Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
35 Selection of Performance-Based Test Methods Laboratory test data were analyzed to identify the tests ap- propriate for identifying RAP and RCP materials intended for use as unbound pavement layers, singularly or in combi- nation with other materials. An adequate test method is expected to measure parame- ters that influence performance and should be capable of dif- ferentiating between sources, types, and blends of recycled aggregate. Toughness and Abrasion Resistance Recycled materials and virgin aggregate toughness and abrasion resistance characteristics were determined using the Micro-Deval test (AASHTO TP 58-00); percent loss for the different materials are shown in Figure 5.1. RCP and RAP in as-received condition exhibited more material loss than vir- gin aggregate material or 50-percent blends of recycled ma- terials with virgin aggregate. For the highest material loss, it appears that the test results were affected by the amount of fines produced during testing. The excess fines were not caused by aggregate degradation; therefore, test results are not appro- priate. Aggregates that exhibit material loss of 17 percent or more have shown fair or poor field performance. (1, 13) The following two statistical hypotheses were used to determine whether the test method differentiated between RAP, RCP, and virgin aggregates at a statistically significant level (5 percent): Null hypothesis, HO: MeanRAP = MeanVirgin Alternate hypothesis, HA: MeanRAP MeanVirgin Hypothesis testing results on Micro-Deval test data are shown in Table 5.1. These results indicate different material loss for different recycled materials, thus indicating that the test differentiates between different materials. Durability Recycled materials and virgin aggregate durability charac- teristics were determined using the Canadian Freeze-Thaw test (MTO LS-614); percent material loss for different aggre- gates are shown in Figure 5.2. These data show that RCP sam- ples and 50-percent blends of RCP with virgin aggregate had the largest material loss. It appears that the results are affected by the production of excess fines from recycled materials dur- ing testing, resulting from the disintegration of the cement paste on the aggregate particles. Test results show the differ- ence between different recycled materials and their blends with virgin aggregate. Results of statistical evaluation of test data are shown in Table 5.2. Frost Susceptibility Frost susceptibility of the recycled and virgin aggregates was determined using the tube suction test. Aggregates were con- sidered acceptable, marginal, or unacceptable if the dielectric constant was less than 10, between 10 and 16, and greater than 16, respectively (12). The order of recycled and virgin aggregate materials with respect to dielectric constant value is shown in Table 5.3. All RAP samples met the dielectric constant criterion; some RCP samples failed this criterion. The statistical test results are shown in Table 5.4. The tube suction test produced different results for the different material types. Static Triaxial Test Figure 5.3 shows the average maximum deviator stress at 15 psi (103.4 kPa) confining pressure sorted by aggregate type, C H A P T E R 5 Analysis of Test Data
36 tested gradation, and material tested. The test yielded differ- ent results for different conditions, although similar results were obtained for materials containing limestone and gravel. Statistical analysis of test data also showed similar trends. Test results were different for RAP and RCP samples with a p-value of 0.027. However, other comparisons indicated mixed results, probably due to limited test data. Figures 5.4 and 5.5 show the maximum deviator stress at 15 psi (103.4 kPa) confining pressure for different percentages of RAP and RCP recycled materials, respectively. For RAP samples, the results of the static triaxial strength on as-received materials indicated a decrease in deviator stress with increasing recycled material content but not for the re-blended OGDL gradation. RCP, RCP-GV, and RCP-LS materials exhibited lower fail- ure strengths than those for the virgin aggregate material or the 50-percent blend with virgin aggregate. However, the fail- ure strength of RCP material with granite increased with in- creased recycled material content. Results of the static triaxial test at 15 psi (103.4 kPa) con- fining pressure indicated differences between RAP and RCP and between recycled materials containing different aggre- gate types. Repeated Load Triaxial Test Figure 5.6 shows the âmeanâ maximum deviator stress at 15 psi (103.4 kPa) confining pressure sorted by aggregate 0 8 6 4 2 10 12 14 16 18 20 50/50 RCP-LS-IL Virgin DGBL Blend#1 RAP-GV-LA 50/50 RCP-GV-LA Virgin OGDL Blend 50/50 RCP-GR-SC RCP-GR-SC 50/50 RAP-GR-CO 50/50 RCP-GV-LA 50/50 RAP-LS-MS RAP-LS-MS RAP-GR-CO RCP-LS-IL RCP-GV-LA Percent Material Loss Figure 5.1. Results of Micro-Deval tests. Null Hypothesis, Ho p - Value Remarks MeanRAP = MeanVirgin 0.030 (< 5%) Test method differentiated between RAP and virgin aggregate MeanRCP = MeanVirgin 0.060 (> 5%) Test method did not differentiate between RCP and virgin aggregate. MeanRecycled = MeanVirgin 0.006 (< 5%) Test method differentiated between recycled and virgin aggregate MeanRAP100% = MeanRAP50% 0.655 (> 5%) Test method did not differentiate between 100 percent and 50 percent RAP samples. MeanRCP100% = MeanRCP50% 0.104 (> 5%) Test method did not differentiate between 100 percent and 50 percent RCP samples. MeanRAP = MeanRCP 0.537 (> 5%) Test method did not differentiate between RAP and RCP samples. MeanRAP50% = MeanRCP50% 0.068 (> 5%) Test method did not differentiate between 50 percent RAP and 50 percent RCP samples. MeanRAP100% = MeanRCP100% 0.736 (> 5%) Test method did not differentiate between 100 percent RAP and 100 percent RCP samples. Table 5.1. Statistical assessment of Micro-Deval test data at 5 percent test significance.
