Practices for Unbound Aggregate Pavement Layers (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.

9 chapter two aGGREGaTE TYPES aND MaTERIaL SELECTION INTRODUCTION Continuous removal of available natural resources and increased material hauling and transportation costs have put an emphasis on finding “acceptable” materials to be used in pavement construction. The performance of any constructed pavement system largely depends on the quality of materi- als used in different layers. To ensure adequate performance of pavements under loading, transportation agencies have developed specifications that address certain minimum prop- erties or qualities of construction materials. Performance of aggregates used in unbound pavement layers under loading is also influenced by such material properties of individual particles and the particle arrangement within the aggregate matrix as a bulk material. This chapter provides a brief overview of the different types of aggregate materials available as natural resources and mined from sand and gravel pits and crushed stone quarry operations throughout the United States. Important aggre- gate properties and quality aspects, which enable aggregate material to meet agency specifications for pavement granular base/subbase use, are summarized to establish guidelines for aggregate source selection. Next, the concept of best value granular material use is introduced for pavement projects with the potential to save energy and material hauling costs through examples of recent sustainable construction practices, high- lighting how local natural aggregates and recycled materials could be better used in granular base/subbase applications. The increasing trend to use recycled granular materials in base and subbase layers is discussed in detail. Important issues are reviewed to shed light onto what tests are used by agencies to characterize recycled materials for unbound granular base/ subbase acceptance and design. aGGREGaTE TYPES aND SOURCES According to the ASTM, aggregates are defined as “granu- lar materials of mineral composition such as sand, gravel, shell, slag, or crushed stone, used with a cementing medium to form mortars or concrete, or alone as in base courses, rail- road ballasts, etc.” Based on the nature of their extraction from natural resources, aggregates used in pavement appli- cations can be divided into two broad categories: (1) stone deposits and (2) sand and gravel deposits (Barksdale 1991). Industrial by-product materials, such as slags, have also been specified and used in granular base and subbase applications in some states. Stone Deposits Stone deposits can be broadly classified into the following three categories: (1) sedimentary rocks, (2) igneous rocks, and (3) metamorphic rocks. A brief discussion of the mech- anism of formation for these three rock types is presented here, along with examples of each rock type. These three rock types usually are obtained from quarries through a blast- ing process and are processed through a series of crushers, pulverizers, and screening units to obtain aggregate materials for pavement and other construction applications. Note that depending on the processing methods, crushed aggregates produced by a crusher operation may be dry or wet immedi- ately after production. 1. Sedimentary Rocks: These rock types are formed by chemical precipitates and the settlement of sediments or organic matter at or near the earth’s surface and usu- ally within bodies of water. Examples of sedimentary rocks include limestone, dolomite or dolostone, shale, and sandstone. The generic name “limestone” is used for com- monly found carbonate rocks, including limestone, dolomite, and marble (Langer 2011). Limestone and dolomite usually form as a result of the consolida- tion and sedimentation of the shells of marine ani- mals and/or plants. They may also form as a result of the precipitation of fine carbonate mud from marine waters. Limestone and dolomite constitute approxi- mately 70% of crushed stone production in the United States (Willett 2011). 2. Igneous Rocks: These rock types are formed by the cooling and solidification of magma or lava. The pro- cess of magma solidification can occur below or on the earth’s surface. Accordingly, igneous rocks formed below the earth’s surface are called intrusive or pluto- nic rocks, whereas those formed on the earth’s surface are called extrusive or volcanic rocks. As a result of the longer duration associated with cooling of magma in the formation of intrusive igneous rocks, the indi- vidual minerals have a chance to grow large enough

10 to see with a naked eye, and the rock has a coarse- grained texture, but extrusive igneous rocks, which cool rapidly from magma at or near the earth’s surface, are too fine-grained to distinguish individual miner- als. Igneous rocks often have high amounts of silica. Examples of igneous rocks used in pavement appli- cations include granite (intrusive), basalt (extrusive), and rhyolite (extrusive). The generic classification “granite” sometimes includes coarse-grained igneous or metamorphic rocks such as true granite, syenite, gneiss, and dark-colored gabbro (Langer 2011). Granites account for approxi- mately 16% of crushed stone production in the United States (9% of total aggregate production). Although the hardness of individual particles leads to granite usually being classified as excellent crushed stone, some gra- nitic type aggregates are weak and brittle because of their poorly bonded mineral grains, usually caused by weathering. Fine-grained igneous rocks are also called “trap rocks.” Trap rocks include dark-colored, fine-grained, volcanic rocks and make up about 9% of the crushed stone production (5% of the total aggregate produc- tion) (Willett 2008). Examples of trap rock are basalt and diabase. Excellent resistance to chemical reac- tions and ability to withstand high mechanical stresses led to the classification of trap rock as an excellent crushed stone material. 