Introduction
Full Depth Reclamation (FDR) is a recycling technique in which all of the asphalt pavement section as well as a predetermined amount of underlying base materials are treated, as defined by the Asphalt Recycling and Reclaiming Association [
1]. This mixture is pulverized, mixed with a stabilizing agent, and compacted to produce a thicker, stabilized base course, shown in Fig. 1. FDR is typically performed to a depth of 100 to 300 mm, or 4 to 12 inches. Often, due to structural capabilities of this blend of material, it is sufficient to act as the base for a new surface wear course without the addition of stabilizing additives. However, if it is determined that the in-situ material is not adequate in structural strength, three different types of stabilization exist: mechanical, chemical, and asphaltic. These types of stabilization can either be performed independently or may also be combined and used in conjunction with one another [
2].
Full Depth Reclamation has several major advantages and benefits. Some of these, as previously mentioned, are the cost efficiency and environmental benefits of FDR. In addition to being a sustainable option, FDR is also an effective option for rehabilitating a deteriorated pavement section and making it more structurally sound. FDR allows for the improvement of the structure of the pavement without changing the geometry of the pavement or requiring shoulder reconstruction. Distresses such as wheel ruts, potholes, irregularities, rough areas, alligator, transverse, longitudinal and reflection cracks can all be eliminated using FDR, while also restoring old pavement to the desired profile, crown, and slope. Eliminating these problems can also improve the ride quality. FDR is also a viable option for pavements with base or subgrade problems [
2].
Asphaltic FDR is a common type of stabilization used in FDR. Asphalt emulsion is used to increase cohesion of the mixture and the load bearing capacity. It also helps to rejuvenate and soften the aged binder in the existing asphalt material, creating a flexible and fatigue resistant layer not prone to cracking. Emulsion is asphalt dispersed through water and chemically stabilized. Typically, asphalt emulsion contains between 40% and 75% asphalt cement, 0.1%–2.5% emulsifier, and 25%–60% water. The asphalt cement droplets are 0.1 to 20 microns in diameter once suspended.
Asphalt foam, which is becoming increasingly popular for use in FDR, has been shown to increase adhesion properties of the asphalt, making it well suited for mixing with cold or moist aggregates. To create the foam, a small percentage of water is added to hot asphalt cement, causing the liquid asphalt to expand in a small-scale explosion. The amount of water added controls the rate and amount of foamed asphalt. As a result, a thin film of asphalt with about ten times more coating potential than typical asphalt cement is created.
Both emulsified and foamed asphalt have many advantages that make them a preferred choice over typical liquid asphalt. One of these advantages is that emulsion and foam have a lower viscosity, allowing them to be constructed at lower temperatures and increasing the paving season. Additionally, the emissions and energy consumed are decreased because they can be used at lower temperatures. In terms of emissions, the greatest contribution during asphalt production comes from drying and heating the aggregates used in the mixtures. This process also requires the greatest consumption of fuel. Using asphalt emulsion and foamed asphalt does not require this significant heating of the aggregate, but rather uses the in-situ material at ambient temperatures. As a result, about half of the energy required for typical Hot Mix Asphalt is consumed. Additionally, emulsions and foam do not contain any volatile chemicals that evaporate into the atmosphere, also making them more environmentally friendly options than typical asphalt cement [
2].
A problem that is currently being experienced with FDR projects in the field is that the reclaimed surface is susceptible to raveling when being released to traffic too early. Raveling is the progressive disintegration of a pavement layer from the surface downward as a result of the dislodgement of aggregate particles. This is occurring before the surface wear course is being applied. The surface wear course cannot be immediately applied due to the possible trapping of water in the lower areas of the stabilized layer. Therefore, there is a time window of when the reclaimed surface is exposed to traffic because it is very costly to keep the road shut down. The stabilized FDR base should be cured before placing the surface for at least 7 days after the completion of construction. Heavy trucks and construction equipment should not be allowed or at least should be limited during curing, to prevent structural damage to the base, such as flexural cracking [
3]. The friction between the vehicle tires and the reclaimed surface is one of the causes of the raveling problems. Since the asphalt emulsion does not have time to form a strong enough cohesive bond to the aggregate, the aggregate may come loose from the surface of the reclaimed layer when the friction force is applied and thus raveling occurs. For asphalt emulsion stabilized base, a tack coat of diluted slow-setting emulsion is usually applied to ensure good bonding of the HMA or cold-mix overlay [
3]. Currently there is no field test to quantify when the traffic should be released onto the FDR section. It is typically based on engineering judgment and projects may be opened to traffic too early, causing raveling, or too late, causing costly user delays. Due to the variability with each project, there is no established method to determine the optimal return to traffic timing to avoid raveling. In this research, the existing raveling test for Cold In-place Recycling (ASTM D7196, Raveling Test of Cold Mixed Bituminous Emulsion Samples) was used to quantify raveling of FDR mixtures.