37 type, tested gradation, and material for tests conducted on un- saturated and saturated samples (OGDL samples were tested under unsaturated conditions only). Test results indicate dif- ferences between different conditions, although the differences between saturated and unsaturated tests were relatively small. Figure 5.7 shows the results of repeated load triaxial tests conducted at 15 psi (103.4 kPa) confining pressure. The RAP samples with granite exhibited relatively higher deviator stress in as-received gradation compared to when blended with 50-percent virgin aggregate. The capacity of RAP samples with limestone and granite aggregate to resist permanent deformation decreased with an increase in the recycled ma- terial content. RAP with granite aggregate exhibited higher deviator stress by itself (100-percent recycled material) com- pared with RAP blended with virgin aggregate material (50- percent blend). Tests conducted on RAP-GR-CO material in both saturated and unsaturated conditions had similar results. Figures 5.8 and 5.9 show the results of repeated load tri- axial tests on RCP samples. In general, there was decrease in de- viator stress when RCP samples were tested in the saturated condition. However, the 100-percent RCP samples had rela- tively higher deviator stress than the 50-percent blend of RCP with virgin aggregate. For most samples tested in the unsatu- rated condition, there was a decrease in deviator stress with an increase in recycled material content. When tested under saturated conditions, 100-percent RCP samples exhibited higher deviator stress relative to RCP samples composed of 50-percent virgin aggregate. The mean deviator stress of the 0 5 10 15 20 25 30 Virgin DGBL Blend#1 RAP-LS-MS 5050 RAP-LS-MS Virgin OGDL Blend 50/50 RCP-GV-LA RAP-GV-LA 50/50 RAP-GR-CO RAP-GR-CO 50/50 RCP-LS-IL 50/50 RCP-GV-LA 50/50 RCP-GR-SC RCP-LS-IL RCP-GV-LA RCP-GR-SC Percent Material Loss Figure 5.2. Canadian Freeze-Thaw test results. Null Hypothesis, Ho p - Value Remarks MeanRAP = MeanVirgin 0.165 (> 5%) Test method did not differentiate between RAP and virgin aggregate. MeanRCP = MeanVirgin 0.004 (< 5%) Test method differentiated between RCP and virgin aggregate. MeanRecycled = MeanVirgin 0.003 (< 5%) Test method differentiated between recycled and virgin aggregates. MeanRAP100% = MeanRAP50% 0.771 (> 5%) Test method did not differentiate between 100-percent and 50-percent RAP samples. MeanRCP100% = MeanRCP50% 0.009 (< 5%) Test method differentiated between 100- percent and 50-percent RCP. MeanRAP = MeanRCP 0.011 (< 5%) Test method differentiated between RAP and RCP samples. MeanRAP50% = MeanRCP50% 0.142 (> 5%) Test method did not differentiate between 50-percent RAP and 50-percent RCP samples. MeanRAP100% = MeanRCP100% 0.012 (< 5%) Test method differentiated between 100- percent RAP and 100-percent RCP samples. Table 5.2. Statistics for Canadian Freeze-Thaw data at 5-percent test significance.