3. Metamorphic Rocks: These rocks are formed by the transformation of existing rocks (may be sedimen- tary or igneous) under heat and pressure. Examples of meta morphic rocks include quartzite, marble, slate, and gneiss. Metamorphic rocks as an aggregate can have widely variable characteristics. Many quartzites and gneiss can have properties similar to those of granite, whereas shale can be slabby and schist can be soft and flaky because of its high mica content. Sand and Gravel Deposits Aggregates are also extracted from sand and gravel pits, where the parent material has been transported from another location by fluvial, glacial, or alluvial processes to form loose deposits of natural sand and gravel. They are usually found in existing or historic river valleys or older, consoli- dated bedrock, glacial deposition, and mountain alluvial fans. Sand and gravel make up approximately 42% of the total aggregate production in the United States (Langer 2011). Depending on agency specifications and the nature of the deposits, aggregates obtained from gravel and sand pits may or may not be processed through a series of crushers before being used for pavement applications. Coarser sand and gravel materials are better for this purpose because the coarse particles can be crushed to smaller sizes. Note that in some cases, cobbles (particles larger than 75 mm or 3 in.) and small boulders (particles larger than 305 mm or 12 in.) in gravel deposits are also crushed. Apart from the above two types of natural sources, other sources of aggregates include recycled materials and indus- trial by-products. A detailed discussion on different recycled materials and industrial by-products used in the construction of UAB and subbase layers is presented later in this chapter. SUPPLY aND DEMaND FOR aGGREGaTES IN THE UNITED STaTES According to the USGS, the demand for all types and uses of aggregates in 2007 and 2008 was on the order of 2.5 to 3.0 billion tons (2.2 to 2.7 billion metric tons) (Meininger and Stokowski 2011). These aggregates are obtained from natural resources or from recycled materials and/or industrial by-products. To improve and maintain the present condi- tions of the nation’s infrastructure at an acceptable level, the demand for aggregates is presumed to increase with time. However, the supply of available natural aggregates is limited and undergoes gradual depletion with continual extraction and usage. Moreover, the availability of natural aggregate resources is constrained by geologic formations, encroach- ing land development, and the resource’s proximity to the intended place of usage. Therefore, although some regions in the country may have abundance of available natural aggre- gate supply, natural aggregate supplies are scarce in most regions of the country. Figure 5 shows the relative locations of aggregate resources in the conterminous United States (Langer 2011). As indi- cated, there is a limited supply of natural aggregates in the Coastal Plain and Mississippi embayment, Colorado Plateau and Wyoming Basin, glaciated Midwest, High Plains, and the nonglaciated Northern Plains. Thus, construction projects in these regions often require transportation of natural aggre- gates from other sources. Moreover, the limestone found in several regions of the country does not meet the hardness and durability requirements for use in pavement base and subbase layer applications. These conditions often demand the transportation of “good quality” natural aggregates from nearby sources to be used in pavement applications. According to a 1998 USGS report, 27% of the crushed stone produced annually in the United States was used in pavement base construction (Wilburn and Goonan 1998). Similarly, 43% of the cement concrete debris produced was used for road base construction. On the other hand, 23% of the total sand and gravel production was used for road base construction, with portland cement concrete (PCC) production accounting for the highest proportion (45%) of use. According to the 2010 Minerals Yearbook published by the USGS, approximately 58.7 million metric tons of crushed stone was used in the United States for graded road base or sub base applications (Willett 2011). Similarly,

11 approximately 83 million metric tons of construction sand and gravel were used for road base and subgrade coverings (Bolen 2012). AGGREGATE PROPERTIES AFFECTING UNBOUND AGGREGATE LAYER BEHAVIOR Physical characteristics of the rocks that govern load- dissipating and particle-interlocking aspects differentiate “good” and “poor” quality aggregates with respect to the suitability for application in pavement unbound base/subbase courses. Moreover, chemical properties of the aggregates governing their durability and soundness are critical to ensur- ing long-lasting pavement structures. NCHRP Project 4-23, NCHRP Report 453: Performance-Related Tests of Aggre- gates for Use in Unbound Pavement Layers, summarizes the most important tests that relate to the performance of aggre- gates in unbound pavement layers (Saeed et al. 2001). Among the tests highlighted, the screening tests (sieve analysis, Atter- berg limits, moisture–density relationship, flat and elongated particles, uncompacted voids), durability test (sodium and magnesium sulfate soundness), shear strength tests [triaxial tests conducted on wet and dry samples and California bear- ing ratio (CBR) test], stiffness test (resilient modulus con- ducted on wet and dry samples), toughness and abrasion resistance tests (Los Angeles Abrasion and Micro-Deval), and frost susceptibility test (tube suction) are the most relevant for unbound aggregate pavement layers. Extensive review of technical literature was conducted to identify the most important physical properties affecting aggre- gate strength, modulus, and deformation behavior in unbound and bound pavement layers. A summary of the findings on important physical properties from the literature review is pre- sented here. Mineralogy Mineral composition of aggregates has a significant effect on the physical and chemical characteristics that ultimately gov- ern the performance of UAB/subbase layers under loading. This is particularly true as far as degradation and polishing of aggregates resulting from interparticle friction is concerned. For example, calcareous aggregates, such as limestone and dolomite, show significantly lower resistance to particle degradation and polishing. Therefore, UAB/subbase layers constructed using these aggregates are likely to undergo significant changes in gradation during compaction and sub- sequently under traffic loading. Note that not many research studies have evaluated directly the effects of aggregate min- eralogy on UAB/subbase performance. On the other hand, research studies have generally focused on evaluating the effects of aggregate physical characteristics influenced by mineralogy on performance. Woolf (1952) presented exten- sive data on the results from physical tests on road building aggregates. From his data, the effects of aggregate miner- alogy on physical characteristics are clearly apparent. For FIGURE 5 Generalized locations of aggregate resources in the conterminous United States (Langer 2011).

