Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling (2021)

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.

99   Several agencies indicated on the survey that the lack of project selection criteria was a barrier to increased use of cold recycling. The first case example was selected to highlight the Indiana DOT process for identifying the types of projects on which CCPR and CIR processes can provide an alternative to traditional maintenance and rehabilitation practices. The survey also showed that CCPR was used by only a limited number of agencies, so the next two case examples were selected to document the implementation and running of a CCPR program. Case example two describes the Minnesota DOT’s initial experience with constructing CCPR test sections. Case example three documents the Maine DOT experience with operating an agency-owned CCPR plant. Case example four describes the new Utah DOT specification for cold recycling mix designs and criteria for when the cold recycling layer can be surfaced. The mix design method incorpo- rates changing environmental temperatures into determining acceptable adjustments to recy- cling agent quantities in the field. The criteria for when the cold recycled layer can be surfaced uses a stiffness measurement rather than either moisture content or a set time. Case example five highlights Caltrans’s experience with applying smoothness specifications to cold recycled mix projects. Case example six provides a brief summary of the Federal Lands Highway CIR experience gained over more than four decades. Case Example One – Indiana DOT Approach to Project Selection The Indiana DOT has been experimenting with cold recycling for a number of years. How- ever, cold recycling was typically considered as an alternative close to the time the project was about to go to bid (N. Awwad, personal communication, July 10, 2020). This meant cold recycling was not included in the early stages of the project development. The agency’s experience led to the development of a decision flow chart to identify when cold recycling can provide the best way to improve the roadway with the least cost. The Indiana Department of Transportation Design Manual, Chapter 602 (Indiana DOT 2020), provides a good example of how to use the pavement type (i.e., composite, full-depth asphalt) and pavement distresses to identify potential cold recycling projects. Indiana roads, both owned by the agency and local municipalities, historically used PCC pavements, so the presence of these old pavements needs to be considered at the start of project selection (Figure 49). C H A P T E R 4 Case Examples Indiana DOT

100 Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling Asphalt pavements with less than 10% full-depth patching and no evidence that an underlying PCC pavement needs repair or recycling are good candidates for CIR with an overlay. When the PCC pavement needs remediation, then the existing asphalt pavement needs to be milled off so the underlying structure can be repaired. In those cases, the milled RAP can be stockpiled and used to produce a CCPR base layer and topped with an overlay (Figure 50). More than 10% full-depth patching, a subgrade CBR of more than 6, and an existing asphalt pavement 5 in. (125 mm) or thicker can be improved by using an emulsion FDR to recycle the asphalt and any aggregate base if present (Figure 51). Since the subgrade provides acceptable support, it should not be disturbed during FDR. When the existing pavement is more than 10 in. (250 mm) thick, pre-milling the surface may be needed. When more than 3 in. (75 mm) needs to be pre-milled, the RAP can be stockpiled and used for a CCPR base layer once the FDR is completed. CIR can be used when the existing pavement is less than 5 in. (125 mm) thick. More than 10% full-depth patching and a subgrade CBR of less than 6 are an indication of poor subgrade support that needs to be addressed (Figure 52). In these cases, cement FDR that includes 50% subgrade material is used to restore support. Either a CCPR base layer with an overlay or multiple lift overlays can be used for higher traffic volume roadways. When the existing asphalt pavement is over 10 in. (250 mm) thick, pre-milling may be needed before proceeding with the FDR. Case Example Two – Minnesota DOT Initial Experience with CCPR The first experience of the Minnesota DOT with CCPR was the construction of four test cells (133, 135, 233, 235) on the MnROAD low-volume test track in Otsego, Minnesota, in the fall of 2017. All four test sections consisted of 3.5  in. (89 mm) of CCPR, after compaction, Source: Indiana DOT 2020, Chapter 602, Figure 602-1a. Figure 49. Initial part of Indiana DOT decision tree. Minnesota DOT

Case Examples 101   Source: Indiana DOT, Chapter 602, Figure 602-1a. Figure 50. Decision tree for when an overlay is considered appropriate. Source: Indiana DOT 2020, Chapter 602, Figure 602-1a. Figure 51. Decision tree for when full depth repair is indicated and CBR is greater than 6.