38 Material Tested Dielectric Constant Value Rating RAP-GV-LA 2.0 Acceptable RAP-LS-MS 2.1 Acceptable 50/50 RAP-LS-MS 3.2 Acceptable RAP-GR-CO 3.3 Acceptable RAP-GR-CO 100%OGDL re-blend 3.4 Acceptable 50/50 RAP-GV-LA 3.6 Acceptable 50/50 RAP-GR-CO 3.7 Acceptable RAP-GR-CO 50/50 OGDL re-blend 3.9 Acceptable Virgin OGDL Blend 8.0 Acceptable RCP-GR-SC 50/50 OGDL re-blend 9.5 Acceptable Virgin DGBL Blend#1 9.6 Acceptable 50/50 RCP-GR-SC 10.0 Acceptable RCP-GR-SC 100%OGDL re-blend 10.4 Marginal Virgin DGBL Blend#2 10.6 Marginal RCP-GR-SC 12.1 Marginal 50/50 RCP-GV-LA 13.5 Marginal RCP-GV-LA 14.3 Marginal RCP-LS-IL 16.3 50/50 RCP-LS-IL 21.6 Unacceptable Unacceptable Null Hypothesis, Ho p - Value Remarks MeanRAP = MeanVirgin 0.009 (< 5%) Test method differentiated between RAP and virgin aggregate. MeanRCP = MeanVirgin 0.033 (< 5%) The test method differentiated between RCP and virgin aggregate. MeanRAP = MeanRCP 0.000 (< 5%) The test method differentiated between RCP and RAP samples. Table 5.3. Tube suction test results. Table 5.4. Statistical test results on Tube-Suction test data at 5-percent test significance. 104 98 87 86 107 96 89 109 104 76 0 20 40 60 80 100 120 Vi rg in G ra ni te G ra ve l Li m es to ne D G BL #1 D G BL #2 O G D L R C P Vi rg in R AP Aggregate Type Gradation Tested Material M ax im um D ev ia to r S tr es s, p si Figure 5.3. Static triaxial test results at 15 psi (103.4 kPa) confining pressure.
39 0 20 40 60 80 100 120 140 0 50 100 Recycled material content, percent M ax im um d ev ia to r s tre ss , p si RAP-GR-CO (DGBL#2) RAP-GR-CO Reblend (OGDL) RAP-GV-LA (OGDL) RAP-LS-MS (OGDL) Figure 5.4. Maximum deviator stress versus recycled material content (RAP). 60 80 100 120 140 0 50 100 Recycled material content, percent M ax im u m d ev ia to r st re ss , p si RCP-GR-SC (DGBL#2) RCP-GR-SC Reblend (OGDL) RCP-GV-LA (DGBL#1) RCP-LS-IL (DGBL#1) Figure 5.5. Maximum deviator stress versus recycled material content (RCP). unsaturated test appeared to be unaffected by the amount of RCP in the test sample (165 psi [1138 kPa]) for 50-percent RCP blends with virgin aggregate compared to 163 psi (1124 kPa) for 100-percent RCP. Statistical significance test results, shown in Table 5.5, in- dicated similar trends. The test method correctly differenti- ated between different materials. Figures 5.10, 5.11, and 5.12 show the order of maximum deviator stress of recycled materials tested in the unsaturated condition in the repeated load triaxial strength at 1-, 2-, and 3-percent strain, respectively. Figure 5.13 shows the shear strengths at 1-, 2-, and 3-percent strain for tests conducted in the saturated condition. Overall, RCP and virgin aggregate ex- hibited higher maximum deviator stress than RAP material. When tested in the saturated condition, the RCP and virgin aggregate showed higher maximum deviator stress than RAP material. The materials selected for laboratory tests were expected to provide a range of expected performance as indicated by shear strength. At 1-percent strain, 100-percent RAP material had the lowest strength, followed by 50 percent blends of RAP with virgin aggregate. The 100 percent RCP and virgin aggregate samples had the highest strengths; the 50 percent RCP blends with virgin aggregate had the second highest shear strengths. Shear strengths estimated at 3 percent strain provided somewhat different order.
