips out of the fresh seal and qualita- affects the ability of the chip seal to withstand brooming and tively judging dislodgement potential. This qualitative evalua- traffic stresses. tion was conducted after rolling. Moisture remaining in the emulsion was determined by placing plywood pads covered 2.3.1 Full-Scale Field Tests with aluminum foil measuring 24 in. by 24 in. in front of the asphalt distributor prior to spraying with emulsion. The pads Three full-scale chip-seal projects were included in this were weighed before and after spraying and chipping, and the research. Test pavements were located on County Road 11 near loss in weight was determined periodically during the day until Frederick, Colorado, approximately 30 miles north of Denver, approximately 95% of the water had evaporated. Figure 5 shows Colorado; the Main Entrance Road in Arches National Park, the setup used to measure the tare weight of the apparatus prior Utah, approximately 15 miles north of Moab, Utah; and US- to spraying and chipping. 101 near Forks, Washington, on the western edge of Olympic The tared pad was placed in front of the asphalt distribu- National Park. tor and chip spreader before chip-seal operations began. After the emulsion and chips were applied to the pavement 2.3.2 Moisture Tests and tared pad, the pad was removed from the pavement and Moisture in a chip seal comes from two sources: the chips re-weighed. As moisture evaporated from the pad the weight and the asphalt emulsion. In addition, on some projects addi- was recorded and the strength of the emulsion residue was tional moisture may be present in the roadway. If the amount evaluated using the 1 to 10 scale. The resulting relationship of moisture in the chips and the emulsion is known at the time between emulsion strength and moisture loss was developed. the chip seal is constructed, the amount of moisture that evap- orates after emulsion and chip application can be measured. 2.4 Emulsion Consistency in the Field The objective of this part of the research was to measure the moisture loss in the three chip-seal projects and develop a The consistency of the emulsion is an important factor that relationship to chip adhesion. influences performance of the chip seal. An emulsion with The amount of moisture remaining in each chip seal was viscosity too low may not have the ability to hold chips in place measured and compared with the relative strength of the or could flow off the pavement. An emulsion with viscosity

OCR for page 66
74 Figure 5. Moisture test pads prior to spraying/chipping. too high could be difficult to spray evenly or may not have was found to be cumbersome to operate and time consuming the wetting ability needed to coat chips. Emulsions are often to clean and not appropriate for use in the field. tested at the point of manufacture and a certificate of compli- The first tests were conducted at the Arches site using the ance is issued by the manufacturer indicating compliance to Wagner funnel with a 4 mm orifice. However, the emulsion state, local, ASTM, or AASHTO specifications. However, required over 90 s to empty the funnel. This resulted in large because changes to physical properties of emulsions used for differences between test results because the emulsion viscosity chip seals can occur during transportation, a means of mea- increased as the temperature decreased, increasing the time to suring the consistency of the emulsion at the construction site empty the Wagner cup. Therefore, the orifice was drilled out is desirable. Some highway agencies have portable laborato- to increase the diameter until the cup emptied in approxi- ries capable of conducting viscosity tests in the field (Santi mately 60 s or less. This process was repeated for the Frederick, 2009). However, most agencies do not have laboratories or CO, and Forks field tests. trained personnel to conduct such tests. Therefore, a simple The test proved simple to conduct, low cost, and required method of verifying the ability of the emulsion to be used as a simple apparatus. Although this test would require more a chip-seal binder was identified in this research. development to be used for determining specification com- pliance in the field, the test will help a field inspector rapidly determine the suitability of an emulsion upon delivery to the 2.4.1 Full-Scale Field Tests construction site. Two simple methods for measuring the consistency of asphalt emulsions in the field were evaluated. One method based on 2.5 Pavement Texture Testing a procedure developed by Wyoming DOT (Morgenstern 2008) requires a Wagner Part #0153165 funnel, wind protec- Adjusting the emulsion spray rate to compensate for dif- tion, 16-ounce plastic cups, thermometer, and a stop watch. ferences in pavement surface texture is one of the most sub- The other method was a falling cylinder viscometer which jective adjustments made during chip-seal construction.

