Abstract
Cold central plant recycled asphalt mixtures (CCPR) have been shown to provide a high-quality, economical, and environmentally conscious asphalt base mixture. Existing literature, however, does not indicate what would be an appropriate future rehabilitation method for a pavement that includes CCPR. Given the previously cited benefits of CCPR, it is logical to ask whether a CCPR can be re-recycled and, if so, how will it perform? This paper presents a study of a test section containing a re-recycled CCPR placed at the National Center for Asphalt Technology Test Track. CCPR from a Virginia Department of Transport study on the Test Track was recovered, re-recycled, and placed at 5-in. thick with a 2-in. asphalt concrete (AC) overlay. In preparing for a re-recycled CCPR layer, milling an existing CCPR layer resulted in a coarser gradation and lower indirect tensile strength values. Initially, the re-recycled CCPR test section did not perform well owing to moisture in the aggregate base present during construction of the re-recycled CCPR layer. It is suspected that this moisture negatively affected the curing of the re-recycled CCPR layer. Thus, the re-recycled CCPR layer was milled and re-recycled again (3rd generation re-recycled CCPR) and placed at 5-in. thick with a 2-in. AC overlay. The 3rd generation re-recycled CCPR performed well through 3 million equivalent single axle loads (ESALs), the same number of ESALs that resulted in fatigue cracking and rutting failures in the initial re-recycled CCPR.
Keywords
Cold central plant recycling (CCPR) is a cold recycling process in which stockpiled or newly milled reclaimed asphalt pavement (RAP) is mixed with a recycling agent (such as emulsified asphalt or foamed asphalt) to create a bound asphalt base layer ( 1 ). CCPR has several cost and environmental benefits ( 2 ) because it reuses existing pavement material and is manufactured without heating the RAP materials. Virginia Department of Transportation (VDOT) has been a leader in investigating the use of CCPR, especially in high traffic applications. In 2011, VDOT contracted to reconstruct a section of Interstate 81 using CCPR and full-depth reclamation (FDR), another pavement recycling process, in the right lane along a 3.7-mi portion of the highway ( 3 ).
To better understand the structural performance of pavements using CCPR, VDOT sponsored the construction and monitoring of three pavement test sections at the National Center for Asphalt Technology (NCAT) Test Track in 2012. Test sections at the NCAT Test Track are subjected to 10 million equivalent single axle loads (ESALs) in 2 years applied by loaded trailers driven around the track ( 4 ). After three track cycles (note that a cycle constitutes 2 years of traffic followed by a year of forensic work and reconstruction of some track sections), encompassing 9 years of testing and more than 30 million ESALs, one of the VDOT test sections placed in 2012 that contained CCPR over an FDR-like foundation was taken out of service ( 5 , 6 ). This was done because analysis of the responses from embedded instrumentation showed that this section was performing like a perpetual pavement ( 7 ) and additional trafficking would most likely only confirm these results. After this section was removed from study, VDOT decided to sponsor the reconstruction of this test section to learn more about what it considered a likely rehabilitation strategy for recycled pavements. VDOT decided to investigate the performance of a re-recycled CCPR material, that is, the performance of a pavement containing a CCPR layer that was produced from material previously recycled in a CCPR process.
Through personal communication with contractors and other agency personnel experienced with pavement recycling, it was found that re-recycling of previously recycled materials has been done elsewhere. However, the authors of this paper were unable to identify documentation of the techniques used or performance of these re-recycled sections through a literature search. Thus, performance data of pavements comprising materials that were previously recycled in a cold recycling process are not widely available.
Objectives and Scope of Work
The objective of this study was to evaluate the performance of a pavement test section that was constructed using CCPR produced from materials that were previously placed as a CCPR layer in 2012 and then subjected to accelerated trafficking. This study focuses on the mix design, preparation and construction, and results of testing for a test section at the NCAT Test Track constructed using re-recycled CCPR. Throughout this paper, the CCPR mixtures will be designated using the terms original CCPR and re-recycled CCPR to denote the sections constructed in 2012 and 2021, respectively.
Materials and Methods
The pavement cross section investigated in this study consisted of 5 in. of re-recycled CCPR and a 2-in. asphalt concrete (AC) stone matrix asphalt (SMA) overlay. The AC layer used to surface the re-recycled CCPR section was an SMA mixture that was made using an accepted VDOT design. The constituent materials to produce the SMA were shipped to the NCAT Test Track from the Virginia asphalt mixture producer and used to locally produce the SMA for the Test Track.