40 The slopes of the deviator stress versus axial strain curve obtained during the load and unload cycles in the repeated load triaxial testing could provide an indication of the per- formance potential of recycled materials. Figure 5.14 shows typical load/unload curves for RCP and RAP materials. Dur- ing static triaxial testing, these materials failed at about 4- to 5-percent strain. However, in repeated load triaxial testing, the sample sustained a higher load due to aggregate interlock and resistance characteristics. Good quality materials indi- cate a large slope (change in deviator stress per unit perma- nent strain) or low curvature at test initiation. The order of tested materials based on initial slope is shown in Figure 5.15. Resilient Modulus Test The resilient modulus (or stiffness) was estimated at dif- ferent bulk stresses from data obtained during repeated load triaxial tests; results at the bulk stress of 100 psi (689.5 kPa) are shown in Figure 5.16. RCP-GR-SC was the least stiff ma- terial. Order-based saturated test results are shown in Figure 5.17. Generally, virgin aggregate and 50-percent blends of re- cycled materials with virgin aggregate exhibited higher stiff- ness than 100-percent recycled materials. Statistical analysis of resilient modulus data, shown in Table 5.6, indicate that test data reveal differences between different materials. 17 6 16 1 15 9 16 6 16 7 17 8 15 4 16 3 17 6 15 8 15 7 16 6 13 9 17 7 16 2 15 7 15 8 15 7 1 73 0 20 40 60 80 100 120 140 160 180 200 Vi rg in Ag g. G ra ni te G ra ve l Li m es to ne D G BL #1 D G BL #2 O G DL R CP Vi rg in Ag g. R AP Aggregate Type Gradation Tested Material M ax im um D ev ia to r S tre ss , p si Unsaturated Tests Saturated Tests Figure 5.6. Repeated load triaxial test results at 15 psi confining pressure. 100 110 120 130 140 150 160 170 180 190 RAP-GR-CO (DGBL#2) RAP-GR-CO (DGBL#2-Sat) RAP-GR-CO (OGDL) RAP-GV-LA (OGDL) RAP-LS-MS (OGDL) M ax im um D ev ia to r S tre ss , p si Virgin aggregate 50/50 Blend 100% Recycled Figure 5.7. Repeated load triaxial test results for RAP samples.
41 120 130 140 150 160 170 180 190 0 50 100 Recycled material content, percent M ax im um D ev ia to r S tre ss , p si RCP-GR-SC (DGBL) RCP-IL-LS (DGBL) RCP-SC-GR (OGDL) RCP-GV-LA (DGBL-Sat) Figure 5.8. Repeated load triaxial test results for RCP samples. 120 130 140 150 160 170 180 190 0 50 100 Recycled material content, percent M ax im um D ev ia to r S tre ss , p si RCP-IL-LS (DGBL-Sat) RCP-GR-SC (DGBL-Sat) RCP-GV-LA (DGBL) Figure 5.9. Repeated load triaxial test results for RCP-DGBL samples. Test Condition Null Hypothesis, Ho p - Value Remarks MeanRCP = MeanVirgin 0.007 (< 5%) Test method differentiated between RCP and virgin aggregate. MeanRAP = MeanVirgin 0.000 (< 5%) Test method differentiated between RAP and virgin aggregate. At OMC MeanRAP = MeanRCP 0.024 (< 5%) Test method differentiated between RAP and RCP samples. MeanRCP = MeanVirgin 0.653 (> 5%) Test method did not differentiate between RCP and virgin aggregate. MeanRAP = MeanVirgin 0.321 (> 5%) Test method did not differentiate between RAP and virgin aggregate. Saturated MeanRAP = MeanRCP 0.023 (< 5%) Test method differentiated between RAP and RCP samples. MeanRAP50% = MeanVirgin .019 (< 5%) Test method differentiated between 50 percent RAP and virgin aggregate. OMC and saturated MeanRCP50% = MeanVirgin 0.048 (< 5%) Test method differentiated between 50 percent RCP and virgin aggregate. Table 5.5. Statistics for repeated load triaxial test data at 5-percent test significance.
42 0 20 40 60 80 100 120 140 RCP-GV-LA Virgin OGDL Virgin #2 Virgin #1 RCP-GV-LA (50) RCP-GR-SC RCP-LS-IL RCP-LS-IL (50) RCP-GR-SC (OGDL) RCP-GR-SC (OGDL) 50 RCP-GR-SC (50) RAP-GV-LA (50) RAP-CO-GR (50) RAP-LS-MS (50) RAP-CO-GR (OGDL) 50 RAP-GV-LA RAP-CO-GR RAP-CO-GR (OGDL) RAP-LS-MS Maximum Deviator Stress at 1% strain, psi Figure 5.10. Shear strength at 1-percent strain in repeated load triaxial test. 0 20 40 60 80 100 120 140 160 RCP-GV-LA Virgin #1 Virgin OGDL RCP-LS-IL Virgin #2 RCP-GV-LA (50) RCP-GR-SC RCP-LS-IL (50) RCP-GR-SC (50) RCP-GR-SC (OGDL) 50 RCP-GR-SC (OGDL) RAP-GV-LA (50) RAP-CO-GR (50) RAP-LS-MS (50) RAP-GV-LA RAP-CO-GR (OGDL) 50 RAP-CO-GR RAP-CO-GR (OGDL) RAP-LS-MS Maximum Deviator Stress at 2% strain, psi Figure 5.11. Shear strength at 2-percent strain in repeated load triaxial test.