OCR for page 66
75 Except for the sand patch test used in South Africa and 2.5.1 Laboratory Texture Testing Australia/New Zealand (Austroads 2006, South African Roads Agency 2007), adjustments in the United States are One part of this research involved testing three slabs of made using judgment based on past experience. The objec- varying surface texture. These test slabs provided a range of tive of this experiment was to provide a more quantitative textures for evaluating three texture measurement techniques. method for evaluating pavement texture and adjustment of The slabs were fabricated to simulate three surfaces ranging emulsion application rate. from very rough, simulating a highly raveled and pocked sur- Macrotexture is the texture type that is relevant to chip seals. face, to very smooth, simulating a very flushed surface. Macrotexture is surface roughness that is caused by the mixture The slabs were cast over asphalt pavements using a very low properties of an asphalt concrete surface or by the finishing/ viscosity self-consolidating concrete. The self-consolidating texturing method of a portland cement concrete surface (Hall concrete was used to make the texture specimens because of et al. 2006). the concrete's ability to flow into the smallest voids in the Previous work has indicated that either the sand patch surface of the asphalt pavements. This created texture test test (ASTM E 965) or the circular texture meter (CT meter) specimens that mimicked the texture of the three pavement profile (ASTM E 2157) can be used to effectively evaluate surfaces. Texture of the three slabs was measured using sand pavement macrotexture (Abe et al. 2001, Hall et al. 2006, patch, CT Meter, and the Aggregate Imaging System (AIMS). Hanson and Prowell 2004). Both of these measurements are easily performed in the field, but traffic control is needed 2.5.1.1 Sand Patch Test during these measurements. The sand patch test has been used for texture measurement because it requires inexpen- The sand patch test (ASTM E 965) is a volumetric technique sive equipment that is easy to obtain, and it provides accept- for determining the average depth of pavement surface macro- able measurements (Austroads 2006, South African Roads texture. A known volume of small particles (either sieved sand Agency 2007). However, conducting the test is slow and or small glass beads) is poured onto the pavement surface and exposes personnel to traffic, and results are influenced by spread evenly into a circle using a spreading tool. Four diame- wind and moisture. ters of the circle are measured and an average profile depth is The CT meter evaluation of surface macrotexture can be calculated from the known material volume and the averaged made more quickly than sand patch testing and therefore circle area. This depth is reported as the mean texture depth exposes the technician to less traffic and accident risk. Also, (MTD). The method provides an average depth value and is the CT meter measurements do not depend upon operator insensitive to pavement microtexture characteristics. skill. Figure 6 shows the interior of the CT meter, which faces The CT meter test method (ASTM E 2157) is used to the pavement when taking measurements. measure and analyze pavement macrotexture profiles with Figure 6. CT meter.

OCR for page 66
76 a laser displacement sensor. The laser sensor is mounted on methods, structural functions, behavioral responses, distress an arm which follows a circular track of 284 mm (11.2 in.) types, and effects of environmental exposure. Therefore, the diameter. Depth profiles are measured at a sample spacing binder grading system, surface performance grading (SPG), of 0.87 mm, and the data are "segmented into eight 111.5 mm was first suggested to classify emulsion residues or hot-applied (4.39 in.) arcs of 128 samples each" (ASTM E 2157). A mean binders for use in chip seals (Epps et al. 2001, Barcena et al. profile depth (MPD) is calculated for each segment, and 2002). This grading system utilizes the same test methods as the an average MPD is then calculated for the entire circular PG system, but applies limits on test parameters that are con- profile. sistent with the mechanics of chip seals rather than hot mix asphalt. An emulsion residue specification requires a standardized 2.5.1.2 AIMS emulsion residue recovery method that produces a material The AIMS was created to quantitatively describe the charac- representative of the emulsion residue in situ. Currently, emul- teristics of aggregates (Masad 2005). The system consists of a sion residues are recovered by distillation (ASTM D 6997) that camera mounted above a table with several lighting arrange- exposes the material to high temperatures and may