Composite Pavement Systems, Volume 1: HMA/PCC Composite Pavements (2013)

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.

14 This R21 project, “Composite Pavement Systems,” focused on providing strong, durable, safe, smooth, and quiet pavements needing minimal maintenance. Two strategies have shown great promise: (1) surfacing a new portland cement concrete (PCC) layer with a high-quality hot mix asphalt (HMA) layer(s), and (2) placing a relatively thin, high-quality PCC surface atop a thicker PCC layer. However, the structural and functional performances of these two types of composite pavements were not well understood or documented. Models for predicting the performance of these pavement systems needed to be developed and/or confirmed for use in design, pavement management, and life-cycle cost analysis (LCCA). In addition, guidance on the development of specifications, construction techniques, and quality management procedures was needed for these technologies to become widely adopted. Research Objectives and Overview The objectives of this research were to investigate the design and construction of new composite pavement systems and specifically not those resulting from the rehabilitation of existing pavements. The goal was to: 1. Determine the behavior of new composite pavement systems and identify critical material and performance parameters. 2. Develop and validate mechanistic-empirical (M-E) based performance prediction models and design procedures that are consistent with the Mechanistic-Empirical Pavement Design Guide (MEPDG). 3. Develop recommendations for construction specifications, techniques, and quality management procedures for adop- tion by the transportation community. This project consisted of the following three phases: • Phase 1 consisted of a literature search, survey of various national and international highway agencies, field survey of composite pavements in three European countries, an evaluation of existing design procedures, development of database for full-scale applications, populating the database with information from available projects, and an initial evaluation of existing data. Phase 1 was completed and the Phase 1 interim report prepared and submitted to SHRP 2 in May 2008. • Phase 2 consisted of further completion of the databases, analyzing the databases, identifying failure mechanisms and other distresses relevant to new composite pavements, performance modeling, and conducting parametric eval- uations of the performance models. Phase 2 also included the development of the detailed research plan for Phase 3. Phase 2 was completed and the Phase 2 interim report pre- pared and submitted to SHRP 2 in May 2009. • Phase 3 consisted of implementing the research plan devel- oped in Phase 2. Full-scale roadway and accelerated pave- ment testing (APT) sections were constructed and tested at MnROAD and the University of California Pavement Research Center (UCPRC), respectively. Field composite pavement sites with long-term performance were surveyed, and detailed information was collected in the United States; Ontario, Canada; and three European countries. The results of these investigations were used to refine and validate the performance models and develop the final design guidelines and procedures. Phase 3 also included the development of construction specifications, design guidelines, and a plan for long-term evaluation and validation of the design models, development of training materials, and delivery of the final report for this research. Overview of Report The purpose of this report is to present the work performed throughout the course of the project. Included are the exec- utive summary and two volumes. Volume 1 covers HMA/ PCC composite pavements and Volume 2 covers PCC/PCC C h a p t e R 1 Introduction and Background