Milling of Existing CCPR
The original CCPR section consisted of approximately 2 in. of SMA mixture above 2 in. of a dense graded asphalt (DGA) mixture. This was placed above 4.3 in. of CCPR and an FDR-like cement stabilized layer ( 4 ). To prepare the section for the re-recycled CCPR experiment, the existing SMA and DGA layers were milled and removed. The milling was done carefully so as not to cause damage to the existing CCPR layer so that its condition could be evaluated (Figure 1a). Next, the CCPR layer was milled (Figure 1b) and stockpiled so these materials could be used to produce the re-recycled CCPR layer. The milling was again completed carefully to ensure that that the underlying cement stabilized layer was not incorporated into the recovered CCPR material (Figure 1c). The milled and stockpiled original CCPR materials (Figure 2) were used to develop the mix design for and to produce the re-recycled CCPR layer. Finally, the FDR layer was removed and replaced with a crushed granite aggregate base material commonly used on the Test Track to enable comparisons with other previously constructed test sections (Figure 1d). More information about the milling process can be found in Bowers et al. ( 6 ).
(a) Milling of SMA layer, (b) Milling of CCPR layer, (c) Clean FDR surface, and (d) Replaced FDR layer with aggregate base.
Stockpiled CCPR.
Mix Design of 2nd Generation Re-Recycled CCPR
The recovered CCPR materials were used to develop a mix design for the 2nd generation re-recycled CCPR following AASHTO MP 38-18 Standard Specification for Mix Design of Cold Recycled Mixture with Foamed Asphalt, and AASHTO MP 94-18 Determination of Optimum Asphalt Content of Cold Recycled Mixture with Foamed Asphalt. Indirect tensile strength (ITS) test specimens were compacted using Marshall compaction, 75 blows per side. Specimens were cured at 40°C for 72 h and then separated into “dry” and “conditioned” specimens. Dry ITS specimens were stored at ambient laboratory conditions for 24 h before testing, whereas the conditioned specimens were soaked in a water bath at 25°C for 24 h and then tested. Testing was conducted in accordance with AASHTO T 283 Standard Method of Test for Resistance of Compacted Asphalt Mixtures to Moisture-Induced Damage.
Test Track Performance Monitoring
Performance characteristics of the Test Track sections are monitored using instrumentation installed during construction, by measurements of the ride quality, wheel path rutting, and cracking, and deflection testing as trafficking is applied. During construction of the 2nd generation re-recycled CCPR section, the section was instrumented with strain, pressure, and temperature sensors. Surface performance and mechanistic response measurements were made on a routine basis. Surface performance measurements obtained with a Pathways van included ride quality (international roughness index [IRI]), wheelpath rutting, and cracking. These measurements were made on a weekly basis in addition to weekly visual inspections of the sections once trafficking began.
The instrumentation plan for the 2nd generation re-recycled CCPR section was instrumented following a well-established layout and procedure that has been used at the Test Track since 2006 ( 8 ). The strain gauges comprised a full Wheatstone bridge with four active strain gauges procured from a commercial supplier and were placed at the bottom of the CCPR to capture tensile strain generated at the bottom of the AC/CCPR layer. The temperature probes, from a different commercial supplier, were thermistors, placed at the top, middle, and bottom of the AC/CCPR and 3 in. into the aggregate base to measure the thermal gradient versus depth. All of the temperature data presented in this paper are from the mid-depth temperature probe as this provided the strongest correlation to measured pavement responses. Pressure plates were also installed in the section; measurements from these sensors are not included in this paper as the focus was on the strain measurements.
Falling weight deflectometer (FWD) testing and backcalculation was also conducted several times per month on the section. FWD testing consisted of testing 12 locations in the sections (4 longitudinal stations × 3 offsets [outside, inside, and between wheelpaths]) with three drop heights (6, 9, and 12 kip) and three replicates at each drop height. A nine-sensor configuration was used to measure deflections and the load plate had a radius of 5.91 in. Backcalculation was conducted using EVERCALC 5.0 where the pavement was divided into three layers: Layer 1 = AC/CCPR; Layer 2 = aggregate base; Layer 3 = subgrade. This setup was consistent with previous backcalculations used on other CCPR sections at the Test Track ( 4 , 9 ); this was predicated on CCPR having a modulus reasonably close to that of AC, being temperature sensitive, and very difficult to separate out during backcalculation.