destroy or ments. Using AIMS, coarse aggregate is characterized by parti- change any polymer networks present in modified emulsion cle shape, angularity, and texture. Samples of coarse aggregate residues. are placed on the AIMS table under the camera and lighted This section describes the experiment used to compare from above, below, or both, and camera images are used to emulsion residue recovery methods, characterizes the emul- quantify the aggregate characteristics. Analyzing macrotexture sion residues by both the PG and SPG grading systems and of coarse aggregates can be compared to measuring macro- some additional tests, and recommends an emulsion residue texture of a pavement surface. recovery method and emulsion residue specification. Using AIMS, microtexture or macrotexture of coarse aggregate surfaces can be quantified using wavelet analysis of 2.6.1 The Surface Performance-Graded a grayscale digital photo. Camera focal length can be adjusted Specification depending on whether macrotexture or microtexture is of interest. Using AIMS, depth measurements were generated The tests used in the SPG grading system are conducted every 1 mm for four scanlines of 100 mm length each, 20 mm with standard PG testing equipment and the analyses are apart, and in two perpendicular directions, for a total of eight performance-based and consistent with chip-seal design, con- scanlines per test slab. The total of eight scanlines at 100 mm struction, behavior, in-service performance, and associated length each was chosen to be similar to the eight segments of distresses (Epps et al. 2001, Barcena et al. 2002). Field validation the CT meter profile. The two sets of four scanlines each of the initial SPG system was completed in Texas (Walubita were taken in perpendicular directions to account for direc- et al. 2005, Walubita et al. 2004) and resulted in the proposed tional differences in pavement texture. This arrangement three SPG grades shown in Table 3. could be used to estimate directional differences in texture, which is texture in the direction of traffic versus texture 2.6.2 Residue Recovery Experiment perpendicular to the direction of traffic. Profiles were gen- erated for the scanlines and analyzed in a procedure simi- The standard PG system (Asphalt Institute, SP-1) and the lar to the CT meter analysis (ASTM E 1845). A mean profile modified SPG system (Epps et al. 2001, Barcena et al. 2002, depth, MPD, was calculated from the AIMS data for each of Walubita et al. 2005, Walubita et al. 2004) were both used to the three test slabs. grade all base binders and corresponding recovered emulsion residues in this experiment. 2.6 Residue Recovery Methods and Properties 2.6.2.1 Materials The performance grading (PG) asphalt binder grading sys- Eight emulsions were included in this research, five of which, tem (Asphalt Institute, SP-1) is widely used as the specification identified as emulsions 1 through 5, were laboratory prepared. for grading and selecting asphalt binders. The PG specification The other three emulsions were obtained from the full-scale test was developed for use in hot mix asphalt concrete (HMAC) pavements in Utah Arches National Park; Frederick, Colorado, pavement layers. However, the PG system is not applicable to CR-11; and Forks, Washington, US-101. Table 4 lists the types classifying and choosing binders for use in pavement chip seals. of emulsions and, when known, the PG grades of the base Chip seals differ from full-depth HMAC layers in construction binders as reported by the supplier.

OCR for page 66
90 Table 8. Strain sweep test results. UNAGED AGED Gi* Emul- % at % at % at % at % at % at % at Recovery Gi* (Pa, (Pa, at sion 0.90Gi* 0.80Gi* 0.50Gi* 0.98Gi* 0.90Gi* 0.80Gi* 0.50Gi* at 1% ) 1% ) 1 base 241,120 21.23 34.74 n/a 987,120 4.95 10.88 12.67* n/a 1 stirred can 326,460 19.20 31.22 n/a 883,620 5.01 11.86 14.15* n/a 1 hot oven 337,500 19.79 32.67 60.06* 844,030 5.53 11.46 14.82* n/a 2 base 248,290 25.72 6.18 84.03* 1,448,300 3.93 7.31 8.62* n/a 2 stirred can 298,170 22.17 38.31 n/a 1,948,300 2.77 5.29 6.40* n/a 2 hot oven 318,660 21.32 36.98 63.37* 1,385,600 4.33 7.68 9.01* n/a 3 base 747,630 14.84 16.71 n/a 3,329,800 2.31 n/a n/a n/a 3 stirred can 825,740 13.26 15.13 n/a 2,811,300 3.62 n/a n/a n/a 3 hot oven 813,970 13.64 15.35 n/a 3,163,400 2.14 n/a n/a n/a 4 base 219,060 25.41 44.14 n/a 954,040 5.24 10.92 13.11* n/a 4 stirred can 289,860 20.77 34.51 n/a 905,480 5.58 11.27 13.82* n/a 4 hot oven 257,750 24.35 34.26 n/a 778,100 4.84 11.11 16.10* n/a 5 base 266,850 22.03 38.45 n/a 1,260,200 4.92 8.81* 9.91* n/a 5 stirred can 297,360 17.79 31.46 67.95* 765,620 5.18 10.76 16.35* n/a 5 hot oven 286,680 17.27 30.57 70.53* 801,740 3.96 10.38 15.54 n/a 6 UT stirred