102 Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling which were placed on 12-in. (305-mm) Class 6 aggregate base over a clay/loam subgrade (Tompkins 2019). Since this was the Minnesota DOT’s first project with CCPR, the mix design and testing were completed as a joint effort between the agency and American Engineering Testing, Inc. Additional CCPR and CIR cells were constructed in 2019–2020 in collaboration with the National Center for Asphalt Technology and the Virginia Transportation Research Council as part of an ongoing NCHRP study. The CCPR mixes for the 2017 MnROAD test cells were produced with the pugmill portion of a CIR recycling train. The RAP was obtained from a single source, and the only gradation preparation was to ensure the maximum particle size requirement was met. The in-place density was defined by establishing the rolling pattern. The Minnesota DOT is continuously monitoring the performance because this is an ongoing research study. Since the construction of short test sections (700 ft) limits experience-based learning, David Rettner of American Engineering Testing, Inc., was asked to outline, based on his previous experience, key points for obtaining density on a typical cold recycled mix project (personal communication, July 10, 2020). Mr. Rettner noted that using a test strip to define the rolling pattern may be problematic. Roller operators tend to increase rolling speed to keep up with paving operations, which can produce ripples in the cold recycled mix surface. For example, he noted, a roller speed of 7 mph (11 km/h) can produce poor compaction and premature rutting of cold recycled mixes. Having more rollers on the job allows the operators to maintain a slower speed while keeping up with the paving operations. Rather than establishing a rolling pattern and specifying the number and types of rollers, a target density needs to be defined; then the contractor can be allowed to do what is necessary to meet the density requirements. Source: Indiana DOT 2020, Chapter 602, Figure 602-1a. 5 to 10 in What is pre-mill depth? < 5 in What is the existing pavement thickness? > 10 in Cement FDR + 50% of subgrade, CCPR (if suggested) Multi-lift overlay Cement FDR + 50% of subgrade < 3 in Proceed with pre-mill > 3 in Use CCPR on pre-milled portion Pavement Condition and Pavement Distresses Indicate Full Depth Repair and CBR is less than 6 Distress Trigger: Bottom up cracking, distresses requiring full depth patching. Figure 52. Further options on decision tree for when full-depth repair is indicated and CBR is less than 6.

Case Examples 103   Mr. Rettner also noted that the density on a typical cold recycled mix project is dependent on the weather throughout the day of construction. Colder pavement temperatures result in lower achievable densities when using a set rolling pattern. To compensate, more water is added to improve the compactability of the mix. The original Road Science design was based on adjusting liquid quantities according to the surface area of the RAP. The minus No. 4 (4.75-mm) RAP is air-dried and sieved down to No. 30 (0.60-mm) for a quick calculation of surface area. Emulsion and foamed asphalt recycling agents were used for test sections, and both recy- cling agents produced visually similar mixes. When asked if foamed asphalt cold recycled mixes tended to be friable and not well coated, Mr. Rettner noted that when the original asphalt pave- ment used a finer gradation with a hard aggregate, milling tended to break up the RAP by separating particles rather than fracturing the aggregate particles. The result is milled RAP that looks like well-coated aggregate with either emulsion or foamed asphalt recycling agents. When the original pavement used a softer aggregate, such as limestone, and coarser gradations, the milling operations tended to fracture the aggregate. In this case, the cold mix can look more like a partially coated aggregate material. Foamed asphalt cold recycled mixes need to cure for at least 3 days before testing to avoid being damaged (i.e., too friable). Both foamed asphalt and emulsion cold recycled mixes can produce similar test results when cured before testing. Case Example Three – Maine DOT Operation of CCPR Mr. Brian Luce (personal communication, July 29, 2020) provided a summary of the Maine DOT cold recycling program. Mr. Luce noted that by the late 1990s, the requirements for high- quality aggregates in conventional asphalt mixes restricted the use of local aggregates in some locations. The high cost of trucking in aggregates was becoming cost-prohibitive, and in 1997 the Maine DOT started to explore using excess RAP to produce cold recycled mixes. The premise was presented to the administration as a way of essentially mining existing pavements as sources of high-quality materials that had already met specification requirements. But as in most areas of the country, the difficulty was finding cold recycling contractors. In 2004, the agency purchased a portable twin shaft pugmill with three feeder bins (primary, secondary, fines). Initially, the expectation was that the cement could be added with the fines bin; but metering the cement accurately was too difficult, and a 30-ton cement silo was added to the portable plant. The cement is used to reduce the curing time and to improve the cold recycled mix. When asked if the agency had any push-back about competing with local contractors, Mr. Luce indicated the CCPR operation only produces the mix. Local trucking firms are used to move materials. Local paving contractors are hired to place the cold recycled mix and to provide the overlays. RAP stockpiles are developed by offering contractors on capital improve- ment projects a disposal site for excess RAP close to their projects. The contractor saves money and time while the agency builds stockpiles of high-quality RAP. The Maine DOT uses a rolling 5-year capital improvement planning process with project funding identified 3 years before building the project. The RAP disposal sites are tentatively identified as part of the 5-year project planning. RAP stockpiles are built over several months and processed just before the production of the cold recycled mix starts. Screening equipment controls the maximum particle size and diverts oversized material for reprocessing. A stacker conveyor moves the RAP to the pugmill cold feed bins. The maximum RAP particle size is eventually reduced to ¾ in. (19 mm) to improve the mix characteristics and uniformity. Recent improvements in miller cutting heads (i.e., micromillers) are making it easier to meet the smaller maximum size as well as producing Maine DOT