Analysis and Discussion
Mix Design
The milled and stockpiled original CCPR materials were processed for mix design use in accordance with AASHTO MP38 and AASHTO PP94 for cold recycled mixtures with foamed asphalt. Foamed asphalt was selected as the recycling agent to facilitate comparisons with the CCPR sections constructed in 2012. First, a “black-rock” gradation (gradation of the RAP with the RAP binder still on the aggregate) was performed to evaluate whether the gradation would pass mix design material requirements, along with a post-ignition gradation to compare to the post ignition gradation of the original 2012 section. It was found that the “black-rock” gradation was sufficient provided 1% cement was added as an active filler. This is common in CCPR mixes, namely with foamed asphalt, as it helps to distribute the foamed asphalt through the mixture ( 1 ). Interestingly, the post-ignition gradation for the 2nd generation re-recycled CCPR was found to be finer than the original CCPR post-ignition gradation, as shown in Figure 3. This makes sense as the material is being pulverized and processed in the mobile plant for a second time.
Black-rock gradation for original CCPR, 2nd generation re-recycled CCPR, and AASHTO recommended gradation limits.
A PG 67-22 binder, the common binder used in Auburn, AL, was used as the recycling agent. The foaming properties of the binder were tested and a 2.3% water content at 165°C was established as the required water content and temperature to achieve the desired expansion ratio and half-life. One percent cement by weight of dry RAP was added as an active filler. The optimum moisture content was found to be 6.8% based on AASHTO T 180, Method D Proctor density curve. To select the recycling agent content, three contents were tested by weight of dry RAP: 1.8%, 2.0%, and 2.2%. The highest dry strength with an acceptable tensile strength ratio (TSR) was selected as the mix design.
Of the three trials, the mix design process found that 2.0% recycling agent content resulted in the maximum strength, identical to the original CCPR design from 2012. When comparing the original CCPR mix design to the 2nd generation re-recycled CCPR mix design, the average dry and conditioned strength ITS values were found to reduce from 83 to 65 pounds per square inch (psi) and from 63 to 55 psi, respectively. However, the strength values for the 2nd generation re-recycled CCPR still exceeded AASHTO MP 38 requirements for dry and conditioned specimens of 45 and 31.7 psi, respectively. Potential causes of the strength reduction are (1) coarser gradation or (2) the presence of additional asphalt binder leading to a softening of the mixture (the original CCPR mixture had 2.0% recycling agent added and the 2nd generation re-recycled CCPR required an additional 2.0%). A comparison of the original CCPR design to the 2nd generation re-recycled CCPR design is provided in Table 1.
Comparison of the Original CCPR Mix Design to the Re-Recycled Mix Design
Note: CCPR = cold central plant recycled asphalt mixture; min. = minimum; NA = not available; psi = pounds per square inch.
Construction
Construction commenced on September 8, 2021. The night before construction of the 2nd generation re-recycled CCPR layer, the Test Track area experienced a significant rainstorm. On the day of construction, the section was found to have water ponded on the surface of the aggregate base. The research team was concerned that the ponded water could cause the aggregate base to lose the necessary support for the CCPR layer to be placed on it. To verify the condition of the aggregate base, dynamic cone penetrometer (DCP) testing was conducted on areas of ponded water and areas that were relatively dry. The results of the DCP testing showed that the two conditions were similar, with penetration index values that ranged from 2 to 5 mm per blow. Based on this information, the research team, in consultation and agreement with VDOT researchers, made the decision to move forward with construction.
The 2nd generation re-recycled CCPR mixture was produced using a Wirtgen KMA 240i mobile CCPR plant, which was located on site. The mixture was loaded into the back of a dump truck, then taken to the test section and loaded into the paver hopper of a conventional paver. Once the 2nd generation re-recycled CCPR was paved, the mixture was compacted using a double drum asphalt roller (vibratory), a pneumatic tire roller, and a double drum asphalt roller (oscillatory). The total number of passes was 11 to 13 depending on location. The average wet density of the CCPR mixture was 130.5 pounds per cubic foot (pcf) during construction, which was the maximum achievable density in the field.