Modified ASTM D7196
Currently there is one laboratory scale raveling test being used to evaluate the resistance to raveling of Cold In-Place Recycled (CIR) mixtures. This test, from the slurry surfacing area, was adapted by Koch Materials as a cohesion test for cold mix, ASTM D7196 [
4]. The test involves abrasion of partly (4 hours) cured, compacted cold mix specimens (150mm) with the Wet Track Abrasion Tester used in slurry testing. A CIR sample is placed under a rotating rubber hose so the sample is abraded. The test is run dry for 15 minutes and simulates raveling, which could occur from too early trafficking. According to the author, a partly cured CIR mix ready for traffic should exhibit less than 2% abrasion loss in the test [
5].
This device was used on both asphalt emulsion samples and asphalt foam FDR samples in an attempt to find a correlation between the curing time and method. Unlike ASTM D7196, which is used for CIR pavements, there is no current test method adapted to this device to determine the raveling characteristics of FDR samples. This device was used in a similar manner as is done for CIR samples. The data obtained from this device was to be used as a standard test method to contrast against the research performed with the testers designed in house. An advantage to this inclusion is the proven precision of the test method, compared to the other testers that were built. If a tester showed a similar trend of data to the CIR ravelling test device, it would exhibit a higher level of promise for further evaluation.
The test was originally intended to follow the CIR testing guideline and run for fifteen minutes and measure the mass loss of each sample. This quickly proved to be impossible due to the lack of early strength that the samples had at early curing times. Since FDR has much larger aggregate than CIR, the raveling head often caught one piece of aggregate and the sample would literally explode. After such a failure, the ravelling head would lose contact with the sample. After observing this phenomenon, the quantification by mass loss was dismissed. Instead, the sample performance was quantified based on the amount of time that a sample was able to last in the tester (Time in Tester) before the ravelling head appeared to exert more of a lateral force on the sample and was no longer exerting a downward force. Determining the point when the sample was destroyed was done by the same individual manually for every test that was run in order to maintain a consistent definition. Once this point was reach the test was stopped and the time recorded. A maximum time of ten minutes (six hundred seconds) was set for the test to run.
Mix design
Full-Depth Reclamation (FDR) is a combination of existing asphalt concrete pavement layers, base material, and potentially subbase and subgrade material. In order to evaluate the raveling test for potential use as an evaluation tool of FDR mixtures, material from a local quarry was utilized. This ensured a reliable and consistent supply of material for the entire study. The mix design included 50% processed Recycled Asphalt Pavement (RAP) and 50% Class 7 base (as defined by Arkansas State Highway and Transportation Department). The mix design was performed using the method for asphalt emulsion as laid out by the Asphalt Academy in South Africa [
6]. The first step of this mix design is determining the gradation. Minimum and maximum gradation bands were provided by South Africa, and the gradation of the 50% RAP/50% Class 7 material was designed to pass between the two bands, and shown in Fig. 2.
Since FDR often has soil as a portion of the reclaimed material, the modified proctor test is used to determine the optimal moisture content [
7]. The modified proctor test was run at 2%, 4%, 6%, 8%, and 10%, and an optimal moisture content was established at 5.4%, with a density of 111.1 pcf. Once the gradation and optimal moisture content was determined, a Troxler 5850 with a six-inch slotted mold was used to compact the FDR samples with asphalt emulsion. Samples were compacted to 30 gyrations at 1%, 2%, 4%, 4.25%, 4.75%, 5.5%, and 6% CIR-EE asphalt emulsion from Ergon Asphalt and Emulsions. Using the NCDOT FDR mix design procedure, the maximum indirect tensile strength test of both conditioned and unconditioned samples determined the optimal emulsion content. Fig. 3 shows the ITS results, and both sample sets had a maximum value of approximately 4%, which was chosen as the optimal emulsion content. In Fig. 3, the upper dashed line indicates the minimum required indirect tensile strength of unconditioned samples, and the lower dashed line indicates the minimum required indirect tensile strength of conditioned samples.
With the materials, water content (5.4%), and optimal emulsion content (4.0%) established, samples were fabricated using the Troxler 5850 and tested using the modified ASTM D7196. The mix design for the asphalt foam samples simply substituted the asphalt cement content of the asphalt emulsion, and did not include Portland Cement. While cement is frequently used in foamed mixtures, it was not included to reduce differences between the asphalt emulsion and asphalt foam samples. The asphalt emulsion residue averaged 70%, so the asphalt cement content for the asphalt foam mixtures was set at 2.8%.