43 0 20 40 60 80 100 120 140 160 180 Virgin #1 RCP-GV-LA RCP-LS-IL Virgin OGDL Virgin #2 RCP-GV-LA (50) RCP-GR-SC RCP-LS-IL (50) RCP-GR-SC (50) RCP-GR-SC (OGDL) 50 RCP-GR-SC (OGDL) RAP-GV-LA (50) RAP-CO-GR (50) RAP-LS-MS (50) RAP-GV-LA RAP-CO-GR RAP-CO-GR (OGDL) 50 RAP-CO-GR (OGDL) RAP-LS-MS Maximum Deviator Stress at 3% strain, psi Figure 5.12. Shear strength at 3-percent strain in repeated load triaxial test. 0 20 40 60 80 100 120 140 RCP-LS-IL Virgin #1 RCP-GR-SC Virgin #2 RCP-GV-LA RCP-GV-LA (50) RCP-LS-IL (50) RCP-GR-SC (50) RAP-CO-GR (50) RAP-CO-GR Maximum Deviator Stress at 1, 2, 3 % strain, psi 1% strain 2% strain 3% strain Figure 5.13. Shear strength materials tested saturated in repeated load triaxial test.
44 0 20 40 60 80 100 120 140 160 180 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 Axial Strain (%) D ev ia to r S tre ss (p si) RCP-GV-LA-Strain First Load Increment RCP-GV-LA-Strain Last Load Increment RAP-LS-MS-Strain First Load Increment RAP-LS-MS-Strain Last Load Increment Figure 5.14. Comparison of load/unload curves for typical RCP and RAP materials. 0 20 40 60 80 100 120 140 160 180 RAP-CO-GR W RAP-MS-LS RAP-CO-GR-OGDL RAP-CO-GR-50-50 RAP-CO-GR RAP-CO-GR-OGDL-50-50 RAP-MS-LS-50-50 RAP-LA-GV RAP-CO-GR-50-50 Virgin DGBL W RCP-SC-GR-50-51 RCP-SC-GR-50-50 Virgin DGBL #3 RCP-LA-GV-50-51 Virgin DGBL #2 RAP-LA-GV-50-50 RCP-IL-LS RCP-IL-LS-50-50 RCP-SC-GR-OGDL-50-50 RCP-IL-LS-50-50 Virgin DGBL RCP-LA-GV-50-50 RCP-IL-LS W RCP-LA-GV W RCP-GR-SC-OGDL RCP-SC-GR-OGDL-50-50 RCP-SC-GR-50-50 W Virgin OGDL RCP-LA-GV Initial Slope (percent change in deviator stress per unit permanent strain) Figure 5.15. Order of materials based on initial load-strain slope.
