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.
3 2 LITERATURE REVIEW 2.1 Cold Recycling Technologies Classification and Performance Cold recycling (CR) is a series of asphalt pavement rehabilitation methods that can be applied at lower cost, have lesser environmental impacts, and are often completed in less time than typical rehabilitation options. These benefits are attainable since CR reuses existing or stockpiled asphalt materials (often in the form of reclaimed asphalt pavement or RAP) without the application of heat. CR is not only effective in correcting many types of asphalt pavement distresses (such as raveling, rutting, cracking, and potholes), but it also conserves non-renewable resources and energy. Based on these merits and more recent success on high-traffic roadways, CR is gaining more attention in the United States. Traditionally, CR consists of two subcategories: cold in-place recycling (CIR) and cold central plant recycling (CCPR). CIR is an in-situ process in which the existing asphalt pavement is recycled using a specialized train of equipment that may include the ability to remove a portion of the existing pavement by milling, add an asphalt-based recycling agent and an active filler, crush or screen the materials, pave, and compact the resultant product in a single pass. CCPR is a process in which the asphalt-based recycling agent, active filler, and any corrective aggregate are incorporated at a central location using a mobile CR plant. The mobile plant could be a specifically designed CCPR plant or a CIR train minus the cold planning operation and set up in a stationary configuration. The materials produced from both the CIR and CCPR processes may consist of up to 100 percent RAP, a predetermined amount of recycling agent (foamed asphalt or emulsified asphalt) with or without a combination of active fillers (e.g., lime, cement, or other), and water to produce the CR mixture. Virgin aggregates may also be added to adjust RAP gradation and improve performance properties of CR mixtures. Active fillers are effective in enhancing mixture stiffness in both the short- and long-term and for reducing susceptibility to moisture damage (Wirtgen 2012, NASEM 2017). CIR treatment depths are generally from 3 to 4 inches but can be up to 5 inches if adequate compaction can be achieved (ARRA 2015). CCPR placement depths are similar to or slightly thicker than CIR and multiple lifts of CCPR can be placed for even thicker structures. CR technologies do not include full depth reclamation (FDR), which reclaims the full thickness of the asphalt layer and a predetermined portion of underlying unbound materials (aggregate base, subbase, subgrade). Similar to CR, asphalt-stabilized FDR also uses a stabilizing agent of foamed asphalt or emulsified asphalt, active fillers and water to produce recycled mixtures. However, FDR often uses a specialized reclaimer that mixes unbound material (aggregate base or soil) with RAP materials (Bowers et al. 2020). FDR treatment depths are typically up to 12 inches, which is greater than that of CR. FDR can completely eliminate asphalt pavement distresses resulting from a weak foundation and produce a strong base layer. Compared to CR, FDR is suitable to rehabilitate pavements with deeper deterioration such as bottom-up fatigue cracking, subgrade rutting, or other structural foundation issues. CR mixtures typically have a higher air void content as compared to hot and warm mix asphalt mixtures and require a surface course to prevent water from intruding into underlying layers and to improve the raveling properties of the pavement. Typically, an asphalt overlay is used for
4 pavement with high traffic volume, while a chip seal, slurry seal, or micro surfacing may be employed for pavements with a low traffic volume. Over the decades, CR has been found to be an economical rehabilitation technique. Kandhal and Koehler (1987) reported that the Pennsylvania Department of Transportation (DOT) completed around 90 CR projects by 1985. Through the case studies, they suggested that at least two applications of a surface treatment be placed over CR to avoid raveling and pothole distresses. Maurer et al. (2007) evaluated 29 different combinations of pavement structural and surface rehabilitation strategies for Nevada DOT and found that CIR with a double chip seal treatment was a cost-effective strategy that suffices to rehabilitate most roads with low and medium traffic volume. Recently, some state agencies, including the Virginia DOT, have proven that CR is also cost- effective for rehabilitation of heavy traffic volume roadways. In 2011, Virginia DOT reconstructed a 3.7-mile section on Interstate 81 (I-81) involving