104 Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling RAP with more consistent gradations. The finer gradation also produces mixes that are less prone to segregation. Water storage is a tank on a trailer that can be filled up at a local water source, such as a fire hydrant. The optimum water content is determined during the mix design phase of the project, but it is actually used to set the maximum allowable amount of water. The water content of the RAP stockpile is frequently monitored during production as it changes throughout the day, at different depths in the stockpile, and with environmental changes such as humidity and rainfall. The RAP moisture content is typically between 2% and 4%. When it is too low, the mix is difficult to work and the emulsion has difficulty coating the RAP particles. When the moisture content gets too high, the mix is tender, slow to cure, and will likely have pavement performance problems. The emulsion tanker acts as the recycling agent tank. The finer RAP gradation has increased the optimum emulsion content to about 2.6% to 3% (residual asphalt content of 1.8% to 2%). The CCPR mix, referred to as “pugmill mix,” is currently designed using 50 gyrations, but this number will likely decrease so the mix design voids more closely replicate those seen in the field cores. The plant is operated by a four-person crew: two loader operators, a plant operator, and one person acting as a troubleshooter and choreographer. Additional day labor is added as needed. The Maine DOT CCPR operation produces between 60,000 and 100,000 tons of cold mix per year. The field target density is set as a percentage of the test strip maximum specific gravity. The final wearing surface is placed about 14 days after the cold recycled mix is placed. Cores can be taken to determine if the overlay can be placed sooner, sometimes as soon as 3 to 4 days after placing the cold recycled mix. When asked if the cores hold together when taken shortly after placement, Mr. Luce noted that if the core held together during coring, the mix had likely cured sufficiently, which is demonstrated by meeting strength requirements. The fine cold recycled mix gradation allows the mix to be used as a leveling course, a base layer, or in multilift designs. The typical lift thickness is 3 in. (75 mm). Additional lifts can be placed once the first lift cures for about 2 to 3 days. As many as three lifts of the cold recycled mix have been used on projects. Regardless of the number of lifts, cold recycled mixes have been typically used on lower traffic volume roadways, although a recent project was completed using a cold recycled mix on 12 miles of interstate highway. None of the projects to date have involved ride quality specifications. But preliminary evalu- ations show that a cold recycled mix layer will reduce an existing International Roughness Index (IRI) of about 350 in./mi to under 90 in./mi when the overlay (one lift) is placed on the cold recycled mix. The current cost of conventional hot asphalt mix is around $110/ton. The cost of the entire cold recycled mix project (material processing, mix production, placing) is between $46 and $47 per ton, which is a cost savings of at least 54%. Case Example Four – Utah DOT Cold Recycling Experiences The Utah DOT recently completed a 5-year research study, which resulted in the implemen- tation of Utah DOT Materials Manual, Part 8-965: Guidelines for Evaluation, Mix Design and Field Acceptance of Cold Recycling of Asphalt Pavements Using Solventless Emulsion (Utah DOT 2017). The original research findings are documented in VanFrank et al. (2014), VanFrank (2015), and VanFrank et al. (2016). Utah DOT