12 example, the average reported loss by abrasion for granite was 4.3%, whereas the corresponding values for limestone and dolomite were 5.7% and 5.5%, respectively. Particle Size Distribution and Fines Content One of the primary variables in any laboratory testing of aggregate materials is the grain size distribution. Differences in aggregate gradations can often lead to significantly differ- ent behavior for the same aggregate type. This is the result of the different packing order and void distributions that play a crucial role in load carrying through particle-to-particle contact in an aggregate matrix. To control the gradation of an individual aggregate sample, sieving and size separation of the aggregate materials need to be undertaken based on washed sieve analysis. Gradation itself is a key factor influ- encing not only the mechanical response behavior charac- terized by resilient modulus, shear strength, and permanent deformation, but also permeability, frost susceptibility, ero- sion susceptibility, and so forth (Bilodeau et al. 2007, 2008). Note that sieve shakers used to separate aggregate sizes based on dry sieving of the aggregate stockpiles can give erroneous size distributions. In a recent study, wet sieving results showed that the actual fines content (note that unless otherwise specified, “fines” in this synthesis refers to material finer than 0.075 mm or passing No. 200 sieve) of an aggre- gate sample was always higher than the target fines content during a blending operation (Tutumluer et al. 2009). This dif- ference in achieved versus target fines content was attributed to the significant amount of fines that remained stuck to the surfaces of larger particles during dry sieving and contrib- uted to changing the performance of the aggregate layer as a whole. For example, aggregate samples blended with tar- geted 0% fines (material passing sieve No. 200 or 0.075 mm) contained 4.4% fines for a limestone and 2.9% fines for an uncrushed gravel material (Tutumluer et al. 2009). There- fore, these fines had to be accounted for appropriately during study of the effects of fines on aggregate strength and defor- mation behavior. Gradation and fines content are interconnected in their effects on strength and resilient and permanent deformation characteristics. For a dense-graded crushed aggregate base material having a 25-mm (1-in.) top size, Gray’s (1962) pio- neering work indicated that maximum strength was achieved at a fines content of about 8%. As the maximum aggregate size increased, the optimum amount of fines that gave the maximum strength typically decreased. Using a directional modulus approach by changing the pulsing direction in repeated load triaxial tests, Tutumluer and Seyhan (2000) also determined an optimum fines content of 7% for a dense- graded crushed limestone aggregate base material. Well-graded aggregates have been found to have higher resilient modulus values to the point at which the fines content of the mixture displaces the coarse particles and the proper- ties of the fines dominates (Jorenby and Hicks 1986; Kamal et al. 1993; Lekarp et al. 2000a). Barksdale and Itani (1989) found a dramatic 60% reduction in the resilient modulus when the fines content was increased from 0% to 10%. Thom and Brown (1988) found that the effect of grading varied with the compaction level; when uncompacted, specimens with uniform grading accumulated the least permanent deforma- tion, whereas the resistance to permanent deformation was similar for all gradations when the specimens were heavily compacted. Kamal et al. (1993) and Dawson et al. (1996) found the effect of grading to be more significant than the degree of compaction (DOC), with the densest mix having the highest permanent deformation resistance. Brown and Chan (1996) successfully reduced rutting in granular base layers by selecting an optimum aggregate material grad- ing that maximized compacted density. These performance characteristics were demonstrated through experiments with two types of wheel tracking and the use of repeated load triaxial tests at the University of Nottingham in the United Kingdom. Increasing the amount of fines in a mix reduces the perma- nent deformation resistance (Barksdale 1972, 1991; Thom and Brown 1988). Moreover, the type of fines (nonplastic or plastic fines) in an aggregate layer has been found to affect the performance significantly. The results of a recent Illinois Department of Transportation (DOT) field study, Experi- mental Feature IL 03-01, indicate that increased aggregate fines had a significant effect on their performance in working platform applications (IDOT 2005). Bilodeau et al. (2009) identified, from a laboratory study conducted on the performance of unbound granular materials with six gradations and three aggregate sources commonly used in Canada, one fines-related volumetric parameter (termed fine fraction porosity, represented as a ratio between the total amount of voids in aggregate matrix to the total amount of voids if the entire matrix comprised coarse particles only) that described satisfactorily not only the mechanical performance but also the environmental stresses sensitivity of materials tested. Also identified from their study were the adapted (or optimized) gradation zones that ensured adequate overall per- formance of those three aggregate sources. Particle Shape, Surface Texture, and angularity The gradation, shape, and hardness have a great influence on the mechanical behavior and the strength properties of aggre- gate particles in contact. In general, it is preferable to have somewhat equidimensional (cubical) and angular particles rather than flat, thin, or elongated particles (Barksdale et al. 1992). Aggregate gradation is also critical for achieving good packing and minimal porosity in an aggregate mix. The maxi- mum size of aggregates, the size distribution, and the shape of the particles determine the packing density that can be