CIR, CCPR, and FDR techniques. Diefenderfer et al. (2012) evaluated three CR pavement structures through the project, including 4-inch asphalt overlay with 5-inch CIR (left lane), 4-inch asphalt overlay with 8-inch CCPR and 12-inch FDR (right lane), and 6-inch asphalt overlay with 6-inch CCPR and 12-inch FDR (right lane). After 5 years of service and approximately 10 million equivalent single axle loads (ESALs), both the CIR and CCPR pavements exhibited excellent ride, rutting and cracking performance (Diefenderfer et al. 2015). One year later, in 2012¸VDOT sponsored three test sections at the National Center for Asphalt Technology (NCAT) Pavement Test Track. These included a section with a 4-inch asphalt overlay over 5 inches of CCPR, a 6-inch asphalt overlay over 5 inches of CCPR, and a 4-inch asphalt overlay over 5 inches of CCPR over an 8- inch-thick cement-stabilized layer similar to FDR. After 28 million ESALs on the Pavement Test Track all three recycled sections showed negligible distress (Timm et al. 2020). In 2015, NCAT constructed four test sections on US 280 in Lee County, Alabama, including CCPR with emulsified asphalt and foamed asphalt, and CIR with emulsified asphalt and foamed asphalt. The CR pavement structures include a 1-inch-thick asphalt overlay, 3.5-inch-thick CIR or CCPR, and a 10-inch-thick existing asphalt treated base. As of 2020, the four test sections carried over 3.0 million ESALs with negligible cracking and rut depth less than 0.25-inch (NCAT 2018). In summary, CR is a leading technology in structural pavement rehabilitation that reuses existing resources and can reduce construction time. The field experience from existing literature shows that this rehabilitation technology is not only suitable for low and medium traffic volume roadways but is also applicable to high traffic volume pavements nationally. 2.2 Cold Recycling Mix Design A job mix formula is often developed in a laboratory setting for most CR projects. The job mix formula identifies the RAP gradation and defines the composition of recycling agent, active filler, water, and other additives. Kim and Lee (2006) and Wirtgen (2012) developed mix design methods for CR asphalt mixtures with foamed asphalt binder. These mix design methods not only define the requirements for RAP materials, foamed asphalt, and active fillers, but also provide the procedures to determine the optimum binder content. With the increase of field
5 experience in CR, Asphalt Recycling and Reclaiming Association (ARRA) (2015) developed new mix design methods for CR mixtures using foamed or emulsified asphalt. These methods better defined the material requirements and the performance criteria for CR mixtures. According to ARRA (2015), CR mix design consists of four key steps, including: 1) processing and sampling of RAP material; 2) selection of recycling agent, active fillers and other additives; 3) mixing, compaction, and curing; and 4) performance testing. 2.2.1 RAP Sampling and Processing For CIR, a representative sample of the pavement to be recycled is most often obtained by coring. These cores are crushed by a jaw crusher, laboratory milling machine, or other suitable method to determine the RAP gradation expected during recycling. Alternatively, the recycler can be used in the field to pulverize the to-be-recycled pavement so that an adequate sample can be collected. For CCPR, RAP is obtained either from an existing stockpile or sampled from the field like CIR. CCPR is unique in that the designer can have more control over the stockpile gradation by fractionating (ARRA 2015), but this is not always implemented in practice. To better ensure uniformity of the CR mixture for laboratory mix design, RAP particles larger than 1 inch may be removed by screening. Other sources in the literature point to fractionating RAP prior to batching (Eller and Olson 2009, Kim et al. 2006) but this is not usually done in practice. RAP gradation is thought to be an important factor affecting mixture performance and availability of fines has been shown to impact the performance of CR mixtures using foamed asphalt (Wirtgen 2012, Asphalt Academy 2009). AASHTO Standards MP 31 and MP 38 provide suggested gradation bands for fine, medium, and coarse RAP, which are shown in Table 2.1. The standards recommend to adjust gradation bands to local conditions and construction equipment, noting that it should be selected to mirror expected field gradations. Table 2.2 summarizes the RAP gradation bands for CR mixtures recommended by Wirtgen (2012) and Asphalt Academy (2009). Kim and Lee (2006) investigated the influence of RAP gradation on mechanical performance of CR mixtures. They found that the CR mixtures with a coarse RAP gradation have lower optimum foamed asphalt binder and lower indirect tensile strength than those with a fine RAP gradation. Table 2.1. RAP Gradation Requirements for Cold Recycled Asphalt Mixture (AASHTO Provisional Standards MP 31 and MP 38) Sieve Size Fine Gradation Medium Gradation Coarse Gradation Percent Passing 31.5 mm (1.25 in.) 100 100 100 25 mm (1 in.) 100 100 85-100 19 mm (3/4 in.) 95-100 85-96 75-92 4.75 mm (No. 4) 65-75 40-65 30-45 600 µm (No. 30) 15-35 4-14 1-7