Case Examples 105   The new standard includes four innovative components: • Adjusting the RAP gradation based on achievable density over a range of temperatures • Setting an optimum moisture content to prevent flash setting of the emulsion • Field testing in nearby field laboratories with quick turnaround times • Shear vane testing to determine when the roadway can be opened to traffic Discussions with Mr. Kevin VanFrank were used to clarify, expand, and summarize these points in the following sections (K. VanFrank, personal communication, July 23 and 31, 2020). Density over a Range of Temperatures The RAP temperature needs to be between 80°F (27°C) and 120°F (47°C) to achieve the densification. There is only a limited change in density when the temperature is below or above this range (Figure 53). The slope of the percent compaction–temperature relationship provides an understanding of how a particular RAP source responds to changes in temperatures. When the slope is flat (i.e., little change in density with temperature), the RAP will behave like a black rock. Steeper slopes indicate the workability and compactability of the cold recycled mix can be significantly influenced by changes in temperatures. This indicates the RAP may be binder-rich, and have less oxidized asphalt, from one or more chip seals. Understanding adjustments that need to be made during the mix design stage helps reduce unanticipated, but needed, changes in the materials and/or quantities during construction. It also highlights the importance of having experienced personnel evaluating and overseeing projects. For mix design purposes, the lowest desirable density was set at 92% of the RAP maximum specific gravity (8% air voids) for RAP-only specimens compacted at 80°F (27°C); the upper desirable density was set at 97% at 3% air voids when compacted at 120°F (47°C). Adjustments to the quantity and type(s) of materials are needed when the RAP-only densities do not meet these limits. Figure 53. Example of the compacted RAP density-temperature relationship.

106 Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling When the percent compaction at 80°F (27°C) is less than 92% (± 1%), fine corrective aggre- gate or Portland cement can be used to fill the voids between the RAP particles. If RAP-only specimens prepared at 120°F (47°C) have compaction levels higher than 97% (± 1%), other materials such as coarse corrective aggregates can help by providing more room in the mix for other materials. Specimens with the additional materials are prepared and tested. If the percent compaction at either temperature still does not meet the requirements, further material adjust- ments may be evaluated at the engineer’s discretion. Alternatively, the project may not be a good candidate for cold recycling. Setting an Optimum Water Content The optimum water content is defined as the water needed to prevent the emulsion from flash setting. The initial laboratory research showed that over 90% of the optimum moisture content needed to prevent flash set was held by the No. 8 (minus 2.36-mm) and No. 30 (minus 0.60-mm) RAP sieve sizes. A vibratory compaction table (15 seconds, 50 Hz), to represent steel wheel rollers in vibratory mode, is used to consolidate a fixed fine RAP gradation. The amount of water needed to show a thin film of water on the surface is defined as the moisture content needed to prevent the emulsion from breaking. This is the moisture content for the fine RAP. The water needed for the entire RAP gradation is defined as the ratio of the percent water needed to prevent the emulsion from breaking to the percent fine RAP in the total RAP grada- tion. The optimum water content includes added water and the water content in the lime slurry used to improve moisture resistance. Field Testing Nuclear density gauges used to measure the in-place density are calibrated with wet theo- retical maximum specific gravities. Wet theoretical maximum specific gravities are calculated as follows: • Set the target density as a percentage of the mix design theoretical maximum specific gravity. • Measure the moisture content of the cold recycled mix, w% (in decimal form). • Divide the mix design theoretical maximum specific gravity by (1 – w%). • Input this value into the nuclear density gauge as the target density. The nuclear gauge readings can be adjusted to reflect density estimates based on current moisture content. The Utah DOT requires the contractor to maintain a field laboratory within 5 minutes of the construction site, and results need to be completed within 30 minutes of obtaining the sample. The distance and time limit were set based on a typical speed of 30 ft/min of paving. Therefore, test results can be obtained for every 1,000 ft (300 meters) of paving. The cold recycled mix needs to exhibit cohesive characteristics (i.e., the emulsion needs to break) before rolling begins. Rolling may need to be delayed until this point is reached. If rolling starts too soon, the mix moves out from under the roller and is not densified. It is best to have more than one steel wheel vibratory (40 ton) breakdown roller and a pneumatic rubber tire roller available to maximize the density in a short time without having to increase the speed of the rollers to keep up with the paver. Rollers that move too quickly create fine ripples on the surface that can influence ride quality. The shear vane test is used to estimate timing for compaction and opening to traffic: • Between 25 and 35 ft-lb is considered the compaction “sweet spot.” • Over 40 ft-lb indicates finish rolling needs to be completed. • Greater than 30 ft-lb indicates the mat can be opened to traffic.