can 1,182,300 9.18 10.56* n/a 2,486,600 2.45 4.46 n/a n/a 6 UT hot oven 1,203,200 9.21* 10.37* n/a 2,886,400 3.33 3.84* n/a n/a 7 - CO stirred can 440,260 18.16 28.36 45.86* 1,235,400 3.36 8.41 10.11 n/a 7 - CO hot oven 444,800 17.92 28.20 45.42* 1,198,900 3.06 7.51 10.36 n/a * Max DSR stress was reached; n/a = test didn't run that far materials and characterize strain tolerance. For comparison, Researchers have conducted testing on binders during cur- the strain sweep data from the stirred can recovery residues for ing and have recommended the following criteria for deter- aged and unaged materials are shown in Figure 22. mining strain tolerance and failure of the emulsion residue Materials with high strain tolerance exhibit slow deteriora- during curing (Hanz et al. 2009): tion of G* with increasing strain level, indicating that the material maintains stiffness and holds together under repeated 10% reduction in G*, or 0.10Gi* characterizes strain toler- and increasing loads. Emulsions 1, 2, 4, and 5 in the unaged ance and indicates that the material is behaving nonlinearly state exhibited this behavior and were visibly more adhesive and is accumulating damage; and elastic when handled in the laboratory. After PAV aging, 50% reduction in G* or 0.50Gi* defines failure of the some materials exhibit less strain tolerance and develop a steep material. decrease in G* with increasing strain. Emulsions 2, 3, and Utah Arches are examples of this type of behavior. These materials Hanz et al. (2009) found that stiffer emulsion residues after were very stiff and broke off of the test plates in a brittle manner PAV aging are difficult to induce 50% Gi* and even 90% Gi* after the strain sweep testing was completed. in strain sweep testing. Most of the unaged and only a few of An asphalt binder must develop enough stiffness (G*) to be the PAV aged materials reached 80% Gi*, as shown in Table 8, able to carry vehicle loads before the chip-sealed pavement is and none reached 50% Gi*. It is possible that intermediate broomed or opened to traffic. The amount of moisture remain- reductions in Gi* could be used to characterize behavior of the ing in the chip seal has been shown to relate to binder strength fully cured residues when 50% or 90% Gi* cannot be attained. development. This moisture level could be correlated with G* Besides differing in the rate at which G* decreased with from strain sweep testing to determine a minimum G* for increasing strain, the materials differed in their original stiff- traffic bearing capacity. ness, Gi*, and the rate of change of Gi* between the unaged and

OCR for page 66
91 3,000,000 Emulsion 3 PAV Aged 2,500,000 Utah Arches PAV Aged Complex Modulus G* (Pa) 2,000,000 Emulsion 2 PAV Aged 1,500,000 1,000,000 500,000 Emulsions 1, 2, 4, and 5 Unaged 0 0 5 10 15 20 25 30 35 40 45 50 Strain (%) Figure 22. G* versus shear strain from stirred can recovery method. the PAV aged states as shown in Table 8. The stiffest material Based on the results of the strain sweep testing, Emulsions in the unaged state was Emulsion residue 6, a latex-modified, 1, 2, 4, 5, and 7 would be expected to resist raveling due to their rapid-setting emulsion. The stiffest material in the aged state high strain tolerances. Emulsions 3 and 6, which had very stiff was the Emulsion 3 residue, a rapid-setting unmodified emul- unaged residues, would be expected to resist flushing and also sion. G* increased the most from the unaged to the aged state might be able to be opened to traffic earlier. However, these for the Emulsion 3 residue. It was followed by the Emulsion 2 emulsions became more brittle with aging and could therefore residue, also a rapid-setting unmodified emulsion, and then exhibit raveling with age. by the Emulsion 6 residue. Residues for polymer modified A comparison between the emulsion residues used at the Emulsions 1, 4, 5, and 7 increased in G* and exhibited aged three field tests and those recommended by the SPG criteria behavior after the PAV aging, but not by as much as the residues are shown in Table 9. In all three cases the materials used were for Emulsions 2, 3, and 6. Also, for Emulsions 1, 4, and 5, the higher temperature grades than the SPG recommended criteria, base binder increased in G* considerably more than the recov- and in the case of Washington and Utah, lower temperature ered residue did, possibly indicating that either the emulsifi- grades, as well. cation process or the residue recovery process reduced the The proposed emulsion residue criteria shown in Table 10 susceptibility of these materials to the PAV aging process. are based on those originally proposed in previous research Table 9. Recommended SPG temperature ranges (C). Actual SPG Field Site Material Recommendation Used Forks, WA 5212 6715 Arches NP, Utah 6112 7915 Frederick, CO 5824 7618