6 Table 2.2. Other RAP Gradation Requirements for Cold Recycled Asphalt Mixture Sieve Size Wirtgen (2009) Asphalt Academy (2009) Percent Passing 37.5 mm 87-100 87-100 31.5 mm (1.25 in.) - - 25 mm (1 in.) 76-100 77-100 19 mm (3/4 in.) 65-100 66-99 12.5 mm (1/2 in.) 55-90 67-87 9.5 mm (3/8 in.) 48-80 49-74 4.75 mm (No. 4) 35-62 35-56 2.36 mm (No. 8) 25-47 25-42 1.18 mm (No. 16) 18-36 18-33 600 µm (No. 30) 13-28 12-27 75 µm (No. 200) 4-12 2-9 2.2.2 Material Selection It is necessary to select an appropriate recycling agent and active filler (as needed) in order to achieve the desired performance of the CR mixture. AASHTO MP 31 recommends two types of cationic emulsified asphalt (CSS-1 and CSS-1h) and three types of anionic emulsified asphalt (HFMS-2, HFMS-2h, and HFMS-2s) that can be used as recycling agents for emulsified asphalt CR. Additional criteria are provided for engineered emulsified asphalt. To ensure the desired performance of a foamed asphalt CR mixture, AASHTO PP 94 requires that foamed asphalt binder meet the minimum requirements of expansion ratio and half-life shown in Table 2.3. If quicklime or hydrated lime is used, it is required to meet the requirements of AASHTO M 216. Active fillers may be added to the mixture to improve early strength, reduce moisture susceptibility, and increase long-term stiffness (Schwartz et al. 2017). If Type I or II cement is used, it is required to meet the requirements of AASHTO M 85. ARRA (2015) notes that lime is typically added at 1.0-1.5% and cement between 0.25-1.0% by weight of RAP. A minimum 3:1 ratio is recommended for asphalt residue to active filler (ARRA 2015); however, the project team has observed minimum ratios of 2:1 and 2.5:1 successfully used in practice (Timm et al. 2020, Schwartz et al. 2017, Diefenderfer, et al. 2015). Note that the current AASHTO provisional standards (MP 31, MP 38, PP 86 and PP 94) do not specify any other additives (e.g., asphalt rejuvenator, fly ash, baghouse fines) for use in CR mixtures. Table 2.3. Minimum Requirements of Foamed Asphalt Properties RAP Temperature Foamed Asphalt Properties Expansion Ratio (times) Half-life (seconds) 10-15°C (50-59°F) 10 8 >15°C (59°F) 8 8 2.2.3 Mixture Preparation Mixture preparation involves three steps: mixing, compaction and curing. Mixing quality is dependent on mixing equipment, mixing time and temperature. There are three types of mixing procedures commonly used in the laboratory including blender-based, bucket-based, and the use of twin-shaft pugmill mixers. Table 2.4 shows the mixing methods adopted in the existing literature.
7 Table 2.4. Existing Mix Design Methods for Cold Recycled Asphalt Mixture Reference Mixing Equipment Initial Mixing Time (seconds) Secondary Mixing Time (seconds) Mixing Temperature Eller and Olson (2009) Blender 40-60 60 Ambient Asphalt Academy (2009) Pugmill - 20-30 Ambient Wirtgen (2012) Pugmill >10 30 Ambient Kuna et al. (2014) Pugmill - 60 20±2 °C ARRA (2016) Bucket/Pugmill - 60 25±5 °C Ma (2018) Pugmill 15 60 Ambient AASHTO PP 86 Bucket/Pugmill - 60 20-25 °C AASHTO PP 94 Bucket/Pugmill - 60 25±2 °C Table 2.5 summarizes the compaction methods for CR mixtures adopted by current studies: modified proctor compaction, vibratory compaction, Marshall hammer compaction, and gyratory compaction. The modified proctor and vibratory methods are typically used for unbound material, while Marshall hammer and Superpave gyratory methods are commonly used for asphalt materials. Diefenderfer et al. (2012) found that CR specimens compacted using the Marshall hammer for 75 blows per side had comparable density to the field in-place density measured by nuclear gauge for one project. Cross (2003) investigated seven CR asphalt emulsion projects and recommended that CR asphalt emulsion specimens be fabricated by Superpave gyratory compactor to 30 gyrations immediately after mixing, and 35 gyrations if emulsion breaks. Ma (2018) verified that this compactive effort is also suitable for CR foaming specimens. AASHTO provisional standards PP 86 and PP 94 specify that 4-inch diameter and 2.5-inch-tall CR specimens should be compacted by Marshall hammer to 75 blows per side, or gyratory compactor to 30 gyrations, and 6-inch diameter and 3.75-inch-tall