45 0 10 20 30 40 50 60 70 Virgin DGBL Blend#1 Virgin DGBL Blend#2 Virgin DGBL Blend RCP-GV-LA 50/50 RAP-GR-CO 50/50 RAP-LS-MS RAP-LS-MS RAP-GR-CO OGDL reblend 50/50 RAP-GV-LA 50/50 RAP-GR-CO OGDL reblend RAP-GR-CO RAP-GV-LA 50/50 RCP-GR-SC OGDL reblend 50/50 RCP-GR-SC 50/50 RCP-LS-IL 50/50 RCP-GV-LA RCP-LS-IL 50/50 RCP-GR-SC OGDL reblend RCP-GR-SC Stiffness, Resilient Modulus (ksi) Figure 5.16. Stiffness estimate using repeated load triaxial data at 100 psi bulk stress. 0 10 20 30 40 50 60 70 Virgin DGBL Blend#2 Virgin DGBL Blend#1 50/50 RAP-GR-CO RAP-GR-CO RCP-LS-IL 50/50 RCP-LS-IL 50/50 RCP-GR-SC 50/50 RCP-GV-LA RCP-GV-LA RCP-GR-SC Stiffness, Resilient Modulus (ksi) Figure 5.17. Estimated stiffness at 100 psi bulk stress (sat. repeated load triaxial test). Test Method Selection Summary The performance potential of an unbound pavement layer depends on its dry and wet shear strength, resistance to freeze-thaw (durability), toughness, and frost susceptibility. These properties were evaluated using selected tests in a lab- oratory investigation. Also, screening tests were conducted to characterize recycled materials. Based on results of the labo- ratory investigation, the following tests were found to relate to performance: ⢠Screening tests for sieve analysis and the moisture-density relationship,
46 ⢠The Micro-Deval test for toughness, ⢠Resilient modulus for stiffness, ⢠Static triaxial and repeated load at OMC and saturated for shear strength, and ⢠Tube suction test for frost susceptibility. Selection of Recycled Materials for Intended Use Recycled materials can be selected for use in a particular traf- fic and climatic condition. Rangarajy et al. (13) developed an approach for evaluating aggregates using selected test param- eters, performance ratings, and traffic and climatic categories. In this approach, tests are conducted in sequence and results are compared to suggested performance levels for specific traf- fic and climatic ranges. Three traffic levels are proposed: ⢠Low traffic (<100,000 ESALs/year), ⢠Medium traffic (100,000â1,000,000 ESALs/year), and ⢠High traffic (>1,000,000 ESALs/year). The climatic conditions of moisture (high/low) and tem- perature (freezing/not freezing) are based on the AASHTO definitions (14). Table 5.7 shows the significance levels of traffic, moisture, and climate combinations on a scale of 1 to 4, where 4 is most significant and 1 least significant on aggre- gate performance potential. Recycled materials could also be selected for use in a partic- ular pavement structure (e.g., doweled PCC, undoweled PCC, or HMA at various traffic levels and climates) for a particular base/subbase application (e.g., strength layer or construction or drainage layer). Different properties would be required of recycled materials for each unique situation. This level of detail has not been considered in this research. Test Condition Null Hypothesis, Ho p - Value Remarks MeanRCP = MeanRAP * 0.066 (> 5%) Test method did not differentiate between RCP and RAP. MeanRAP = MeanVirgin 0.003 (< 5%) Test method differentiated between RAP and virgin aggregate. MeanRCP = MeanVirgin 0.000 (< 5%) Test method differentiated between RCP and virgin aggregate. MeanRCP50 = MeanRAP50 0.028 (< 5%) Test method differentiated between RAP and RCP samples containing 50-percent virgin aggregate. At OMC MeanRCP100 = MeanRAP100 0.747 (> 5%) Test method did not differentiate between 100 percent RCP and 100-percent RAP. MeanRCP = MeanRAP * 0.462 (> 5%) Test method did not differentiate between RCP and RAP. MeanRAP = MeanVirgin 0.077 (> 5%) Test method did not differentiate between RAP and virgin aggregate. Saturated MeanRCP = MeanVirgin 0.020 (< 5%) Test method differentiated between RCP and virgin aggregate. MeanRCP = MeanRAP * 0.038 (< 5%) Test method differentiated between RAP and RCP. MeanRAP = MeanVirgin 0.000 (< 5%) Test method differentiated between RAP and virgin aggregate. MeanRCP = MeanVirgin 0.000 (< 5%) Test method differentiated between RCP and virgin aggregate. MeanRCP50 = MeanRAP50 0.009 (< 5%) Test method differentiated between RAP and RCP samples containing 50-percent virgin aggregate. OMC and Saturated MeanRCP100 = MeanRAP100 0.641 (> 5%) Test method did not differentiate between 100 percent RCP and 100-percent RAP. * Includes 100-percent and 50-percent blend recycled material samples. Table 5.6. Statistics for Resilient Modulus data at 5 percent test significance. Traffic Temperature Condition Moisture Condition High Medium Low High 4 4 3 Freezing Low 4 3 2 High 3 2 2 Non Freezing Low 3 2 1 Scale of 1 to 4 with 4 = Most significance, 1 = least significant Table 5.7. Significance level of intended use on aggregate performance potential.