OCR for page 66
Table 10. Proposed emulsion residue criteria. Performance Grade* SPG 61 SPG 64 SPG 70 12 18 24 30 12 18 24 30 12 18 24 30 Average 7-day Maximum Surface Pavement Design <61 <64 <70 Temperature, C Minimum Surface Pavement Design Temperature, >12 >18 >24 >30 >12 >18 >24 >30 >12 >18 >24 >30 C Original Binder Dynamic Shear, AASHTO TP5 * G , Minimum: 0.65 kPa 61 64 70 Sin Test temperature @10 rad/s, C Shear Strain Sweep % strain @ 0.8Gi*, Minimum: 25 Test Temperature @10 rad/s linear loading from 1 25 25 25 50% strain, 1 s delay time with measurement of 20 30 increments, C PAV Residue (AASHTO PP1) PAV Aging Temperature, C 100 100 100 Creep Stiffness, AASHTO TP1 S, Maximum: 500 MPa 12 18 24 30 12 18 24 30 12 18 24 30 m-value, Minimum: 0.240 Test Temperature @ 8 s, C Shear Strain Sweep Gi*, Maximum: 2.5 MPa 25 25 25 Test Temperature @10 rad/s linear loading at 1% strain and 1 s delay time, C *This table presents only three SPG grades as an example, but the grades are unlimited and can be extended in both directions of the temperature spectrum using 3oC and 6oC increments for the high temperature and low temperature grades, respectively.

OCR for page 66
93 shown in Table 3, including equivalent testing and perfor- At 20% embedment, the measured diameters are reasonably mance thresholds for parameters measured in the dynamic close to the theoretical diameters. However, at 82% embed- shear rheometer and bending beam rheometer for unaged, ment, the measured diameters are significantly less than the high-temperature and aged, low-temperature properties, calculated values. respectively. Additional testing and performance thresholds At 20% embedment, voids are deep, requiring many beads, were added based on strain sweep testing conducted as part and the procedure of spreading the beads from particle peak of this project and other research (Hanz et al. 2010) and to peak contributes less to error than it does at higher embed- the significantly different performance of Emulsion 3 and the ment percentages when the amount of beads between the Utah Arches emulsion. The thresholds provided for the DSR aggregate voids is relatively less. At 80% embedment, many and BBR parameters are based on validation with Texas field particles were fully covered by asphalt, making it impossible test sections adjusted for climates in Utah and Colorado. to spread the glass beads between these aggregates. Test results indicate this procedure to be useful when chip- 3.7 Estimating Embedment seal particle embedments are closer to 50% or not submerged in the Field and chips have low flakiness index. Embedment depth is usually determined during construc- 3.7.2 Constant Diameter Method tion by pulling several chips out of the binder and visually estimating the amount of embedment. This practice is prob- This method of estimating embedment depth used a mold lematic even if chips have a very low flakiness index because it of constant diameter in which glass beads were poured on is difficult to assess quantitatively with any precision. There- top of the aggregate chips and the volume measured. Using fore, two methods based on the sand patch test were developed weight-to-volume relationships for the materials, the volume to estimate embedment depth: the constant volume method of glass beads required to fill the mold was calculated as a and the constant diameter method. function of the aggregate embedment depth. A comparison between the calculated volume of glass beads required to fill 3.7.1 Constant Volume Method the mold and the actual volume measured for the limestone Glass beads were spread out in a circle on top of chips embed- and granite aggregates embedded to 20% and 82% is shown ded to 20% and 80% of the chip average least dimension. The in Figure 24. diameter of the circle was compared with the theoretical diam- At 20% embedment, measured values deviate 10% from eter that should result based on weight-to-volume characteris- the theoretical values. At 82% embedment, the deviation is tics of the materials presented in Section 2.7.1. Results of this less at 5% from theoretical. Deviations were similar for the experiment are shown in Figure 23. limestone and the granite at both levels of embedment. 100.0% 90.0% 80.0% Aggregate Embedment, % 70.0% 60.0% 50.0% Granite Limestone 40.0% Theoretical Theoretical Sand Patch Diameter Sand Patch Diameter at 82% Embedment at 82% Embedment 30.0% 20.0% Granite and Limestone Sand Patch Diameter at Actual 82% Embedment 10.0% 0.0% 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 Sand Patch Diameter (Dia), cm Figure 23. Comparison of calculated to measured embedment depth.