specimen should be compacted by gyratory compactor to 30 gyrations. Table 2.5. Existing Compaction Methods for Cold Recycled Asphalt Mixture Reference Compaction Method Compactive Effort Lee and Kim (2003) Gyratory 25 gyrations Kim et al. (2007), Cox and Howard (2016), ARRA (2016), Ma (2018) Gyratory 30 gyrations Cross (2003) Gyratory 30-35 gyrations Kim et al. (2007), Wirtgen (2012), ARRA (2016) Marshall 75 blows per each side Wirtgen (2012) Modified Proctor 55 blows per each layer Asphalt Academy (2009) Vibratory 10-25 seconds per each layer AASHTO PP 86 Marshall 75 blows per each side Gyratory 30 gyrations AASHTO PP 94 Marshall 75 blows per each side Gyratory 30 gyrations Laboratory curing aims to simulate field curing conditions that are important for CR asphalt mixtures to achieve adequate strength. Bowering (1970) proposed to cure CR foamed mixtures at 60°C (140°F) for 3-days. Maccarrone (1995) found that CR foamed mixtures cured at 60°C (140°F) for 3-days had similar resilient moduli to those cured in the field for 12-months. Ruckel et al. (1980) recommended a protocol of 1-day at 40°C (104°F) to simulate intermediate field curing (7-14 days of field curing) and 3-days at 40°C (104°F) to simulate long-term curing (30-
8 200 days of field curing). Ma (2018) confirmed that curing CR foamed mixtures including cement or lime as active fillers at 40°C (104°F) for 3-days achieved comparable indirect tensile strength to those cured in the field between 30 and 100 days. Kim et al. (2006) compared two laboratory curing protocols for CIR foamed mixtures, including 3-days at 40°C (104°F) and 2- days at 60°C (140°F). They found that CIR foamed specimens cured at 60°C (140°F) for 2-days showed significantly higher indirect tensile strength than those cured at 40°C (104°F) for 3-days. The authors recommended adopting the curing protocol of 3-days at 40°C (104°F) because the conditioning temperature was closer to field pavement temperatures. Yeung and Braham (2018) also compared these two laboratory curing protocols for CR emulsified mixtures. They found that CR mixtures cured at 60°C (140°F) for 2-days did not follow the traditional parabolic trend in the curve of Marshall stability versus emulsified asphalt content, whereas those cured at 40°C (104°F) for 3-days did. Thus, the authors also recommended to cure CR emulsified mixtures at 40± 1°C (104 ± 2°F) for 3-days prior to conducting any performance testing. According to AASHTO provisional standard PP 86, CR mixtures should be cured at 60 ± 1°C (140 ± 2°F) to reach a constant mass but should not be cured for more than 48 h or less than 16 h. But AASHTO provisional standard PP 94 requires that CR foamed asphalt should be cured at 40 ± 1°C (104 ± 2°F) for 3-days. Table 2.6 summarizes the laboratory curing protocols for CR asphalt mixtures found in the literature. Table 2.6. Summary of Laboratory Curing Protocols for Cold Recycled Asphalt Mixtures References Recycling Agent Type Curing Methods Bowering (1970), Maccarrone (1995) Foam 3 days at 60°C Ruckel et al. (1980) Foam 1 day at 40°C for intermediate curing, 3 days at 40°C for long-term curing Kim et al. (2006), Ma (2018) Foam 3 days at 40°C Yeung and Braham (2018) Emulsion 3 days at 40°C AASHTO PP 86 (date) Emulsion 16-48 hours at 60°C AASHTO PP 94 (date) Foam 3 days at 40°C 2.2.4 Mixture Performance-Based Testing In practice, performance-based testing is conducted using primarily the indirect tensile strength for foamed asphalt mixtures and the Marshall stability test for emulsified mixtures. Certain agencies in the western US also include a Hveem cohesiometer measurement for emulsified mixtures and some agencies use the ITS test for both foam and emulsified asphalt mixtures. These tests are most often used during mixture design to determine the optimum recycling agent content or need for active fillers. During construction, these same tests may be used for quality assurance purposes. Research testing in the US most often assumes that the CR mixtures have asphalt-like behavior and thus dynamic modulus and resilient modulus may be conducted for stiffness assessments (Schwartz et al. 2017, Diefenderfer et al. 2016, Diefenderfer and Link, 2014) while flow number testing may be used to measure rut susceptibility (Schwartz et al. 2017, Apeagyei and Diefenderfer 2013). In addition, balanced mixture design approaches have been considered for CR materials (Diefenderfer and Boz 2019, Cox and Howard 2016) to determine the optimum combination of recycling agent and active filler for adequate rutting and cracking resistance. While performance-based tests have been studied, research is still considered to be limited and thus far they have not been used in mix design.