47 0 2 4 6 8 10 12 14 16 18 20 50/50 RCP-LS-IL Virgin DGBL Blend#1 RAP-GV-LA 50/50 RCP-GV-LA Virgin OGDL Blend 50/50 RCP-GR-SC RCP-GR-SC 50/50 RAP-GR-CO 50/50 RCP-GV-LA 50/50 RAP-LS-MS RAP-LS-MS RAP-GR-CO RCP-LS-IL RCP-GV-LA Percent Material Loss 4 Significance Level 3 2 and less Figure 5.18. Performance potential based on toughness (Micro-Deval) test. Traffic H M H L M L Moisture H L H L H L H L Tests (Test Parameters) Temperature F NF F NF Micro-Deval Test (percent loss) < 5 percent < 15 percent < 30 percent < 45 percent Tube Suction Test (dielectric constant) 7 10 15 20 OMC, c = 5psi 100 psi 60 psi 25 psi Not required Static Triaxial Test (Max. deviator stress) Sat. c = 15psi 180 psi 135 psi 60 psi Not required OMC, c = 15psi 180 psi 160 psi 90 psi Not required Repeated Load Test (Failure deviator stress) Sat. c = 15psi 180 psi 160 psi 60 psi Not required Stiffness Test (Resilient modulus) 60 ksi 40 ksi 25 ksi Not required Table 5.8. Recommended tests and test parameters for levels of intended use. shown Figure 5.18, indicate that recycled materials are generally appropriate for use in medium to low traffic conditions in non- freezing climates with low and high moisture contents. RCP- GR and RAP-GV seem appropriate for use in high traffic areas with non-freezing temperatures or in low and medium traffic areas in freezing climates with low moisture conditions. Adding virgin aggregate to recycled materials improves the performance potential (based on the toughness test). For ex- ample, virgin aggregates and 50-percent blend of recycled materials with virgin aggregate are appropriate for use in low and medium traffic areas in freezing climates with low mois- ture conditions or high traffic areas with non-freezing tem- peratures with high or low moisture conditions. None of the Proposed ranges for selected test parameters that relate to performance are shown in Table 5.8 for various levels of cli- matic and traffic condition. These ranges determine the traf- fic and climatic conditions where these recycled materials and their blends can be used. However, results from acceler- ated pavement tests and/or in-service test pavement evalua- tions are needed to confirm or refine these ranges. Selection Based on Toughness Test Recycled materials and virgin aggregate toughness and abra- sion resistance characteristics were evaluated using the Micro- Deval test. The test results and recommended test parameter,
48 Selection Based on Repeated Load Triaxial Test Failure deviator stress for repeated load triaxial tests conducted at OMC, shown in Figure 5.21, indicate that vir- gin aggregates are appropriate for use in high traffic condi- tions (significance level 4). RCP and 50-percent RCP blend with LS and GR are appropriate for use in conditions rep- resenting significance level 3 (i.e., high traffic level in non- freezing temperatures, medium traffic level in freezing temperature in the presence of low moisture, and low traf- fic level in freezing temperatures). RAP and 50-percent RAP blends are generally appropriate for use in conditions representing significance level 2. Failure deviator stress for repeated load triaxial tests conducted in saturation condition, shown in Figure 5.22, indicate that RAP-GR and RCP-LS are appropriate for use in high moisture conditions with low or medium traffic and non-freezing temperatures. The other materials are appropriate for use in conditions representing significance level 2. Selection Based on Material Stiffness Most recycled materials and 50-percent blends with virgin aggregate were shown to be appropriate for use in conditions representing significance level 3, as shown in Figure 5.23. Vir- gin aggregates were shown to be appropriate for use in con- ditions representing significance level 4. 0 5 10 15 20 25 RAP-GV-LA RAP-LS-MS 50/50 RAP-LS-MS RAP-GR-CO RAP-GR-CO 100%OGDL reblend 50/50 RAP-GV-LA 50/50 RAP-GR-CO RAP-GR-CO 50/50 OGDL reblend Virgin OGDL Blend RCP-GR-SC 50/50 OGDL reblend Virgin DGBL Blend#1 50/50 RCP-GR-SC RCP-GR-SC 100%OGDL reblend Virgin DGBL Blend#2 RCP-GR-SC 50/50 RCP-GV-LA RCP-GV-LA RCP-LS-IL 50/50 RCP-LS-IL Final Dielectric Constant 4Significance Level 3 12 Figure 5.19. Performance potential based on frost susceptibility. tested materials (virgin, recycled or 50-percent blend of virgin and recycled materials) are appropriate for use in perfor- mance significance level 4 (high traffic locations with freezing temperatures and low and high moisture conditions). Selection Based on Frost Susceptibility Test Frost susceptibility of different recycled and virgin aggre- gates was determined using the tube suction test; results are shown in Figure 5.19. The results indicate that RCP materials are appropriate for use only in performance significance level 2 (medium traffic no freezing) and level 1 (low traffic, no freez- ing, low moisture). Blending RCP with virgin aggregate in- creased the performance potential to the next level, and thus would be appropriate for use in high traffic (no freezing) and medium traffic (freezing with low moisture condition). RAP and 50-percent blends with virgin aggregate are appropriate for use in high traffic conditions. Selection Based on Static Triaxial Test The results of the static triaxial test, shown in Figure 5.20, indicate that most of the RCP materials and their blends with virgin aggregate are appropriate for use in extreme traffic and climatic conditions (significance level 4). RAP, on the other hand, is appropriate for use in conditions representing sig- nificance level 3 (i.e., high traffic level in non-freezing tem- peratures, medium traffic level in freezing temperature in the presence of low moisture, and low traffic level in freezing temperatures).