OCR for page 66
94 100.0% 90.0% 80.0% Aggregate Embedment, % 70.0% 60.0% 50.0% 40.0% 30.0% Limestone- Actual 20.0% Granite- Limestone- Actual Granite- 10.0% Calculated Calculated 0.0% 0 20 40 60 80 100 120 140 160 180 Volume of Sand below ALD ht, cm3 Figure 24. Embedment depth from constant diameter method.

OCR for page 66
95 CHAPTER 4 Practical Application of the Research Five new products were identified in this research to improve 4.3 Pavement Texture the design and construction of chip seals. This chapter describes these products and their application. A direct correlation between the sand patch test and the CT meter indicates that pavement texture measurements can be made with the CT meter and used as a substitute for the sand 4.1 Modified Sweep Test patch test results in the design process. This texture measure- and Critical Moisture Contents ment can then be used to adjust emulsion spray rates during This test provides a method to determine the timing for construction. Recommended adjustments are provided in the chip-seal brooming and opening to uncontrolled traffic. The manual. test determines the moisture content of the chip seal, which corresponds to adhesion needed to retain chips under traffic 4.4 Residue Recovery loads. The moisture content of the chip seal can be monitored and Desirable Properties during construction to determine when the desired moisture content is reached. This moisture content ranged from about The SCERR method is recommended for obtaining emul- 15% to 25% of the total chip-seal moisture. A description of sion residues for use in tests proposed for measuring physical the test method is provided in the appendix to the manual. properties. Proposed emulsion residue criteria are listed in Results of the modified sweep test indicated that an aggre- Table 10. The test method is provided in the appendix to the gate in the saturated surface dry condition provides better manual. adhesion than dry aggregates. This finding suggests that chip- seal aggregates be moistened prior to construction. 4.5 Measuring Aggregate Embedment in the Field 4.2 Field Consistency Test Two methods for measuring aggregate embedment in the A Wagner cup viscometer was used in this research to field have been developed. The constant volume method is a measure the consistency of emulsions. simple method, using a constant volume of glass beads spread By conducting the test at a variety of temperatures in the lab- on the pavement surface in a circle. By measuring the diameter oratory, a temperature versus flow time relationship can be of the circle, the embedment of the aggregate can be estimated. produced. Flow in the field can then be measured and com- However, this procedure becomes less accurate at embedment pared with laboratory results to determine actual viscosity at over 50%. An alternative procedure, the constant diameter the field temperature. The test method is described in the method, can be used to estimate embedment up to 80%. These appendix to the manual. test methods are provided in the appendix to the manual.

OCR for page 66
96 CHAPTER 5 Conclusions and Recommendations 5.1 Conclusions sure texture effectively, the CT meter and AIMS apparatus were found to be faster and provide very similar results. This report documents laboratory testing and field evalu- Extensive testing was done to evaluate new methods of ation of several new procedures suitable for use by highway emulsion residue recovery. The methods included hot oven agencies, consultants, contractors, and others involved in the (with nitrogen blanket), stirred can (with nitrogen purge), design and construction of chip seals. These new procedures and warm oven. Residues recovered using these methods were were developed to add objective measurement capability to tested using the Superpave PG test methods and chemical some of the largely subjective judgments made during chip- analysis to determine which recovery technique mimicked seal design and construction. the base asphalts closest and resulted in the least amount of A laboratory test that simulates the sweeping action of water remaining in the samples. These tests indicated that the rotary brooms during chip-seal construction was developed. SCERR method is rapid and provides a good simulation of This test simulates the shear forces applied by brooms and the base asphalt material properties. Also, recovered emul- uncontrolled traffic to fresh chip seals, and can be used to sion residues were shown to be different from their base predict the time required before brooms or uncontrolled traf- binders at