Raveling test results
In order to quantify the ravelling test results of asphalt emulsion and asphalt foam Full Depth Reclamation (FDR), several comparisons were made. Figs. 4-6 show the effect of asphalt emulsion cured at ambient temperatures (approximately 25°C), the effect of asphalt emulsion cured in an oven at 40°C, and the effect of asphalt foam cured at ambient temperatures. Each of these figures show an exponential fit of the data out to 48 hours. Each data point represents three replicates. However, since this test would provide the highest benefit in quantifying the minimum curing time necessary before traffic is released, the data was also examined in more detail up to four hours. In addition to showing data up to 48 hours, Figs. 4-6 also show the data up to four hours with a linear fit of the data. Error bars showing the standard deviation with 95% confidence are also shown for the data up to four hours. Discussion on each of comparisons follows.
Fig. 4 shows the effect of ambient curing temperatures on asphalt emulsion FDR. It is interesting to note that the only sample to last the full ten minutes (600 seconds) in the raveling tester was the sample cured to 48 hours. This indicates that either the larger aggregate size of FDR mixtures, or the lower cohesive nature of mixture, prevented the samples from retaining their shape during the testing. Unfortunately, there was a weak linear correlation for curing times under four hours. This can be attributed to the high standard deviation of the test, which approached 25 seconds (corresponding to a coefficient of variation of over 60%) on multiple occasions. This builds the argument that the raveling test, even in the modified form, may not be appropriate for testing FDR mixtures. A separate study investigates different potential raveling test [
8]. After these relatively poor correlations, and low time in tester results, it was decided to explore a higher curing temperature in the hopes of increase cohesiveness of the samples and develop stronger trends at the shorting curing times. Fig. 5 shows the effect of 40°C curing temperatures on asphalt emulsion FDR.
Curing the FDR samples at 40°C increased the number of samples that lasted the entire 600 seconds in the raveling test, from just 0 hours to 12, 24, and 48 hours. This decreased the fit of the exponential function, but it may provide a better indication, or at least ranking potential between options, of when traffic can be returned. This is supported by looking at the data only up to four hours of sample curing time. In this figure, it is evident that there is a more linear relationship versus ambient temperature, and the standard deviation is lower across the majority of the curing times. However, there is still a weak linear correlation between the time in tester and sample curing time. The final comparison performed utilized asphalt foam samples cured at ambient temperatures, and can be seen in Fig. 6.
In Fig. 6, it is immediately apparent that the asphalt foam had a much lower time in tester than asphalt emulsion at ambient temperatures. In fact, the samples only lasted on average 43 seconds after 48 hours of curing. While FDR mixtures are not designed to withstand significant traffic loads, and generally are covered before significant raveling can occur, using the modified raveling test indicates that the asphalt foam has the potential to deteriorate extremely quickly under traffic loads. However, there is a decent correlation between time in tester and sample curing time, with extremely small standard deviations. Although the asphalt foam did not perform as well as the asphalt emulsion FDR mixtures, calibration to field performance could possibly generate useful quantification of the modified raveling test to actual field raveling performance.
Conclusions
Full-Depth Reclamation (FDR) is a proven rehabilitation technique to increase the structural capacity of a roadway using in-place material. This study explored the possibility of using ASTM D7196, “Raveling Test of Cold Mixed Bituminous Emulsion Samples,” in order to quantify when FDR jobs can be returned to traffic without significant raveling occurring between opening the road and placing a surface course. It was found that the ASTM procedure had to be modified for FDR, as samples would frequently disintegrate before the full fifteen minutes had past. Therefore, based simply on visual observation, the time before the ravelling head began to exert more lateral versus vertical force on the sample was used instead. This was a subjective observation and more research would enhance the decision. In addition, the maximum test time was set to ten minutes (600 seconds) for the modified procedure. Using the NCDOT FDR mix design, asphalt emulsion cured at ambient temperatures, asphalt emulsion cured at 40°C, and asphalt foam cured at ambient temperatures was explored. The following conclusions were found:
(1) At ambient temperatures, the asphalt emulsion FDR only lasted to 600 seconds after 48 hours of curing, and showed poor correlation up to four hours of curing
(2) At 40°C, the asphalt emulsion FDR lasted to 600 seconds after 12, 24, and 48 hours of curing, with decent correlation up to four hours of curing
(3) At ambient temperatures, the asphalt foam FDR did not last more than 43 seconds after 48 hours of curing, but showed decent correlation with small standard deviation up to four hours of curing
(4) It appears that with the larger aggregate size and lower cohesive nature (compared to Cold In-place Recycling), ASTM D7196 may not be appropriate for use in determining return to traffic characteristics of FDR.
Since this work is a preliminary lab study, it is recommended that more mixtures be evaluated and compared to actual field return to traffic performance. FDR performance is highly dependent on in-place materials, so a wider range of soil types would help understand performance. In addition, correlation with field would ideally create conversion factors for the ravelling test, so the times recorded in the lab could be directly related to actual times of return to traffic. It is believed that with a wider range of materials, and correlations to field performance, acceptable criteria could be developed for QC/QA purposes.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(463KB)