Samples of the 2nd generation re-recycled CCPR were collected from the CCPR plant and tested for recycling agent content, moisture content, and compacted for ITS testing. Specimens for ITS testing were compacted to 30 gyrations using a Superpave Gyratory Compactor and a 6-in. diameter mold. The specimens were cured according to the mix design curing regime. As shown in Table 2, the dry and conditioned strength requirements were still met, though there was a reduction in comparison to the design. As shown in Table 2, the dry lab density was within 1.2 pcf between the design and construction specimens compacted in the laboratory. This was considered acceptable by the team as there will inevitably be slight variations between the materials used for lab versus a complete construction project, along with differences in mixing and time-to-compaction, which could potentially affect dry density. Ultimately, the mix was deemed acceptable during construction and the 2-in. SMA overlay was applied the next day. A PG 67-22 tack coat was applied before placement of the overlay. The final constructed thickness of the SMA overlay was measured at 1.7 in. and the average measured mat compaction was 97.9% of the maximum mixture specific gravity (Gmm) as measured with a nuclear density gauge.
Comparison of the Re-Recycled CCPR Mix Design Versus the 2nd generation Re-Recycled CCPR Mixture During Construction
Note: CCPR = cold central plant recycled asphalt mixture; pcf = pounds per cubic foot; psi = pounds per square inch; na = not applicable.
Field Performance
Surface performance monitoring was conducted on a weekly basis, which included measurements of rutting, cracking, and ride quality. As shown in Figure 4 the section rapidly reached 0.5 in. of rutting, considered the failure point at the Test Track, after approximately 2 million ESALs. Figure 5 shows standing water in the wheelpaths after a rain event at approximately 1.75 million ESALs and 0.35 in. of rutting.
Rutting and ESALs versus time.
Water in wheelpaths at 1.75 million ESALs and 0.35 in. of rutting.
Although the section was clearly experiencing significant levels of rutting, the more severe distress was wheelpath cracking that appeared consistent with previous studies at the Test Track, which were eventually shown to be bottom-up fatigue cracking ( 10 ). The cracking in the section was first measured in mid-January 2022 at approximately 400,000 ESALs. Figure 6 plots the amount of cracking and ESALs versus time with a dramatic increase in the amount of cracking in the spring months (April to May) of 2022. The percentages represent the percent of lane area and percent of wheelpath area cracked, respectively. The cracking was mostly confined to the second half of the test section, with greater severity apparent near the end of the section. This was consistent with the location of ponded moisture present at the time of construction. Figure 7 shows the nature and extent of cracking at the end of February 2022 after approximately 800,000 ESALs, whereas Figure 8 shows the section at the end of July 2022 at 2.3 million ESALs when the amount of cracking had mostly stabilized but the severity had increased greatly. Both Figures 7 and 8 are from the end of the section.
Percent area cracked and ESALs versus time.
Cracking at 800,000 ESALs.
Cracking at 2.3 million ESALs.
Though the section began experiencing structural distresses, as described, relatively early in the test cycle, the distresses did not adversely affect ride quality until much later. As shown in Figure 9, the IRI values began increasing dramatically and reached unacceptable levels approaching 200 in./mi at about 2 million ESALs (July 2002), which necessitated section replacement.
Ride quality and ESALs versus time.
Roughness
While the distresses detailed above were being measured at the pavement surface, subsurface strain responses were being recorded on a weekly basis and are shown in Figure 10. The strains represent the 95th percentile best-hit response measured from 12 asphalt strain gauges under single axles from approximately 20 truck passes, with each truck pass representing five legally loaded single axles. From the very beginning, the tensile strain levels were extraordinarily high, ranging from 1,000 to 1,800 microstrain, in comparison to other previously built CCPR sections at the Test Track that had strains generally below 1,000 microstrain ( 5 ). These high strain levels were thought to have led to the observed fatigue cracking. Figure 10 shows that the strain levels also increased with increasing temperature, especially from mid-January to the end of June 2022. This type of behavior is normal and expected from healthy pavements: the temperature increases, the modulus of the AC decreases, and strains consequently increase. However, close examination of the data through mid-January 2022, when cracking first became evident, showed an increase in strain over that period with decreasing temperatures. This indicated that cracking was affecting the measurements, which lent support to the hypothesis that the cracks were coming from the bottom rather than top-down. Furthermore, after the end of June 2022, the strain measurements had high variability, indicative of a failed pavement. For the intervening time between mid-January and June 2022, when the strain levels increased with temperature, it is difficult to discern what effect accumulating damage had on the strain measurements because temperature and damage are both strong influencing factors.