13 derived with an aggregate sample, assuming sufficient com- paction is provided. Angularity, a measure of crushed faces and sharpness of edges in an aggregate, is important because it determines the level of internal shear resistance that can be developed in the particulate medium. Round, uncrushed aggregates such as gravel, particularly with a smooth surface texture, tend to “roll” out from under traffic loads with low rutting resistance. Increasing particle angularity and roughness increase the resilient modulus while decreasing the Poisson’s ratio (Hicks and Monismith 1971; Allen and Thompson 1974; Thom 1988; Thom and Brown 1988; Barksdale and Itani 1989). The reported research indicates that aggregates made with uncrushed or partially crushed gravel particles have a lower resilient modulus than do those with angular crushed particles. This effect has been attributed to the higher num- ber of contact points in crushed aggregates, which distrib- ute loads better and create more friction between particles (Lekarp et al. 2000a). Allen (1973) and Barksdale and Itani (1989) investigated the effects of the particle surface characteristics of unbound aggregates and found that angular materials resisted perma- nent deformation better than did rounded particles because of the improved particle interlock and higher angle of shear resistance between particles. Similarly, Thom and Brown (1988) observed that permanent deformation was primar- ily affected by visible roughness of particles. Barksdale and Itani (1989) also concluded that blade-shaped crushed par- ticles are slightly more susceptible to rutting than are other types of crushed aggregate and that cube-shaped, rounded river gravel with smooth surfaces is more susceptible than is crushed aggregate. In the base courses, although compaction is important from a shear resistance and strength viewpoint, the size, shape, angularity, and texture of coarse aggregates are as important in providing stability (Barksdale 1991). Field tests of conventional asphalt pavement sections with two different base thicknesses and three different base gradations showed that crushed-stone bases gave excellent stability because of a uniform, high degree of density and little or no segregation (Barksdale 1984). Rounded river gravel with smooth sur- faces was found to be twice as susceptible to rutting as was crushed stone (Barksdale et al. 1989). Based on a review of several studies, Janoo (1998) con- cluded that shape, angularity, and roughness have significant effect on base performance and there could be as much as 50% change in resilient modulus of base materials owing to geometric irregularities of coarse and fine aggregate par- ticles. Saeed et al. (2001) showed a linkage between aggre- gate properties and unbound layer performance. That study showed that aggregate particle angularity and surface texture mostly affected shear strength and stiffness. Rao et al. (2002) studied the impact of imaging-based aggregate angularity index variations on the friction angle of different aggregate types and reported an increase in aggre- gate performance when the percentage of crushed particles was increased. An increase in crushed materials beyond 50% significantly increased the friction angle obtained from rapid shear triaxial tests, indicating a higher resistance to perma- nent deformation accumulation. Coarse aggregate angularity provides rutting resistance in flexible pavements as a result of improved shear strength of the UAB. The interlocking of angular particles results in a strong aggregate skeleton under applied loads; whereas, round particles tend to slide by or roll past each other, resulting in an unsuitable and weaker struc- ture. Later, Pan et al. (2006) prepared unbound specimens by blending six aggregate materials with uncrushed gravel and tested for resilient moduli. The modulus values of the aggre- gate specimens blended in different percentages were linked to the imaging-based shape indices. As the aggregate angu- larity and surface roughness increased, the resilient moduli were considerably improved, which was primarily because of the increased shear strength, with better aggregate inter- lock and frictional properties and the increased confinement levels expressed by higher bulk stresses. The NCHRP 4-30A project, Test Methods for Charac- terizing Aggregate Shape, Texture, and Angularity (NCHRP Report 555), recommended the Aggregate Image Measure- ment System (AIMS) and the University of Illinois Aggre- gate Image Analyzer (UIAIA) as viable imaging systems for analyzing aggregate morphology and quantifying aggregate morphologic effects to influence strength and permanent deformation behavior of unbound aggregate materials (Masad et al. 2007). Using the UIAIA system, Uthus et al. (2007) studied the aggregate morphologic property changes resulting from the rounding of aggregate particles in a ball mill drum. For cubical aggregates, the changes in angularity and surface texture appeared to have a significant effect on the elastic and plastic aggregate shakedown threshold limits, which will be discussed in more detail in chapter four. Tutumluer and Pan (2008) reported that aggregate blends comprising angular, rough particles consistently showed lower permanent defor- mation accumulations when studied using the UIAIA system. The angularity property was found to contribute mainly to the strength and stability of aggregate structure through confine- ment, whereas the surface texture property tended to mitigate the dilation effects through increasing friction between indi- vidual aggregate particles. A recent study (Gates et al. 2011) sponsored by the FHWA conducted an interlaboratory study using the recently improved Aggregate Image Measurement System 2 (AIMS2) device. Analyzing results obtained across 32 laboratories, the study concluded that aggregate size and shape proper- ties determined using the AIMS2 device showed reason- able coefficients of variation for all aggregate particle sizes greater than 0.075 mm. Findings from the study have led to