15 composite pavements. Each volume includes six chapters. Chapter 1 presents the introduction and background. Chapter 2 includes details of test sections, and Chapter 3 covers the vari- ous aspects of the research relevant to analysis and modeling. Design and construction guidelines are included in Chapters 4 and 5, respectively. Chapter 6 includes product summary, conclusions, and recommendations for future research. Definitions HMA/PCC composite pavement systems for the purposes of this research are defined as relatively thin HMA layers over a newly placed, but sufficiently hardened, PCC layer (Figure 1.1). The term “HMA” is used to indicate all types of asphalt-based products, including stone matrix asphalt (SMA), dense and porous HMA (including polymer-modified asphalt [PMA]), asphalt rubber friction course (ARFC), and others. The wear- ing surface is a relatively thin, high-quality type of HMA that could consist of one or more layers of HMA with or without special layers or materials to retard reflection cracks. The PCC layer can consist of jointed plain concrete (JPC) or continuously reinforced concrete (CRC). The PCC materials of this layer can consist of conventional PCC, roller compacted concrete (RCC), or a lower cost PCC (such as with a softer large-aggregate or recycled PCC material, or what has previously been called lean concrete base [LCB]). The PCC substructure is the primary load-carrying layer and is designed to provide a durable, long- lasting pavement with low fatigue damage and a strong base, whereas the HMA layer is primarily a functional layer with excellent surface characteristics that can be renewed rapidly. history HMA/PCC composite pavements are by no means a recent development. They have been constructed since the 1950s using a cementitious base with an HMA wearing surface by various national, state/provincial, and local highway agencies, such as the states of New Jersey and Washington; Ontario; and the cities of Toronto, New York, Washington, D.C., and Columbus, Ohio. Columbus has constructed many composite pavements consisting of HMA over RCC on residential and collector streets in the past 15 or so years. Arizona has con- structed many new composite pavements over the past 17 years consisting of an ARFC over a thick JPC or CRC layer. Several European countries, such as the Netherlands, the United Kingdom, Germany, and Italy, have constructed major com- posite pavement projects with low noise HMA surfacing and either CRC or JPC as the lower layer. HMA/PCC composite pavements also are constructed routinely by many highway agencies when widening existing PCC pavements or existing overlaid HMA/PCC pavements. Appendix A provides a review of the history and background of HMA/PCC composite pavements. Major thrusts toward an engineered composite pavement began in the late 1950s, through the guidance of the Commit- tee on Composite Pavement Design of the Highway Research Board. An important task of this committee was to develop a precise definition of “composite pavement” because by some definitions, any pavement consisting of varied layer materials could be considered a composite structure. The eventual defini- tion decided on by the committee was (Smith 1963): A structure comprising multiple, structurally significant, layers of different, sometimes heterogeneous composition. Two layers or more must employ dissimilar, manufactured binding agents. As part of the movement toward a broader use of compos- ite pavements, numerous design possibilities were suggested for study (Van Breemen 1963), including the HMA/PCC composite pavement detailed in this report. Early full-scale test section research into the construction and evaluation of composite pavements with numerous layering options was conducted in Ontario (Smith 1963, Ryell and Corkill 1973). The focus of the study was multifold, including addressing the following questions: • Can a smooth-riding pavement be constructed easily by surfacing a concrete base with HMA layers? • What is the best combination of thicknesses of concrete base and HMA surface for a high-class type of pavement designed to carry heavy traffic with high structural capacity? • How can reflective cracking be prevented or reduced? Between the 1950s and the 1970s, several long-term studies on the performance of composite pavements were conducted in the United States and Canada. These studies include the Ontario Highway 401 Study (Smith 1963, Ryell and Corkill 1973); New Jersey Composite Pavement Study (Baker 1973); Federal Highway Administration (FHWA) Zero Maintenance Pavement Study (Darter and Barenberg 1976), which identified HMA/PCC composite pavement as one of the most promising low maintenance pavements; and FHWA Premium Pavements Figure 1.1. Typical cross section for HMA/PCC composite pavements. Hot Mix Asphalt (SMA, PMA , Superpave, etc. ) PCC Lay er RCC . Hot Mix Asphalt (SMA, PMA uperpave, etc.) P C Layer (JPCP, CRCP, RCC, etc.)