9 Kim et al. (2006) utilized two performance tests, the indirect tensile strength test and Marshall stability test, to determine the optimum foamed asphalt content in CR mixtures. They defined that the optimum foamed asphalt binder content corresponded to the highest Marshall stability or indirect tensile strength of the CR foamed mixtures. They also found that the indirect tensile strength showed a clearer peak than the Marshall stability when varying the foamed asphalt content. Similarly, Yeung and Braham (2018) used the Marshall stability test to determine the optimum emulsified asphalt content corresponding to the peak of Marshall stability. AASHTO provisional standards MP 31 and MP 38 specify the Marshall stability and indirect tensile strength tests to determine the optimum recycling agent (emulsified or foamed asphalt) content. The detailed requirements of mixture performance testing are presented in Table 2.7. AASHTO provisional standards PP 86 and PP 94 define the minimum asphalt content satisfying the mixture performance criteria as the optimum content. Table 2.7. Performance Testing Requirements for Cold Recycling Mix Design Mixture Type Test Test Standard Test Parameter Criteria Foam Indirect Tensile Strength AASHTO T 283 Dry Tensile Strength Min. 45 psi (310 kPa) Tensile Strength Ratio (TSR) Min. 0.70 (cement) Min. 0.60 (hydrated lime or no additive) Emulsion Marshall Stability Test AASHTO T 245 Marshall Stability Min. 1,250 lbs (5,560 N) Retained Marshall Stability Min. 0.70 Raveling Test ASTM D 7196 (Cured in accordance to AASHTO PP 86) Raveling Loss Max. 7.0% 2.3 Impact of Active Fillers and Environmental Conditions on Mixture Performance There are a variety of mineral fillers used in CR mixtures, including but not limited to hydrated lime, Portland cement, fly ash, and baghouse fines. Existing studies found that these additives are effective for improving tensile strength and rutting resistance as well as reducing moisture susceptibility. For example, Nosetti et al. (2016) noted that the addition of 0.5% of cement or hydrated lime (by dry weight of RAP) significantly improved the wet indirect tensile strength and tensile strength ratio (TSR) of CR asphalt mixtures, but increasing the additive content from 0.5% to 2.0% showed limited improvements. Thanaya et al. (2009) reported that the CR mixtures had much higher early-age and long-term strengths when adding 1%-2% cement. Cross and Young (1997) found that the Type C fly ash reduced the moisture susceptibility of CR asphalt mixtures, but also decreased the resistance to fatigue and thermal cracking. Ma et al. (2018) examined the effect of active fillers on indirect tensile strength of CR foamed asphalt mixtures. The authors found that cement provided the greatest improvement in both dry and wet indirect tensile strength, while fly ash only enhanced the wet indirect tensile strength and baghouse fines only increased the dry indirect tensile strength. The authors also confirmed that higher additive content (>2%) did not necessarily improve the indirect tensile strength of CR mixtures. Niazi and Jalili (2009) evaluated the rutting resistance of CR mixtures using a dry wheel tracking test. They
10 concluded that the rutting resistance of CR mixtures was dependent on the type and dosage of additives, in particular on the dosage of cement. The CR mixtures with a dosage of cement less than 1.5% generally have less rutting resistance than a control hot mix asphalt (HMA). Ma (2018) summarized the dosage requirements of active filler in CR mixtures by different organizations, which are shown in Table 2.8. To prevent brittle behavior of the CR mixture, AASHTO provisional standard PP 86 specifies that the cement content should be not greater than one-third of the emulsified asphalt residue content and the lime content should be not greater than 1.5%, while AASHTO provisional standard PP 94 requires that the ratio of cement content to foamed asphalt content be a maximum of 1:2.5. Table 2.8. Summary of Active Filler Requirements by Different Organizations Reference Active Filler Type Cement Hydrated Lime Asphalt Academy (2009) â¤1% and less than asphalt residue content â¤1.5% Wirtgen (2012) â¤1% â¤1.5% AASHTO PP 86 Less than one-third of asphalt residue content â¤1.5% AASHTO PP 94 Less than two-fifths of foamed asphalt content Typically 1-1.5% Gu et al. (2019) investigated the impact of climatic condition on the modeled long-term performance of CR pavements using the Pavement ME Design software. Note that the performance models in the Pavement ME Design software were not calibrated for CR mixtures or CR pavement structures. They selected four locations, including Montgomery, AL, Los Angeles, CA, Bozeman, MT, and Minneapolis, MN, to represent the four climatic regions defined in the long-term pavement performance program. They found that the MT section had the lowest rut depth, while the AL section had the highest rut depth. This was because that asphalt surface mixtures and CR mixtures in freeze regions were much stiffer than those in non- freeze regions. They also pointed out that the climatic condition significantly influenced the moisture damage and freeze-thaw effects of CR asphalt pavements. 2.4 Construction Practices for Cold Recycling ARRA (2015) comprehensively documented the recommended construction practices for CR, including quality testing, equipment requirements, and milling, paving, and compaction processes. To ensure expected pavement performance, there are five critical topics needed in a quality assurance plan, including milling of existing pavement, addition of recycling agent, addition of compaction water, and compaction and curing. Table 2.9 summarizes the controlled process variable, process control test options, and challenges in each process step.