14 increased use of the AIMS2 device as an automated device capable of providing objective and reproducible shape char- acterization of aggregates. Degree of Compaction Before the aggregate samples are tested for strength, mod- ulus, and deformation behavior, the first task is to compact them at the corresponding gradation to determine their moisture–density relationships. Because pavement layers in the field often are compacted to predetermined percentages of the maximum dry density (MDD) values, it is important to establish the values of MDD and optimum moisture content (OMC) for each aggregate gradation. Thus, the objective of compaction is to improve the engineering properties of the soil mass. Through compaction, strength can be increased, deformation tendency can be reduced in the field, bearing capacity of the granular layer can be improved, and undesir- able volume changes (such as those caused by frost action, swelling, and shrinkage) may be controlled (Holtz 1990). Compaction methods applied on aggregate samples have a considerable effect on the moisture–density relationship for determining MDD and OMC. Commonly, an impact type compaction effort (similar to Proctor compaction) is applied on aggregate samples using the methods specified in the AASHTO T 99 Standard and AASHTO T 180 Modified test procedures (also ASTM D 698 and D 1557). The MDD values obtained from impact-hammer based methods, such as the AASHTO T 99 and AASHTO T 180, are subsequently corrected, as per AASHTO T 224, to compensate for par- ticles larger than 19.0 mm (¾ in.). Note that other laboratory compaction procedures, such as the vibratory and gyratory compaction techniques, have been shown to be more realistic for providing adequate modulus and strength in laboratory- compacted samples and simulating properly field loading and applied stress conditions under vibratory rollers (Adu- Osei et al. 2000). Although the use of vibratory compaction for establishing the compaction characteristics of granular soils is covered under ASTM D 7382, no such specification is provided by AASHTO. Kaya et al. (2012) compared the effects of two different compaction methods (impact com- paction and vibratory compaction) on the mechanical behav- ior of UAB materials. Comparing the gradation of aggregate specimens before and after compaction, Kaya et al. observed that impact compaction caused a change in aggregate grada- tion through crushing and particle breakage. This ultimately resulted in an increase in the OMC value. No such particle crushing and resulting change in gradation were observed for specimens prepared using the vibratory compaction method. Although the vibratory compaction method resulted in higher CBR values, the resilient modulus (MR) values for specimens prepared using impact compaction were consistently higher, except for one aggregate type. Density is used in pavement construction as a QC measure to help determine the compaction level of the constructed lay- ers. Holubec (1969) found that increased density improves properties of unbound aggregates with angular particles more than for aggregates with rounded particles, provided there is no increase in the pore pressure during repetitive load- ing. Generally, increasing the density of a granular ma terial makes the aggregate layer stiffer and reduces the magni- tude of the resilient and permanent deformation response to both static and dynamic loads (Seyhan and Tutumluer 2002). Although some have found the research on density to be ambiguous with regard to the resilient behavior of soils causing little change in the resilient modulus (Knutson and Thompson 1977; Elliott and Thornton 1988; Lekarp et al. 2000), others have found that there is a general increase in the resilient modulus with increasing density (Rowshanzamir 1995; Tutumluer and Seyhan 1998). The impact of density appears to be larger on the permanent deformation behavior of aggregates. Decreased density, as measured by DOC, substantially increases permanent defor- mation. Barksdale (1972) found that decreasing the DOC from 100% to 95% of maximum dry density increased permanent axial strain by 185% (on average). Increasing density from the standard Proctor to the modified Proctor maximum density decreased permanent deformation 80% for crushed limestone and 22% for gravel (Allen 1973). The DOC was reported as the most important factor controlling permanent deformation development by Van Niekerk (2002), who observed that 50% to 70% higher axial stresses were needed to cause similar mag- nitude of permanent deformation when the DOC increased from 97% to 103% for the investigated gradations (see Fig- ure 6). Note that in Figure 6, “UL,” “LL,” and “AL” refer to the finest allowable grading, the coarsest allowable grading, and the average of upper and lower limits, respectively. Moisture Content Moisture has been widely recognized to adversely affect the performance of unbound aggregate layers in pavement structures and can affect aggregates in three different ways: (1) make them stronger with capillary suction, (2) make them weaker by causing lubrication between the particles, and (3) reduce the effective stress between particle contact points resulting from increasing pore water pressure, thus decreas- ing the strength. Holubec (1969) conducted repeated load triaxial tests on crushed aggregates and gravel sands over a range of moisture contents. He reported an increase in permanent deformation by 300% for crushed aggregates and 200% for gravel sands when the moisture content was increased by 2.8% and 3.6%, respectively. Thompson and Robnett (1979) and Dempsey (1982) found that open-graded aggregates did not develop pore pressures upon loading, but uniformly graded dense