16 Study (Von Quintus et al. 1980). Transverse reflective crack deterioration was the major distress type observed on these composite pavements. Rutting was rated as only “minor” to “moderate” even under very heavy traffic (Darter and Barenberg 1976). The thinner HMA over a PCC slab seemed to have a definite effect on minimizing HMA rutting. Ryell and Corkill (1973) concluded that better performance may be achieved if the wide transverse cracks were prevented from occurring, which may be accomplished by the use of “transverse crack inducers” (joints) in the concrete base at approximately 15-ft centers. The authors suggested that the extra cost of this could be offset through use of lower quality concrete in the slab. In Spain, the technique of forming of joints in RCC and cement treated layers under HMA to control reflective cracking has been used since 1984 (Jofre et al. 1996). The joints were sawed in the beginning, but since 1991 wet-forming methods have been used. Long-term results show the effectiveness of wet-formed joints every 8 to 13 ft in terms of a reduction in deflections and high values of joint load transfer. This study also showed that short joint spacing led to fewer transverse reflection cracks, tighter cracks, and improved performance. As noted, urban areas have used HMA/PCC composite pavements as their primary pavement design strategy for many years because of the perceived benefits regarding ease of maintenance from the HMA wearing surface and better load carrying capacity of the PCC base. One example is the city of New York, which has been using composite pavements since the 1990s. New York has found that reflective cracking is the primary distress that limits the performance of this design strategy. The city sponsored and built an experimental project that included HMA over jointed PCC (new construction) with various treatments and techniques to retard and prevent the deterioration of reflective cracks in the HMA wearing surface. The reflective cracking treatment that was found to be most economical and has provided consistently good performance was the saw and seal method. This has also worked well for HMA overlays of JPC for many years in many states. For more than 15 years, Arizona has been building a thin ARFC on all JPC pavements constructed in urban areas to provide a low noise surface. Although Arizona has had success with this type of pavement, performance data on this type of pavement in other parts of the country are limited. agency Survey The research team conducted a survey of U.S. and international highway agencies to assess the state of the practice and knowl- edge regarding composite pavement systems. The goals of this survey included • Assessing the interest of various highway agencies in designing/building composite pavements within their jurisdiction. • Identifying agency contacts and projects that can be used in the R21 database for development of the performance models. • Gathering information on individual agencies’ experiences with composite pavements and identifying the appropriate contacts for development of guidelines and construction specifications. A list of key agencies to be contacted was developed. These agencies included the 50 states of the United States, the District of Columbia, the provinces of Ontario and Quebec in Canada, and Austria, Belgium, Czech Republic, Germany, United Kingdom, the Netherlands, Italy, France, Spain, Sweden, South Africa, and Australia. The initial request consisted of a few questions, and agencies that responded positively were contacted for additional information on spe- cific field sections. Responses were received from 35 of 51 (69%) of the U.S. agencies and 7 of 14 (50%) of the inter- national agencies contacted. The results of the survey are summarized in Figures 1.2, 1.3, and 1.4 and are detailed in Appendix C. Summary of european practices Many European countries have been constructing HMA/PCC composite pavements on major projects for several decades and have substantial experience with the design and construction of composite pavements (Hassan et al. 2008). Members of the SHRP 2 R21 research team conducted a trip to some of these European countries to better understand and document their experiences with the construction of composite pavements. Tompkins, Khazanovich, and Darter (2010) described case study projects visited in the Netherlands, Germany, and Austria. Figure 1.2. Pie chart depicting agency response to the question “Has your agency constructed new HMA/PCC composite pavements in the past 20 years?” 24 11 1 6 U.S. - No U.S. - Yes International - No International - Yes

17 The Netherlands has built porous HMA over recently placed CRC on a number of major projects during the past 13 years. These projects are all performing very well with very low noise levels and no reflection cracking from the CRC despite their relatively thin HMA layer of about 2 to 3 in. Germany has built SMA surfaces on JPC and most recently over CRC. One SMA/ JPC section was 15 years old under heavy traffic with sawed and sealed joints that had performed very well. Germany has recently constructed SMA over CRC. The United Kingdom also has constructed major projects and researched thin surface course systems (TSCS) over CRC and found significant technical and functional benefits (Hassan et al. 2008). In reviewing these case studies and discussing the composite pavements with the host engineers and practitioners, a number of benefits to importing and implementing European techniques were identified. Dutch, German, and Austrian researchers report that com- posite pavements provide similar structural performance as an equivalently thick single layer at the same price in Europe, with the added benefits of higher quality and longer life friction and noise reduction due to the high-quality top layer. Furthermore, composite pavements allow for the optimization of costs and materials throughout the pavement cross section because • High-quality materials can be used in lesser quantities in the upper layer, where they will be of the most benefit to the system. • Lower quality (cheaper) materials can be used in greater quantities and in the lower layer, where they will contribute structurally without detracting from the quality and perfor- mance of the overall pavement. Although there are obstacles to the adoption of composite paving in the United States, it is clear from the European experience that overcoming these obstacles will result in high-quality, durable, and sustainable composite pavements. Distress Mechanisms The distress mechanisms for HMA/PCC composite pave- ments are a combination of those of both HMA and PCC pavements and can be divided into three basic categories: fracture, distortion, and disintegration, as shown in Table 1.1. Details of these distress mechanisms and how they relate to the design of composite pavements are discussed in Appendix D. Reflection Cracking Reflection cracking is the most common distress observed in HMA/PCC composite pavements and the most serious in terms of requiring maintenance and rehabilitation. Reflection 11 13 7 4 No Yes Maybe No response Figure 1.3. Pie chart depicting agency response to the question “Is the design and construction of new HMA/PCC composite pavements of interest to your agency?” 0 2 4 6 8 10 12 Cost Reflection cracking and AC/PCC bond Industry acceptance and related issues Lack of experience with this type of construction Lack of long-term data (e.g. for local calibration and LCCA) Rehabilitation and characterization of underlying PCC Construction time and related issues Surface durability (studded tire wear of AC layer) Figure 1.4. Key concerns of agencies regarding construction of HMA/PCC composite pavements in their jurisdictions.