11 Table 2.9. Key Process Topics for a Quality Assurance Plan (ARRA 2015) Process Topic Controlled Process Variable Process control test options Potential Challenges Mill existing pavement Uniform cutting head / teeth Grade control Pavement retainer bar Milling speed Monitor milling speed Check cutting depth Visually check RAP uniformity Monitor RAP gradation Most tests are visual rather than quantitative Add recycling agent Equipment temp booster Binder flowrate Foam water flowrate Foam proportion meter Distribution bar pressure Agent metering system Spray bar condition Uniform production speed Calibrate binder and water flow rate Calibrate proportion metering system Measure foam expansion & half- life Calibrate stabilizing agent metering system If not measured directly on equipment, can be difficult to verify mass of materials coming into the project Add compaction water Amount added in emulsion Amount required to achieve compaction Field adjust as conditions warrant Uniform production speed Visual inspection of material Continuous moisture sensor Compaction moisture need changes during the day Compaction Proper roller size and type for lift thickness Proper roller speed and passes Field test for density Field test for moisture Field test for strength gain Field measure final layer thickness Ensuring uniform application of compaction; Using nuclear density gauge for moisture content; Field-based strength test that can be related to design Curing Changes in curing time with respect to environmental conditions Check moisture content Field test for density Field test for moisture Field test for strength Using nuclear density gauge for moisture content; Field-based strength test that can be related to design Milling is a key step to achieve a good performing rehabilitated layer. The mix design is developed based on an assumed gradation that the contractor will try to achieve during production. Also, milling prior to recycling establishes the reference that is used for paving the recycled layer and is used to establish a smooth recycled layer surface. In addition, milling to the proper depth ensures that the recycling operation can get beneath the deteriorated level (especially true for CIR projects where cracking in the upper part of the pavement is to be removed). The amount of recycling agent is critical to the mechanical performance of the CR mixture. During the construction process, the recycling equipment must add the correct amount of recycling agent into the material. Both foam and emulsion-based equipment must have a properly functioning distribution system to meter the recycling agent at the prescribed amount. Foam recycling agents are particularly dependent on a good distribution system because the foam must properly disperse to achieve bonding of the recycled mixture matrix. Too little recycling agent (whether foamed or emulsified asphalt) can lead to an unstable layer that is
12 susceptible to raveling and have low strength properties while too much recycling agent can lead to a layer that is prone to rutting under traffic. Similar to soils applications, CR technologies depend on achieving the optimum water content to achieve the maximum density. The amount of water must be adjusted during production to account for the changes in moisture content of the parent RAP (whether stockpiled or milled from the project), length of haul, changes in gradation, time between mixing and compaction, ambient temperature, wind speed, humidity, amount of sunlight, etc. Properly compacted CR layers provide a structural layer that is capable of supporting traffic soon after recycling as well as providing the intended engineering properties after placement of the wearing surface. Generally, a higher degree of compaction results in better performance of a CR mixture. Bowers et al. (2020) reviewed 45 CIR and 14 CCPR specifications gathered from North America and all but three specifications contained a density requirement. Examples of density requirements are shown in Table 2.10. While the compaction specification provides a target for compaction effort during construction, there is limited literature that correlates compacted density to material stiffness. Table 2.10. Selected Specification Requirements for Density of Cold Recycled Layers Agency Requirement Iowa > 94% of lab density (high traffic)> 92% of lab density (low traffic) Kansas > 97% of control strip density Texas Establish rolling pattern Colorado > 100% of lab density California 95-105% of control strip density Rhode Island not specified Virginia > 98% of lab density Minnesota not specified Maine > 98% of control strip density A nuclear density gauge is commonly used in the field to measure the compacted density and moisture of the CR layer. There are typically two modes to operate a nuclear density gauge to measure density: direct transmission and backscatter mode. The nuclear density gauge source rod emits low levels of radiation that then interacts with electrons in the material and loses energy. The in-place density of material is calculated by correlating the material density to the lost energy of radiation. The water mass per unit volume is also determined by the thermalizing or slowing of fast neutrons by hydrogen. The neutron source and the thermal neutron detector are both located at the surface of the material being tested in the backscatter mode while a probe is inserted into a specially prepared hole in the direct transmission mode. The high-energy neutrons are slowed when they collide with hydrogen atoms, and the detector then counts the slowed neutrons. This count is proportional to the materialâs water content, based on the assumption that the hydrogen in this water is responsible for almost all the hydrogen found in the layer. When the
13 nuclear gauge is used to determine the moisture content of the recycled or reclaimed mixes the measured moisture content must be adjusted for the asphalt content in the material. For example, a calibration procedure has been used by the Iowa DOT to determine the actual moisture content of recycled mixes by applying a correction factor to the measured results from the nuclear gauge. In addition to the nuclear density gauge method, the electromagnetic method (ASTM D7830), the rubber balloon method (ASTM D2167), and ground penetrating radar can also be used to measure the mat density during the compaction process (Diefenderfer et al. 2012, Minotra et al. 2015, and Nielsen et al. 2007) although some of these processes may also require a calibration. CR mixtures undergo a curing process in which the strength / stiffness of the material increases as the internal moisture content decreases over time. As defined by ARRA (2015), curing is divided into three stages, initial, intermediate, and final. Initial curing is relatively short (usually 0.5-1 hour) and allows the stabilized mix to gain sufficient cohesion. Intermediate curing takes more time and depends on the moisture content of the CR mixture, the amount and type of recycling agent, presence of active fillers, and the ambient climate conditions. Intermediate curing is required to allow CR mixture to build up sufficient strength and/or to allow sufficient moisture to escape to successfully apply the wearing course. Final curing is the time it takes CR mixture to reach its ultimate strength and can be a very long period (several months or years). Final curing takes place after the wearing course has been applied and is dependent on the amount and type of recycling agent, presence of active fillers, and ambient conditions. Although the curing process is known to occur, a quantified relationship between curing (as measured by the moisture content) and strength / stiffness is not well understood. From the perspective of the construction process, determining the moisture content is often used as a guide for when to place the surface layer. Table 2.11 summarizes selected specifications and their criteria for considering a layer to be cured. Typically, this is assessed either in terms of the moisture content of the layer or by a number of days since the layer was placed. Table 2.11. Selected Specifications for Assessing Curing Time of Cold Recycled Mixtures Agency Curing Criteria Maximum Moisture Content (%) Minimum Curing Days Indiana 2.5 NA1 Maine NA 4 Minnesota 1.5 NA New York NA 7 Pennsylvania 2 7 Virginia 50% OWT2 NA Iowa 2 NA Kansas 2 NA Texas 2 NA Colorado 1 NA California 2 10 Note: 1 NA = not specified; 2 OWT = optimum mixing water content
14 Although seldom measured directly in the field, the in-place stiffness properties (or other properties that can be correlated to stiffness) of CR mixtures may be assessed using the falling weight deflectometer (FWD), lightweight deflectometer (LWD), dynamic cone penetrometer (DCP), Geogauge, or portable seismic pavement analyzer (PSPA). Through the specification review, it was found that Minnesota and Nebraska use the LWD to assess deflection rather than density during construction. 