15 aggregates with higher fines contents did develop pore pres- sures that resulted in a reduction in resilient modulus val- ues. Thom and Brown (1987) found that no noticeable pore water pressures developed below 85% saturation and that most of the reduction in resilient moduli was the result of the lubricating effect of the water. It can also be assumed that increasing the water content in a soil reduces the capil- lary suction between particles, thus decreasing the effective stress and the resilient moduli. Therefore, moisture can have a positive effect on unbound granular materials as long as the moisture increases the capillary suction between parti- cles. Once the saturation reaches a point at which it reduces the capillary suction, the moisture becomes a detriment to preventing residual deformation and can cause a lubricating effect. At even higher saturation levels, where excess pore water pressure can develop and reduce the effective stress, the rutting resistance can decrease dramatically, resulting in deeper ruts (Thom and Brown 1987). Maree et al. (1982) conducted Heavy Vehicle Simulator (HVS) tests on pave- ments with untreated granular bases and reported higher permanent deformation for layers with higher moisture con- tents. Moreover, he observed that “unstable”’ conditions in unbound aggregates were triggered at lower values of stress ratio (defined as the ratio of applied stress to aggregate shear strength) when the degree of saturation was increased. Degree of saturation is a factor that reflects the combined effect of density and moisture content. The resilient modulus is strongly correlated with degree of saturation (Thompson and Robnett 1979). Based on the comprehensive subgrade soil resilient modulus testing study, Thompson and LaGrow (1988) proposed using the following “moisture adjustment” factors to adjust resilient modulus values for moisture con- tents in excess of optimum. For example, resilient modulus of a silt loam soil may decrease approximately 1,500 psi for a 1% increase in moisture content (Thompson and Robnett 1979). Wetting up from a shallow groundwater table (GWT) by capillarity or by increase in the GWT level reduces suction and may cause a constructed unbound pavement layer to deform permanently. Moisture sensitivity varies depending on specified gradations and the amount and plasticity index (PI) of the fines: that is, percent passing No. 200 sieve (P200). Tutumluer et al. (2009) compared relative impacts of mold- ing (as-compacted) moisture content and plasticity of fines on the permanent deformation behavior of crushed (dolomite) and uncrushed (gravel) aggregate materials with P200 = 12% (see Figure 7). A drastic reduction in aggregate performance can be seen when plastic fines are combined with increased molding moisture: that is, compare permanent deformation of gravel at 110% of the optimum moisture content (wopt) with plastic and nonplastic fines in Figure 7b. Accordingly, the specification limits for compaction moisture content are best based on accumulated permanent deformation. 0 50 100 150 200 250 300 350 400 Degree of compaction [%] and grading (UL-AL-LL) σ 1 [ kP a] 1%-N=1000000 171 257 130 194 180 276 125 151 234 5%-N=1000000 184 270 155 205 202 299 162 190 247 10%-N=50000 198 317 172 261 238 323 201 264 266 UL-97% UL-100% UL-103% AL-97% AL-100% AL-103% AL-105% LL-97% LL-100% LL-103% UL-upper limit AL-average limit LL-lower limit effect of DOC effect of grading for UL-AL-LL FIGURE 6 Stress (s1) levels at which ep = 1%, 5%, and 10% at N = 106, 106, and 50,000, respectively, at DOC 97%, 100%, 103%, and 105% (Van Niekerk 2002). Key Lesson The following factors have been identified as primarily affecting UAB/subbase layer performance under load- ing: (1) aggregate mineralogy, (2) gradation, (3) fines content (material passing No. 200 sieve), (4) type of fines (plastic or nonplastic), (5) particle shape, texture and angularity, (6) DOC, and (7) moisture content.

16 TESTS TO CHECK AGGREGATE QUALITY FOR PAVEMENT APPLICATIONS Background An extensive review of published literature indicates the pre- viously discussed properties are critical in governing the per- formance of UAB and subbase layers in pavement systems. Accordingly, agency specifications for aggregate usage in pavement base/subbase applications often include require- ments related to gradation (particle size distribution), degree of crushing (100% crushed, 100% uncrushed, number of fractured faces), plasticity (liquid limit and plasticity index), durability, and soundness (Barksdale 1991). Commonly used specifications include those developed by ASTM, AASHTO, the U.S. Army Corps of Engineers (USACE), and individual state and provincial transportation agencies. Based on their underlying philosophy, material speci- fications can be divided into the following four general categories: (1) methods or “recipe” specifications, (2) pro- prietary product specifications, (3) performance specifica- tions, and (4) end result specification-statistically based. Among these different specification categories, end result specifications commonly are employed for aggregate usage in pavement base/subbase layer applications. Discussions on the other specification types are presented elsewhere (Barksdale 1991). AASHTO specification M 147-65, Materials for Aggre- gate and Soil-Aggregate Subbase, Base, and Surface Courses, suggests several tests for sampling and testing of aggre- gates before their use in pavement applications. Differ- ent tests recommended by AASHTO for material quality (a) (b) FIGURE 7 Relative effects of varying moisture content and plasticity of fines on permanent deformation behavior of crushed dolomite and uncrushed gravel aggregates (Tutumluer et al. 2009).