Case Examples 107   The equipment consists of a 5-lb sledgehammer, a torque wrench capable of reading to at least 150 ft-lb, a segmented vane [approximately 3 in. by 3 in. (75 mm by 75 mm)], and a 15�16-in. socket (Figure 54). The vane is driven into the mat with the sledgehammer, and the socket and torque wrench are used to rotate the vane 90° in 10 seconds within the mat surface. The original research study also used a Marshall hammer to evaluate the effort needed to make an indentation in the mat. But continued work with both methods showed the shear vane test was the most useful method. Case Example Five – Caltrans Experience with Cold Recycling Smoothness Specification Numerous agencies have implemented the use of IRI ride quality specifications for accep- tance of conventional pavement projects. However, no information was found in the literature review that evaluates the ability of cold recycling processes to meet these specifications. Over the past decade, Caltrans has transitioned from using the California profilograph to inertial profilers for acceptance. Caltrans was also developing cold recycling specifications over the same period. Mr. Allen King of Caltrans worked with Don Matthews of Pavement Recycling Systems to document the ride quality of the existing pavement, the cold recycled mix layer, and the final overlay surface. Caltrans defines ride quality as the mean roughness index (MRI), which is the average of the wheel path IRI. Data were collected to establish the ability of cold recycling to improve existing ride quality as well as to evaluate the typical smoothness of the cold recycled mix layer (without an overlay) (Table 63). Regardless of the existing MRI, the CIR layer consistently has an MRI between about 75 and 90 in./mi. While CIR improves a rough existing pavement, it can make a smooth pavement somewhat rougher. Data were also collected for cold recycling projects’ ride quality after the final wearing surface, an overlay, was placed (Table 64). These projects then placed a 4-in. (0.33-ft) CIR layer with an overlay from 1.2 to 1.8 in. (0.1 to 0.15 ft) thick. The average MRI was 55 in./mi once the overlay was placed, and the average improvement in the ride quality was 41%. Source: VanFrank 2015. Figure 54. Components of shear vane test equipment. Caltrans

108 Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling MRI Range Segments Existing MRI Average CIR MRI Average Percent Improvement ≤60 34 53.7 76.6 −42.6% 61 to 70 35 65.7 83.8 −27.5% 71 to 80 44 76.4 89.3 −16.9% 81 to 90 56 86.6 84.0 3.0% 91 to 100 34 96.2 85.8 10.8% 101 to 120 54 110.9 88.7 20.0% 121 to 140 23 130.7 89.4 31.6% 141 to 180 77 164.0 90.4 44.9% 181 to 240 166 211.6 86.4 59.2% 241 to 300 85 268.1 88.5 67.0% 301 to 400 84 342.5 89.4 73.9% ≥401 28 441.0 89.4 79.7% Total 720 Source: King and Matthews 2019. Table 63. CIR ride quality measurements, without overlay. Treatment Data Points Existing MRI HMA Pavement MRI % Improvement CIR with Overlay 0.15' HMA-A / 0.33' CIR 33 77.0 48.7 36.8% 0.15' HMA-A / 0.33' CIR 39 95.1 57.0 40.1% 0.15' HMA-A / 0.33' CIR 150 87.3 50.6 42.0% 0.10' RHMA-G / 0.33' CIR 53 121.2 70.1 42.2% CIR Totals Average 275 93.7 55.0 41.3% Source: King and Matthews 2019. Table 64. Smoothness improvements for CIR with overlay, cold planing with rubberized hot mix asphalt surface course, and bonded wearing course options. Case Example Six – More than Four Decades of Federal Lands Highway Experience with Cold Recycling The Federal Lands Highway (FLH) is the roadbuilding component of the FHWA. The core business of the FLH is providing engineering and construction services for highway and trans- portation facility projects on, or serving, federally owned land. Over the past 10 years, FLH has used in-place recycling to rehabilitate 1,430 miles with FDR (85%) and CIR, usually with an overlay (15%) (Voth 2019). The FLH in-place recycling program started in the early 1980s and routinely constructs CIR on roadways in extreme environments. A CIR with an overlay was constructed in 2006 on