49 70 80 90 100 110 120 130 140 150 160 170 180 190 RAP-GV-LA (OGDL) RAP-GV-LA (50) (OGDL) RCP-SC-GR (OGDL) RAP-GR-CO (50) (OGDL) RAP-LS-MS (OGDL) RCP-GV-LA (50) (DGBL) RAP-LS-MS (50) (OGDL) RCP-SC-GR (50) (OGDL) RAP-GR-CO (OGDL) RAP-GR-CO (50) (DGBL#2) RCP-GR-SC (DGBL) Virgin OGDL RCP-GR-SC (50) (DGBL) RCP-IL-LS (DGBL) RAP-GR-CO (DGBL#2) RCP-GV-LA (DGBL) RCP-IL-LS (50) (DGBL) Virgin (DGBL#2) Virgin (DGBL#1) Failure Deviator Stress, psi Significance Level 2 431* * Test not required for conditions representing significance level 1 Figure 5.21. Performance potential based on repeated load triaxial test (OMC). 0 20 40 60 80 100 120 140 RAP-GV-LA RAP-GR-CO RAP-LS-MS RAP-GR-CO 50/50 R RAP-GR-CO 50/50 Virgin (OGDL) RCP-GV-LA RCP-LS-IL RCP-GR-SC 50/50 R RCP-LS-IL 50/50 Virgin (DGBL#2) RCP-GR-SC 50/50 RCP-GR-SC R Virgin (DGBL#1) RAP-GR-CO R RCP-GR-SC Maximum Deviator Stress, psi (at 15 psi confining stress) * Test not required for conditions representing significance level 1 Significance Level 2 4 3 1* Figure 5.20. Performance potential of recycled materials based on static triaxial test.
50 0 10 20 30 40 50 60 70 RCP-GR-SC 50/50 RCP-GR-SC OGDL reblend 50/50 RCP-GR-SC OGDL reblend 50/50 RCP-GR-SC RAP-GV-LA 50/50 RCP-LS-IL RAP-GR-CO RCP-LS-IL 50/50 RAP-GR-CO OGDL reblend 50/50 RCP-GV-LA RAP-GR-CO OGDL reblend 50/50 RAP-GV-LA 50/50 RAP-GR-CO RAP-LS-MS RCP-GV-LA 50/50 RAP-LS-MS Virgin DGBL Blend#2 Virgin OGDL Blend Virgin DGBL Blend#1 Resilient Modulus (stiffness) at 100 psi Bulk Stress, ksi Significance Level 2 4 31* * Test not required for conditions representing significance level 1 Figure 5.23. RAP and RCP performance potential based on stiffness (at OMC). 100 110 120 130 140 150 160 170 180 190 RCP-GV-LA (DGBL) Virgin (DGBL#2) RCP-IL-LS (50) (DGBL) RCP-GV-LA (50) (DGBL) RCP-GR-SC (50) (DGBL) RCP-GR-SC (DGBL) RAP-GR-CO (50) (DGBL#2) RAP-GR-CO (DGBL#2) Virgin (DGBL#1) RCP-IL-LS (DGBL) Failure Deviator Stress, psi Significance Level 2 4 3 Figure 5.22. Performance potential based on repeated load triaxial test (saturated).