18 cracking is caused by horizontal and or differential vertical movements between different layers at a discontinuity in the underlying PCC layer. Thus, the jointing or cracking of the underlying PCC slab is one factor that is critical to reflection cracking, with shorter spaced joints and cracks minimizing the extent and severity of reflection cracking. The load transfer efficiency of these joints and cracks over time is also critical to their occurrence and deterioration. Various systems have been used to retard reflection cracking of HMA layers of new composite structure, including • Bond breakers: Bond breakers (such as stone dust) have been used to isolate the movements of the PCC layer from the HMA layer. However, these are not effective in isolating the movements and the lack of bond between the HMA and PCC layer results in other distresses, such as fatigue cracking, potholes, and slippage cracks. • Asphalt-rubber interlayer: Stress-absorbing membrane inter- layer (SAMI) or strain-relieving interlayer and cushion courses (crack relief layer) are used at the underlying joints to reduce the effect of horizontal and vertical movements in the PCC layer. The stress-absorbing membrane and cushion courses are used to minimize the occurrence of reflection cracks and are placed between the HMA and PCC layer. An example of an SAMI type device is the interlayer stress- absorbing composite (ISAC) of three-layer design: low stiff- ness geotextile (next to PCC layer), viscoelastic membrane, and high stiffness geotextile (next to HMA layer). • Fabrics and geotextiles: Certain paving fabrics and geotextiles that retard reflection cracks have been used with varying degrees of success. • Geogrids: Geogrids and other reinforcing materials have been placed within the HMA to prevent or delay the cracks from propagating to the surface of the HMA. • Reflection crack relief interlayer (RCRI): This can include a cushion course of high-void coarse open-graded HMA mix containing 25% to 35% interconnecting voids and composed of 100% crushed material or can include an unbound aggregate base. • Thicker HMA layer: A thicker HMA layer has been shown to slow the occurrence and progression of reflection cracks but can result in higher costs and thicker pavement structures, which could affect bridge clearances. • Saw and seal the HMA layer: Sawing and sealing the HMA layer above the joints in the PCC base has been the most successful method for controlling reflection cracking in HMA/PCC pavements and has been used by many states for more than 40 years. Although the occurrence of a reflection crack is the only aspect considered in most of these control techniques, the severity of the crack also is critical. Only the saw and seal method directly addresses this aspect. In addition, reflection cracks and deterioration of reflection cracks are primarily an issue only if the PCC is jointed and is not an issue if the PCC is a properly designed CRC with adequate steel reinforcement and PCC thickness. Fatigue Cracking Area fatigue cracking, which is typically observed in flexible pavements, does not usually initiate at the bottom of the HMA layer for HMA/PCC composite pavements because the HMA is almost always in compression, unless there is a loss of friction between the HMA and PCC layers. Fatigue cracks in HMA/PCC composite pavements normally initiate at the bottom (and top) of the PCC layer (bottom of the HMA layer only if HMA does not exhibit friction with the JPC) and propagate to the surface with continued traffic applications. The primary pavement response related to fatigue cracking is the maximum tensile stress at the top and bottom of the PCC layer. HMA/CRC also can develop fatigue-related distress in the form of edge punchout. This distress initiates with repeated load erosion of support, the deterioration of closely spaced transverse cracks, and fatigue damage at the top of the slab to create a “ladder” longitudinal crack about 48 in. from the pavement’s edge. Table 1.1. Pavement Distress Mechanisms for HMA/PCC Composite Pavements Distress Category Fracture Distortion Disintegration Fatigue cracking of JPC Rutting Raveling Edge punchouts of CRC Reduced skid resistance Reflection of transverse cracks and joints Freeze–thaw degradation Reflection of longitudinal joints Spalling at shoulders Debonding/loss of friction between HMA/PCC