Case Examples 109   Badwater Road in Death Valley, and after 11 years, the pavement condition rating was 95.8. In 1988, the FLH constructed the first CIR project in California on Ice House Road in the Eldorado National Forest. The 13-mile roadway was 4 to 5 in. (100 to 125 mm) of CIR with a 2-in. (50-mm) overlay and needed to withstand 1,000 AADT with heavy logging trucks. After 33 years, the roadway is still in good shape, albeit with fine block cracking (Figure 55). Candidate CIR projects are selected based on a visual distress evaluation, a soil analysis, and the existing pavement thickness (M. Voth, personal communication, July 27, 2020). Localized areas of webbing (i.e., alligator cracking), which indicates localized loss of support, can be managed with spot repairs or edge drains. Soil testing includes determining the R-value and gradations to determine the soil classification. Pavement thickness is a key rehabilitation selection criterion. Rural roadways can be very thin, sometimes only 1.5 to 2 in. (37 to 50 mm) of asphalt pavement topped with multiple chip seals. At least 2 in. (50 mm) is needed to provide adequate material for CIR processes. CIR proj- ects can recycle 100% of the existing pavement, down to the top of the subgrade. When asked if support for the construction equipment is a problem when all the existing pavement is removed, Mr. Voth noted that old pavement without evidence of support-related distress is typically a good indication of adequate support, which can be confirmed with the soil testing. Experience has shown that if 2 in. (50 mm) or more of the existing pavement cannot be left after milling, then the entire pavement layer needs to be recycled. Leaving only 1 in. (25 mm) of old pavement when using CIR is not recommended because the equipment can break up the thin layer of old asphalt, creating an uneven paving surface. If the pavement is too thin for CIR [usually less than 2 in. (50 mm)], then some form of FDR is likely the best option. Mr. Voth indicated the FLH has more experience with emulsion than foamed asphalt recy- cling agents. The reason is the availability of materials rather than a materials selection consider- ation. The limited availability of recycling contractors and the types of processes the contractors can provide frequently restrict recycling and material options. The density that needs to be met during construction is usually based on a test strip. The wet density is measured with a nuclear density gauge. This provides a baseline for density measure- ments but not an absolute measurement of density. Traffic is usually controlled with pilot cars during construction and with signage for speed control until the wearing surface is placed. The wearing surface is placed once the moisture content is below 2.5% but no more than 14 days after the CIR is placed. Secondary compaction can improve density, but when the compaction is the most beneficial depends on the site-specific environmental conditions. In dry climates, secondary rolling can Figure 55. Ice House Road, 33 years of performance. Source: Mike Voth, FLH.

110 Practice and Performance of Cold In-Place Recycling and Cold Central Plant Recycling be completed the next day, and then the surface can be fog sealed. In humid climates, secondary compaction may need to wait for several days until the CIR layer cures sufficiently. The timing tends to be at the discretion of the contractor. When the roadway needs to meet a ride quality specification, more experienced contractors profile the roadway before construction. The data are used to identify opportunities for pre- milling for profile corrections that can improve the final ride quality. When the CIR is done correctly, and once the overlay (usually only one lift) is placed, the contractor can frequently meet incentive requirements.