17 testing, selection, and control testing of aggregates are listed here: • AASHTO T 2: Standard Method of Test for Sampling of Aggregates • AASHTO T 11: Standard Method of Test for Materials Finer than 75-µm (No. 200) Sieve in Mineral Aggre- gates by Washing • AASHTO T 19: Unit Weight and Voids in Aggregate • AASHTO T 27: Standard Method of Test for Sieve Analysis of Fine and Coarse Aggregates • AASHTO T 84: Specific Gravity and Absorption of Fine Aggregate • AASHTO T 85: Specific Gravity and Absorption of Coarse Aggregate • AASHTO T 88: Standard Method of Test for Particle Size Analysis of Soils • AASHTO T 89: Standard Method of Test for Determin- ing the Liquid Limit of Soils • AASHTO T 90: Standard Method of Test for Determin- ing the Plastic Limit and Plasticity Index of Soils • AASHTO T 96: Standard Method of Test for Resis- tance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Machine • AASHTO T 104: Soundness of Aggregate by Use of Sodium or Magnesium Sulfate • AASHTO T 112: Clay Lumps and Friable Particles in Aggregate • AASHTO T 113: Lightweight Pieces in Aggregate • AASHTO T 146: Standard Method of Test for Wet Preparation of Disturbed Soil Samples for Test • AASHTO T 176: Standard Method of Test for Plastic Fines in Graded Aggregates and Soils by Use of the Sand Equivalent Test • AASHTO R 58: Standard Practice for Dry Preparation of Disturbed Soil and Soil-Aggregate Samples for Test • AASHTO T 210: Aggregate Durability Index • AASHTO T 248: Reducing Field Samples of Aggregate to Testing Size • AASHTO T 255: Total Moisture Content of Aggregate by Drying Similarly ASTM specification D 2940 Standard Specifi- cation for Graded Aggregate Material for Bases or Subbases for Highways or Airports (ASTM D 2940 2009) specifies the following test methods to evaluate the quality of aggre- gates for use in pavement base and subbase layers: • ASTM D 75: Standard Practice for Sampling Aggregates • ASTM C 136: Standard Test Method for Sieve Analysis of Fine and Coarse Aggregates • ASTM D 422: Grain-Size Analysis (Wet Sieving and Determination of Subsieve Size Fractions, by Hydrom- eter Analysis) • ASTM D 4318: Standard Test Methods for Liquid Limit, Plastic Limit, and Plasticity Index of Soils • ASTM D 2419: Standard Test Method for Sand Equiva- lent Value of Soils and Fine Aggregate • ASTM D 4792: Standard Test Method for Potential Expansion of Aggregates from Hydration Reactions The following test methods have been used by agen- cies for characterizing the toughness/abrasion resistance of aggregates: • Los Angeles abrasion (AASHTO T 96) • Aggregate impact value (British) • Aggregate crushing value (British) • Micro-Deval abrasion (AASHTO T 327)—coarse and fine aggregates • Degradation in the SHRP Superpave® Gyratory Compactor Similarly, the following test methods are used to charac- terize the soundness and durability of aggregates: • Sodium and magnesium sulfate soundness tests (AASHTO T 104) • Freezing and thawing soundness (AASHTO T 103) • Aggregate durability index (AASHTO T 210) • Canadian freeze-thaw test Wu et al. (1998) evaluated different toughness/abrasion resistance as well as durability/soundness tests for character- izing aggregates used in asphalt concrete. Testing aggregates from sources with poor to good performance histories and correlating the laboratory test results with field performance, they concluded that the Micro-Deval Abrasion and Magne- sium Sulfate soundness tests provided the best correlation with field performance. The survey of state and Canadian provincial transportation agencies conducted under the scope of this synthesis study aimed to assess the state of practice in aggregate quality checking before their use in UAB and subbase layer construction. Current Practices on Tests to Check the Quality of aggregate Sources Figure 8 shows the relative distributions of different test meth- ods used by state transportation agencies to check the quality of virgin aggregates for use in UAB/subbase layers. Forty-three of 46 respondents use sieve analysis as the primary method of aggregate quality check for virgin aggregate sources. More- over, sodium sulfate/magnesium sulfate (Na2SO4/MgSO4) soundness test, some form of abrasion tests (Los Angeles abra- sion or Micro-Deval), and percent deleterious materials were also found to be common practices among agencies. Some transportation agencies also use tests, such as absorption and specific gravity, Atterberg limits, and state-specific degrada- tion tests for checking the quality of aggregate sources. Frequency of Checking Aggregate Sources for Quality The survey of state and Canadian provincial transportation agencies also gathered information on the frequencies of