19 Because the PCC layer is the primary load-carrying com- ponent of the HMA/PCC composite pavement system, the material properties of the PCC layer (along with thickness and joint spacing) affect the structural capacity of the composite pavements. Key material properties include flexural strength, elastic modulus, coefficient of thermal expansion (CTE), per- manent built-in temperature gradient, thermal conductivity, and heat capacity. In addition, the moisture content at the top of the slab beneath an HMA surface is near saturation, as is the bottom of the slab, thus eliminating a moisture gradient through the slab. One or more of these factors are affected by cement type, cementitious material content in the mix, water– cement ratio, aggregate type, ultimate shrinkage, and curing. Rutting Rutting is a materials-related issue and has been observed on a limited basis in HMA/PCC composite pavements. Rutting can be prevented through the mixture design and materials selection process. The “flexible” layers above the PCC layer usually consist of one or multiple layers of HMA. The most important property of the HMA layer is its stability of resistance to permanent or plastic deformation. The fatigue resistance of the HMA mixtures is less important because of the PCC layer that reduces the deflections and horizontal strains throughout the HMA layers. Thickness of the HMA layer will affect rutting potential. Rutting in the HMA layer (which is the only place where HMA/PCC can permanently deform) is related to the state of stress or strain in the HMA layers. When a heavy load is applied and repeated on a composite pavement, some per- manent deformation develops in the HMA layer. Permanent deformation may, depending on the stress levels, develop in asphalt-bound or unbound layers beneath the PCC slab after the slab cracks into many pieces. Permanent surface deflection reflects the permanent deformation of the HMA layer only, whereas the total loaded deflection (elastic and plastic) reflects the permanent deformation of the HMA layer, as well as any permanent deformation in the layers beneath the PCC slab. The magnitude of the rutting in the unbound aggregate base layers and subgrade is nonexistent under the intact PCC layer because the vertical compressive strains in those unbound layers are small. Debonding or Loss of Interlayer Friction Another problem that has been observed on a few HMA/PCC composite pavements is the lack or loss of interface friction or bond between the HMA surface and PCC base. Inadequate bond will result in fatigue cracking, potholes, and slippage cracks. A permanent and full adhesion and friction between the HMA and rigid layer is critical for the durability of the entire structure. Thus, a tack coat has to be applied. A tack coat is a light application of asphalt, normally asphalt emulsion (asphalt diluted with water). It is also beneficial to texture the PCC surface to enhance permanent bonding and friction between the HMA and PCC layer with no slippage. A detailed literature review on friction and debonding is included in Appendix O. Low-Temperature Thermal Cracking Low-temperature thermal cracking is a minor issue that is not very likely to occur in HMA/PCC composite pavements, assuming that adequate bond is retained between the different layers. Thermal cracking results when tensile stresses, caused by temperature variations or low temperatures, exceed the material’s fracture strength. Joint reflection cracks relieve low-temperature stresses, but regular low-temperature cracks have not been observed in HMA/CRC pavements in cold areas. Longitudinal Cracking Longitudinal cracking is not a critical issue for HMA/PCC composite pavements and usually results from paving operation problems. Surface-initiated cracks for good quality composite pavements are small because the tensile strains at the surface of the HMA layer are small, if they occur at all. Freeze–Thaw Degradation The freeze–thaw durability of the underlying PCC layer is a key factor that affects the long-term performance of HMA/ PCC composite pavements. Although the surface layer can be expected to be removed and replaced every 10 to 15 years, the underlying PCC layer is expected to be designed for more than 30 years. Freeze–thaw durability is particularly important if recycled concrete aggregate (RCA) is used in the underlying PCC layer. Construction Defects Construction defects that occur during placement can result in distresses on HMA/PCC composite pavements. Construction defects include segregation in the HMA (both longitudinal and truck-to-truck segregation), inadequate densities along longitudinal construction joints, centerline streak down the center of the paver, and so forth. These defects can be related to placement or to the materials used. Segregation is probably the most common defect that has been exhibited on many HMA layers. Segregation will result in raveling and cracking of the HMA layer. Construction defects can be reduced only with an adequate quality control (QC) and quality assurance (QA) and inspection program.