Tensile strain and mid-depth temperature versus time.
Backcalculating the layer properties of the section proved challenging because of the relatively high root mean square error (RMSE) obtained in the backcalculation process. A commonly used RMSE maximum is 3% to discern between acceptable and unacceptable results. Of the 714 deflection basins measured within the section from the start of traffic until it was taken out of service in mid-October, only 4.4% generated an RMSE at 3% or less, which could indicate a poor match between the actual behavior of the section and layered elastic theory. Given the extensive and rapidly forming cracking in the section, this is most likely the case. The AC/CCPR backcalculated moduli that passed the 3% RMSE check, and normalized to a reference temperature of 68°F, are shown in Figure 11; the large gaps in time correspond to stretches where no usable data were obtained (i.e., RMSE > 3%). The data showed generally low moduli ranging from 100 to 300 kips per square inch (ksi), an average of 180 ksi, and with no obvious trend in the data over time. Unfortunately, this data set was less conclusive than the others in characterizing the in situ structural performance, but the relatively low moduli did support the relatively high measured tensile strain values.
Backcalculated AC/CCPR modulus versus time (RMSE < 3%).
Forensics
On February 28, 2022, approximately 5.5 months after construction, cores were taken from the cracked and uncracked areas to identify the source of the cracking. However, no fully intact cores could be extracted from either cracked (wheelpath) or uncracked (between wheelpaths) areas of the pavement. Examples of the cores are shown in Figure 12: the SMA layer is intact and the bottom of the CCPR layer has disintegrated. Although it is not uncommon for core collection to be difficult from newly paved CCPR ( 1 ), it is typically possible after several weeks.
Cores taken from cracked and uncracked areas.
As noted previously, there was a precipitation event that occurred just before construction. Based on the results shown in Figure 12, it is suggested that the 2nd generation re-recycled CCPR layer experienced rapid disintegration owing to a lack of curing. The 2nd generation re-recycled CCPR layer most likely never reached a cured condition because of excessive moisture being retained within the aggregate base during construction.
Experimental Continuation: 3rd Generation Re-Recycled CCPR
Once the research team had reviewed the results of testing and coring, the lack of curing resulting from excessive moisture was considered to be the most likely culprit of the early failure of the 2nd generation re-recycled CCPR layer. To continue the experiment and learn about re-recycling of CCPR, VDOT decided to reconstruct the section by re-recycling the CCPR again (henceforth called 3rd generation re-recycled CCPR) and placing a new 2-in. SMA overlay.
Collection of Materials for Mix Design
In September 2022, materials were collected to perform the mix design on a 3rd generation re-recycled CCPR. A slab was cut in the pavement and removed using a forklift. At this time, the 2nd generation re-recycled CCPR was found to be intact and well bonded to the SMA overlay, though the CCPR was friable (Figure 13). This further supported the curing hypothesis presented previously, as by nearly 1 year after construction, the 2nd generation re-recycled CCPR layer had had additional time to cure during the summer months. CCPR materials were obtained from the slabs and placed in 5-gal buckets, and the SMA was also maintained in case it was required in the mix design.
Slab extraction to collect materials for mix design.
Material Processing
Once millings of both the SMA and 2nd generation re-recycled CCPR layers were collected, they were processed through an aggregate jaw crusher and passed through a 1.5-in. sieve. All passing material was retained for use in mixture design. After processing, the material was transported to the NCAT laboratory where a blend was created using 15% of the SMA layer and 85% of the 2nd generation re-recycled CCPR layer. A portion of the SMA was used in the blend to ensure that sufficient material could be generated to produce the 3rd generation re-recycled CCPR. This blend was then subjected to a modified Proctor test following AASHTO T180, Method D, to determine the optimum moisture content, resulting in an optimum moisture content of 6.8%. The virgin asphalt binder used for this mixture design, a PG 67-22, and optimum foaming properties were again achieved at a temperature of 165°C and a water content of 2.5%.