high temperatures before PAV aging, but similar to fic can be allowed on the surface of the chip seal. The test indi- the base binders at cold temperatures after PAV aging. cated the following: The residues obtained from the emulsions used in the three field test sites were characterized using the Superpave PG The moisture content at which 90% of the aggregate chips binder tests. The result of this work is a performance-based are retained during the sweep test is the "critical moisture criterion for chip-seal residues. Initially, this SPG criterion content" corresponding to very high residue adhesive was calibrated using field test sections in Texas. However, strength at which traffic could be allowed onto the chip-seal results of this research indicated that adjustments to the orig- sections. inal criteria should be made to accommodate other climates. Significantly higher chip loss was measured for sweep test Therefore, characterization of residues should be done by specimens fabricated with dry aggregates than with satu- evaluating the complex shear modulus, G*, over a range of rated surface dry aggregates. shear strains to evaluate strain resistance. These results could No significant difference in chip loss was measured either be used to predict when emulsion-based chip seals will develop at 40% or 80% moisture loss between cationic and anionic resistance to raveling and enough stiffness to be opened to emulsions used with either calcareous or siliceous aggregates. traffic, both in newly constructed chip seals and after weather- ing and aging. The Wagner cup viscometer for measuring the consistency of paints was successfully adapted to measuring viscosity of 5.2 Recommendations emulsions. The test is inexpensive, field portable, repeatable, simple to operate, and can be correlated to laboratory tests. The findings of this project were based on a significant An adjustment to the emulsion spray quantity should be amount of field and laboratory measurement. However, made to account for pavement surface texture. This process additional studies would help improve upon these findings is often done subjectively or measured using the sand patch and the recommendations. Such studies may include the test in other countries. Although the sand patch test can mea- following:

OCR for page 66
97 Further sweep testing with other sources of aggregate tively inexpensive alternative to the sand patch test and the and emulsion to verify the validity of the test as a means of CT meter measurements. measuring chip adhesion. Monitoring the performance of the three test sites con- Evaluation of grayscale photography and image analysis structed as part of this research to provide additional for quantifying macrotexture of pavement surfaces (Pid- validation for setting thresholds in the proposed SPG spec- werbesky et al. 2009) to possibly provide a viable and rela- ification.

OCR for page 66
98 References Abe, H., T. Akinori, J. J. Henry, and J. Wambold. Measurement of Pave- Techniques, Paper #09-2877, Presented at the 88th Annual Meet- ment Macrotexture with Circular Texture Meter. Transportation ing of the Transportation Research Board, Washington, D.C., 2009. Research Record: Journal of the Transportation Research Board, No. Hanz, A., Z. A. Arega, and H. U. Bahia. Rheological Behavior of Emul- 1764, TRB, National Research Council, Washington, D.C., 2001. sion Residues Recovered Produced by an Evaporative Recovery Anderson, V. I. and R. A. McLean. Design of Experiments: A Realistic Method, Paper #10-3845 [CD-ROM], Presented at the 89th Approach, Marcel Dekker, ISBN: 0824761316, 1993. Annual Meeting of the Transportation Research Board, 2010. Asphalt Institute. Superpave Performance Graded Asphalt Binder Specs Howard, I. L., M. Hemsley, Jr., G. L. Baumgardner, W. S. Jordan, III. and Testing, SP-1, ISBN: 978193415416. Chip and Scrub Seal Binder Evaluation by Frosted Marble Aggre- Austroads. Austroads Technical Report AP-T68/06, Update of the Aus- gate Retention Test, Paper #09-1662, Presented at the 88th Annual troads Sprayed Seal Design Method. Alan Alderson, ARRB Group, Meeting of the Transportation Research Board, 2009. ISBN 1 921139 65 X, 2006. Kadrmas, A. Report on Comparison of Residue Recovery Methods and Barcena, R., A. E. Martin, and D. Hazlett. Performance Graded Binder Rheological Testing of Latex and Polymer Modified Asphalt Emul- Specification for Surface Treatments. Transportation Research Record: sions. Annual Meeting of the American Emulsion Manufacturers Journal of the Transportation Research Board, No. 1810, Transporta- Association, Report #4-AEMA ISAET 08, 2008. tion Research Board of the National Academies, Washington, D.C., Kucharek, A. Measuring the Curing Characteristics of Chip Sealing Emul- 2002. sions. Paper Presented at the Joint ARRA-ISSA-AEMA Meeting, Barnat, J., W. McCune, and V. Vopat. The Sweep Test: A Performance Asphalt Emulsion Manufacturers Association and McAsphalt, 2007. Test for Chip Seals. Asphalt Emulsion Manufacturers Association, Masad, E. Aggregate Imaging System (AIMS): Basics and Applications, San Diego, CA, February, 2001. Implementation Report FHWA/TX-05/5-1707-01-1, Texas Trans- Cornet, E. ESSO Abrasion Cohesion Test: A Description of the Cohe- portation Institute (TTI), College Station, TX, 2005. sive Breaking Emulsions for Chip Seal. International Symposium McLeod, N. Basic Principles for the Design and Construction of Seal on Asphalt Emulsion Technology: Manufacturing, Application and Coats and Surface Treatments. Proceedings of the Association of Performance, 346355, 1999. Asphalt Paving Technologist, Vol. 29, 1960, 1150. Epps, A. L., C. J. Glover, and R. Barcena. A Performance-Graded Binder McLeod, N. W. Seal Coat Design. Proceedings of the Association of Specification for Surface Treatments, Report No. FHWA/TX- Asphalt Paving Technologists, Vol. 38, February 1969. 02/1710-1, Texas Department of Transportation and U.S. Depart- Morgenstern, B. Wyoming 538.0 (under review), Field Emulsion Viscos- ment of Transportation, Federal Highway Administration (FHWA), ity Test, 2008. Washington, D.C., 2001. Pidwerbesky, B., J. Waters, D. Gransberg, and R. Stemprok. Road Sur- Epps, J. A., B. M. Gallaway, and C. H. Hughes. Field Manual on Design face Texture Measurement Using Digital Image Processing and and Construction of Seal Coats. Texas Transportation Institute, Information Theory, Land Transport New Zealand Research Report Research Report 214-25, July 1981. 290, New Zealand, 2009. Gransberg, D. and D. M. B. James. NCHRP Synthesis 342: Chip Seal Best Prapaitrakul, N., R. Han, X. Jin, C. J. Glover, D. Hoyt, and A. E. Martin. Practices, Transportation Research Board of the National Academies, The Effect of Three Asphalt Emulsion Recovery Methods on Recov- Washington, D.C., 2005. ered Binder Properties. Petersen Asphalt Research Conference, 2009. Hall, J. W., K. L. Smith, L. Titus-Glover, J. C. Wambold, T. J. Yager, and Santi, M. Case Study of a State Emulsion QA Program. Presentation to Z. Rado. NCHRP Web-Only Document 108: Guide for Pavement the Rocky Mountain Pavement Preservation Partnership, Airport Friction, Final Report, Project 1-43, Transportation Research Board Hilton, Salt Lake City, UT, October 30, 2009. of the National Academies, Washington, D.C., 2006. http://www.trb. Shuler, S. Design and Construction of Chip Seals for High Traffic Vol- org/Main/Blurbs/161756.aspx. ume. Flexible Pavement Rehabilitation and Maintenance. ASTM Spe- Hanson, D. I. and B. D. Prowell, Evaluation of Circular Texture Meter cial Technical Publication (STP) Number 1348. American Society for Measuring Surface Texture of Pavements. National Center for for Testing and Materials. West Conshohocken, PA, 1998. Asphalt Technology (NCAT), Auburn, AL, 2004. South African National Roads Agency. Technical Recommendations Hanz, A., Z. A. Arega, and H. U. Bahia. Rheological Evaluation of for Highways, TRH3 2007, Design and Construction of Surfacing Emulsion Residues Recovered Using Newly Proposed Evaporative Seals, May 2007.

OCR for page 66
99 Walubita, L. F., A. E. Martin, D. Hazlett, and R. Barcena. Initial Validation Texas Department of Transportation and U.S. Department of Trans- of a New Surface Performance-Graded Binder Specification. Trans- portation, FHWA, 2005. portation Research Record: Journal of the Transportation Research Washington State Department of Transportation. Asphalt Seal. Technol- Board, No. 1875, Transportation Research Board of the National ogy Transfer, March 2003, 130. Academies, Washington, D.C., 2004. Woo, J. W., E. Ofori-Abebresse, A. Chowdhury, J. Hilbrich, Z. Kraus, Walubita, L. F., A. E. Martin, and C. J. Glover. A Surface Performance- A. E. Martin, and C. Glover. Polymer Modified Asphalt Durability Graded (SPG) Specification for Surface Treatment Binders: Devel- in Pavements, Report #FHWA/TX-07/0-4688-1, Texas Transporta- opment and Initial Validation, Report No. FHWA/TX-05/0-1710-2. tion Institute, 2006.

OCR for page 66
100 APPENDIX A THROUGH APPENDIX J Appendices A through J of the contractor's final report for NCHRP Project 14-17 are available on the TRB website at http://www.trb.org/Main/Blurbs/164090.aspx.