2.5 Structural Design of Cold Recycled Asphalt Pavements The purpose of structural design is to determine appropriate layer thickness and composition of pavement materials to carry the desired loading for the intended service life. Currently, there are two structural design methods most often used for CR asphalt pavements, including the AASHTO 1993 design method and the Pavement ME Design method. In the AASHTO 1993 design method, the key component is the structural layer coefficient of pavement material. Sebaaly et al. (2004), Diefenderfer et al. (2015) and Diaz-Sanchez et al. (2017) determined the layer coefficients of cold in-place and cold central plant recycled asphalt mixtures relying on an empirical relationship between layer coefficient and back-calculated resilient moduli. This relationship was originally developed for hot mix asphalt (Huang 2004), but whether it is suitable for CR asphalt mixtures is still not clear. Pinto and Buss (2020) investigated about 100 CIR projects in Iowa and established a correlation between the layer coefficient of CIR mixture and the international roughness index of pavement. Table 2.12 shows examples of layer coefficients of CR asphalt mixtures documented in the existing literature. Table 2.12. Layer Coefficients of Cold Recycled Asphalt Mixtures Reference Material Type Traffic Level Layer Coefficient (1/inch) Diefenderfer et al. (2015) CCPR-foam High 0.37-0.44CIR-foam High 0.39 Diaz-Sanchez et al. (2017) CCPR-foam High 0.36-0.39 Sebaaly et al. (2004) CIR-emulsion Low/Medium 0.26 Pinto and Buss (2020) CIR NA 0.26-0.32 Recently, as more highway agencies are adopting the Mechanistic-Empirical Pavement Design Guide, the knowledge of changing material behavior with temperature has become more important. In the current Pavement ME Design program, the CR asphalt mixture can be considered as a bound base material, which means that users only need to assign a constant resilient modulus. However, this assumption contradicts the fact that the CR mixture exhibits thermo-viscoelastic characteristics that make it act more like an asphalt layer, having lower stiffness at high temperatures and greater stiffness at lower temperatures. Kim et al. (2009), Khosravifar et al. (2015), Diefenderfer et al. (2016), and Lin et al. (2017) conducted dynamic modulus testing for various CR mixtures. They found that the stiffness of the CR mixtures exhibited temperature and frequency dependency, although to a lesser degree than for HMA. Schwartz et al. (2017) considered the viscoelastic characteristic of bituminous stabilized recycled materials, and comprehensively quantified the dynamic modulus and repeated load permanent deformation properties of FDR, CIR, and CCPR field cores for use in Pavement ME Design. They did not perform any fatigue characterization for CR mixtures, since bottom-up fatigue cracking was found to be an insignificant distress mode for CR asphalt pavement. Gu et al. (2019) confirmed that CR asphalt pavement in heavy traffic application had negligible bottom-up fatigue cracking. They also found that the Pavement ME Design over predicted the rut depth of
15 CR asphalt mixture. These findings, among others, strongly suggest that Pavement ME Design distress models need to be calibrated for asphalt-stabilized recycled mixtures. 2.6 Economic and Sustainability Assessments of Cold Recycled Asphalt Pavements Life-cycle cost analysis (LCCA) is used to evaluate the overall long-term costs of investment alternatives. Considering anticipated costs over the life of different pavement alternatives, LCCA is a process for identifying the best long-term value among pavement alternatives. Life-cycle assessment (LCA) is a tool that quantifies the environmental performance of pavement system over the entire life cycle. Cross et al. (2011) compared the environmental impacts of CIR with that of a conventional rehabilitation technique (3-inch mill and overlay) using Pavement Life- Cycle Assessment Tool for Environmental and Economic Effects (PaLATE). They found that CIR effectively reduced energy consumption and greenhouse gas emissions at the stages of material production and transportation. Santos et al. (2015) developed a comprehensive life- cycle inventory database for the activities associated with CIR and CCPR. They found that CR technologies reduced the overall life-cycle energy consumption by 30% when compared to the mill and overlay method. Pakes et al. (2018) conducted an LCA using the PaLATE tool for nine CIR projects in Wisconsin, and concluded that CIR reduced energy use by 23%, water use by 37%, and virgin aggregate use by 37% on average. Gu et al. (2019) conducted a case study of LCCA and LCA for four different CR technologies, including CIR-foam, CIR-emulsion, CCPR-foam, and CCPR-emulsion. They demonstrated that all four CR technologies reduced net present costs by 12-32%, energy consumption by 56-67% and greenhouse gas emissions by 39-46%. They reported that the most cost-effective CR technology was CCPR-foam, and the most energy-saving technology was CIR-foam. Zhou et al. (2020) conducted LCA and LCCA on two CR projects on I-81 and I-64 in Virginia and found that CR technologies saved approximately $5 million for a 7-mile segment in material production and construction costs, and about 40% energy consumption due to the reduction of virgin materials usage. Overall, the existing studies demonstrated that CR technologies are more cost-effective and sustainable than other non-recycling pavement rehabilitation techniques.