20 Use of hMa/pCC Composite pavements A key question that often is asked with regard to HMA/PCC composite pavements is “Where will composite pavements be used, and what will be the demand?” HMA/PCC composite pavements allow the pavement designer to design pavements using the best qualities of both HMA and PCC pavements to produce a more functional and economical structure that generally is more cost-effective in terms of service through- out its life. Specifically, HMA/PCC composite pavements are optimal solutions for the following situations: • In some design situations, based on materials, climate, traffic, and support conditions, the life-cycle costs for HMA/PCC composite pavements can be lower than those for conventional HMA or PCC pavements. This is particu- larly true when the thickness of the PCC can be reduced because of the decrease in thermal and moisture gradients in the PCC relative to bare PCC pavement. Comparative designs included in this report show a significant reduction in PCC slab thickness. • In urban areas with high costs of lane closures, rapid renewal is paramount. HMA/PCC pavements can be designed for the PCC to have a long life, structurally speaking (if durable materials are used). The HMA can be milled and replaced rapidly with minimal disruption to traffic. • In some situations, agencies want to design and construct pavements with the structural capacity of PCC pavements but functional characteristics of HMA surfacing—premium pavement. • Where high-quality aggregates for PCC are not available (or expensive because of long haul distances), local PCC aggregates may be susceptible to polishing and other durability-related distresses. In these situations, HMA surfaces can protect the structural integrity of the PCC and be milled and rapidly replaced as needed. • Similarly, the HMA layer can be used as a sacrificial layer, rather than providing a thicker PCC layer (designed for future milling/grinding), where studded tires are an issue. The HMA layer can be milled and rapidly replaced as needed. • Many urban areas and some rural areas exist with old PCC pavements that can be removed and processed and recycled directly back into lower layer PCC for use in HMA/PCC composite pavements. This provides excellent improved sustainability opportunities for pavements. • In urban areas with large populations in close proximity to the pavements, low pavement noise is needed. HMA surfacings of PCC provide riding surfaces with low noise, good friction, and good ride quality. Arizona has built many miles of major freeways with porous rubberized asphalt surface over new JPC and CRC to minimize noise. Noise considerations also are a major reason HMA/PCC composite pavements are constructed in many European countries. • Where transverse cracks and deterioration of transverse cracks are a problem, HMA/CRC is a good alternative to eliminate reflection of transverse cracks. • HMA/PCC composite pavements may be used for widening existing PCC or HMA/PCC pavement such that the widened section is compatible structurally with the existing pavement. The new and existing lanes typically are covered with one or more lifts of HMA. Differences Between HMA Overlay of Old Concrete and New HMA/PCC Composite Pavement There are several key differences that should provide for supe- rior performance of a new HMA/PCC composite pavement as compared with an HMA overlay of existing jointed plain concrete pavement (JPCP): • The concrete slab is undamaged. 4 No fatigue damage or fatigue cracks exist in the concrete slabs; thus, with proper design, fewer fatigue cracks are expected to develop over the design life. 4 No durability-related distresses or spalling exist in the concrete slabs, thus minimizing the chances of localized failures of the HMA surface. 4 A new concrete slab is less likely to have localized areas that rock and cause reflection cracks through the HMA surface. 4 New transverse joints have much higher load transfer, leading to lower deterioration rates for the functional thin HMA surface. This is a major difference that reduces the deterioration of reflection cracks from transverse joints. • The new PCC layer should be built to smoothness specifi- cations, and this provides the opportunity to build a very smooth HMA surface on top. • There is improved bond between the HMA surface and the concrete slab because of the tack coat and because it is cleaner and textured for a mechanical.