Mixture Design
The mix design process found that 2.1% recycling agent content resulted in the maximum strength of the three trials. When comparing the original CCPR mix design to the 2nd generation re-recycled CCPR mix design, the average dry and conditioned ITS values were found to reduce from 65 to 53 psi and from 55 to 34 psi, respectively. The strength values for the 3rd generation re-recycled CCPR mix design exceeded AASHTO MP 38 requirements for dry specimens of 45 psi. However, the TSR for the mix design was less than the MP 38 minimum value of 70% of the dry ITS value. The research team decided that the minimum conditioned ITS value reflected in MP 38 (70% of 45 psi, or 31.7 psi) was more important and thus the design was approved. It was decided during production to reduce the recycling agent content to 2.0% from 2.1%, which aligned more closely with the 2nd generation re-recycled CCPR placed in 2021. This judgment was based on the design range used (1.8%, 2.1%, and 2.4%) and that the dry ITS value at 1.8% foamed asphalt was just below the target dry ITS value (44 psi) and the wet ITS value was identical to that of 2.1% foamed asphalt (34 psi). A comparison of the 2nd generation re-recycled CCPR design (2021) and the 3rd generation re-recycled CCPR design (2022) is provided in Table 3.
Comparison of the 2nd generation Re-Recycled CCPR Mix Design to the 3rd generation Re-Recycled CCPR Mix Design
Note: CCPR = cold central plant recycled asphalt mixture; min. = minimum; pcf = pounds per cubic foot; psi = pounds per square inch.
Mixture Production and Placement
The test section was milled in two lifts, with the SMA being milled first followed by the CCPR layer, with each mixture being stockpiled separately. In October 2022, the milled SMA and 2nd generation re-recycled CCPR were used to construct the 3rd generation re-recycled CCPR layer. The SMA and 2nd generation re-recycled CCPR were blended at their respective 15% to 85% ratio in the Wirtgen KMA mobile CCPR plant. The correct dosages of water, active filler, and recycling agent were added and the mixture was discharged into a dump truck that fed material into the paver hopper. The section was placed at 5-in. thick with a 2-in. SMA overlay. Placement occurred as described previously.
Preliminary Performance
The 3rd generation re-recycled CCPR was again subjected to the same trafficking and performance monitoring activities as described earlier. At the time of writing this paper, the 3rd generation re-recycled CCPR section has experienced approximately the same amount of traffic (3 million ESALs) as the 2nd generation re-recycled CCPR section built in 2021 that failed prematurely.
Overall, the 3rd generation CCPR section is performing well through 3 million ESALs. No cracking is evident at the surface of the pavement. Rut depths, shown in Figure 14, have been steady over time. Initial rut depths started at about 0.15 in. but this is believed to be a function of the automated scanning lasers measuring outside the width of the newly constructed lane into the adjacent inside lane and shoulder. This phenomenon has been observed on several other sections at the Test Track using the same rutting measurement technology ( 6 ). Finally, the IRI has been steady or falling since traffic began on the section, indicating good surface performance, as shown in Figure 15.
Rutting and ESALs versus time in 3rd generation re-recycled CCPR section.
Surface roughness and ESALs versus time in 3rd generation re-recycled CCPR section.
The measured tensile strain responses in the 3rd generation re-recycled CCPR section are shown in Figure 16 and are mostly between 800 and 1,400 microstrain, which is markedly lower than those measured in the 2nd generation re-recycled CCPR section built in 2021 (Figure 10). There is also a strong correlation between temperature and strain, shown in Figure 17, which suggests that there is no unobserved subsurface cracking affecting the readings.
Tensile strain versus time in the 3rd generation re-recycled CCPR section.
Tensile strain versus mid-depth AC/CCPR temperature.
The backcalculated AC/CCPR modulus of the 3rd generation re-recycled CCPR section versus time suffered from the same RMSE problem encountered with the original section, but not to the same extent. As mentioned, only 4.4% of the data met the 3% RMSE criteria in the 2021 re-recycled section, whereas the 3rd generation section had 8.1% of the 402 deflection tests generating usable backcalculated data, which are shown in Figure 18 normalized to the same 68°F reference temperature. Again, the results to date are limited, however, nearly twice the data are usable within a similar time frame. The deflection data results have not been found to change with respect to time and range from 100 to 200 ksi with an average of 140 ksi as shown in Figure 18.
Backcalculated AC/CCPR modulus versus time (RMSE < 3%) in the 3rd generation re-recycled CCPR section.
Findings, Conclusions, and Recommendations
The following findings, conclusions, and recommendations can be made based on this study:
Findings
The 2nd generation re-recycled CCPR mix design (2021) exhibited a finer gradation as well as a lower overall strength than the original CCPR mix design (2012).
The 2nd generation re-recycled CCPR (2021) was found to experience bottom-up fatigue cracking after approximately 400,000 ESALs and reached the Test Track rutting threshold of 0.5 in. after approximately 1.75 million ESALs.
The 2nd generation re-recycled CCPR (2021) experienced extraordinarily high tensile strain at the bottom of the layer, which led to the fatigue cracking observed at the surface of the pavement.
Backcalculating modulus from FWD data proved challenging owing to the cracking and subsequently high RMSE. From the limited data that were collected, the moduli of the 2nd generation re-recycled CCPR materials (2021) were lower than expected, ranging from 100 to 300 ksi, giving an average of 180 ksi.
During forensic coring of the 2nd generation re-recycled CCPR (2021) it was found that the cores were not intact, indicating a lack of curing.
The 3rd generation re-recycled CCPR mix design (2022) exhibited lower dry and wet ITS values than the 2nd generation re-recycled CCPR mix design (2021).
The 3rd generation re-recycled CCPR section (2022) has experienced approximately the same amount of traffic (3M ESALs) as the failed 2nd generation re-recycled CCPR section (2021), but has not exhibited cracking or rutting, and has seen a reduction in IRI since trafficking began.
The 3rd generation re-recycled CCPR section (2022) is exhibiting much lower tensile strains at the bottom of the CCPR layer than the failed 2nd generation re-recycled CCPR section (2021), and the strain correlates well with temperature, indicating that there is no unobserved cracking.
The 3rd generation re-recycled CCPR section (2022) has the same challenges with backcalculated modulus from FWD data with respect to RMSE, however, more data comprising a lower RMSE are available. The range of modulus for the 3rd generation re-recycled CCPR is smaller and lower than that of the failed 2nd generation re-recycled CCPR section (2021) at 100 to 200 ksi and an average of 140 ksi.
Conclusions
It is possible to re-recycle CCPR and get a passing mix design as well as good early field performance.
Potential causes of the strength reduction between the original CCPR mix design and 2nd generation re-recycled CCPR mix design (2021) are (1) change in gradation or (2) the presence of additional asphalt binder leading to a softening of the mixture (the original CCPR mixture had 2.0% recycling agent added and the 2nd generation re-recycled CCPR required an additional 2.0%).
Curing, and thus performance of the CCPR layer, can be negatively affected by the level of moisture in the underlying layers and should be considered during construction.
The 3rd generation CCPR mix design TSR value was lower than the AASHTO MP 38 minimum TSR value, but the mix design was deemed acceptable by the project team based on the actual wet ITS value.
It is possible to achieve good early performance (beyond 3 million ESALs to date) in the field under heavy loading with re-recycled CCPR and a 2-in. AC overlay, as shown by the 3rd generation re-recycled CCPR.
Recommendations
Studies should be conducted to determine whether it is more appropriate to consider a minimum wet ITS value for CCPR mix designs versus a minimum TSR value.
VDOT should continue testing the 3rd generation CCPR section to determine the performance of the material.
Footnotes
Acknowledgements
The CCPR sections were sponsored by Virginia Department of Transportation (VDOT). The authors are grateful for the support and cooperation of the many individuals from VDOT and the contractors involved in the experiments.
Author Contributions
The authors confirm contribution to the paper as follows: study conception and design: B. Bowers, B. Diefenderfer, D. Timm; data collection: B. Bowers, D. Timm, B. Diefenderfer, E. Turochy; analysis and interpretation of results: B. Bowers, B. Diefenderfer, D. Timm, E. Turochy; draft manuscript preparation: B. Bowers, B. Diefenderfer, D. Timm, E. Turochy. All authors reviewed the results and approved the final version of the manuscript.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Virginia Department of Transportation.