18 quality assurance tests on virgin aggregate materials. Results from the survey are presented in Figure 9. Apart from the testing frequencies shown in Figure 9, several other agencies also reported policies for aggregate material quality testing based on the quantity of aggregate used in a particular project. For example, two states reported requirements for conducting at least one quality assurance check for every project per every 2,000 and 2,500 tons of aggregate used, respectively. Crushed versus Uncrushed Aggregates Particle shape and angularity, often expressed as “crushed” or “uncrushed” particles, play an important role in govern- ing the behavior of unbound aggregate layers under loading. Aggregate layers with uncrushed particles undergo signifi- cant particle reorientation under loading, thus accumulating large amounts of permanent deformation, which ultimately may lead to internal shear failure. A recent study at the Uni- versity of Illinois (Mishra 2012) evaluated the effects of particle shape and angularity on unsurfaced pavement per- formance. Through laboratory testing and accelerated load- ing of full-scale unsurfaced pavement test sections, the study highlighted the increased potential for internal shear failure within uncrushed aggregate layers. It is therefore important for transportation agencies to distinguish between crushed and uncrushed aggregates while developing material specifi- cations for aggregates to be used in base and subbase layers. Continued research on the quantification of aggregate par- ticle shape, surface texture, and angularity indices through imaging-based methods may lead to the establishment of an aggregate packing index representing the degree of particle interlock in aggregate matrices. Such a packing index poten- tially could highlight the differences between uncrushed and crushed particles as far as packing within in aggregate matrix and load transfer mechanisms are concerned. An equal number of agencies (20 each of 46 respondents) replied “Yes” or “No” when asked whether uncrushed aggre- gates were allowed in UAB and subbase layers. The remain- 57% 85% 94% 65% 48% (26) (39) (43) (30) (22) 0 10 20 30 40 0% 20% 40% 60% 80% 100% Na2SO4 / MgSO4 Soundness Test Los Angeles Abrasion and/or Micro Deval Test Sieve Analysis Percent Deleterious Materials Other (Atterberg limits, etc.) Number of Responses Percentage of Survey Respondents 46 survey respondents FIGURE 8 Tests used by agencies for evaluating quality aspects of virgin aggregate materials for pavement base and subbase applications (46 respondents). 39% 4% 2% 11% 7% 37% (18) (2) (1) (5) (3) (17) 0 10 20 30 40 0% 20% 40% 60% 80% 100% Prior to the use on major construction projects More than twice every year Twice every year Once every year Less than once every year Other (production basis, etc.) Number of Responses Percentage of Survey Respondents 46 survey respondents FIGURE 9 Frequencies of aggregate acceptance checks by state transportation agencies (46 respondents).

19 ing six agencies require partially crushed particles for base course applications (often by requiring at least one fractured face or by specifying a minimum proportion of fractured particles in the aggregate blend). Moreover, several agencies allow the use of uncrushed aggregates in subbase layers but prohibit their use in base courses. Maximum Allowable Aggregate Particle Size Particle size distribution or gradation has been found to be the most important parameter affecting aggregate perfor- mance in unbound and bound pavement layers. State and Canadian provincial transportation agencies were surveyed for the maximum aggregate particle sizes allowed in dif- ferent types of aggregate base and subbase layers, and their responses are reported in Figures 10 to 13. As shown in Figures 10 to 13, no consistent practice exists among transportation agencies regarding the maximum aggregate particle size allowed in UAB and subbase layers. Nevertheless, most of the respondents reported similar max- imum aggregate particle size limits for dense-graded base and subbase layers. For example, 32 agencies limit the maxi- mum aggregate particle size for dense-graded base courses to 1.0 or 1.5 in. Similarly, 20 agencies limit the maximum aggregate particle size in dense-graded subbase layers to between 1.5 and 2.0 in. In general, larger top-size aggregates are allowed in dense-graded subbase layers compared with those in dense-graded base layers. No such clear trend was observed from comparing the specifications for open-graded drainage base and subbase layers. Type and Amount of Fines The type of fines, often indicated by the PI value (PI test usually conducted on material finer than 0.425 mm or pass- ing No. 40 sieve), plays an important role in governing the shear strength, resilient modulus, and permanent deformation behavior of unbound aggregate layers in pavement structures. As mentioned, unless otherwise specified, the term “fines” in the current synthesis refers to material finer than 0.075 mm or passing the No. 200 sieve. Aggregate materials with high amounts of plastic fines exhibit higher moisture susceptibility and undergo significant reduction in the shear strength in the presence of moisture when compared with aggregates with nonplastic fines. Recent research at the University of Illinois 1 11 21 1 4 2 1 1 0 5 10 15 20 25 N um be r o f A ge nc ie s Maximum Allowable Aggregate Particle Size FIGURE 10 Base layer dense-graded unbound aggregate: maximum aggregate particle size allowed by agencies. 1 1 1 11 1 8 1 5 2 1 1 2 0 2 4 6 8 10 12 14 N um be r o f A ge nc ie s Maximum Allowable Aggregate Particle Size FIGURE 11 Subbase layer dense-graded unbound aggregate: maximum aggregate particle size allowed by agencies.

20 (Mishra et al. 2010a, 2010b; Mishra and Tutumluer 2011; Mishra 2012) has established the increased moisture suscep- tibility of aggregates with high amounts of plastic fines, and thus has emphasized the importance of specifying different values for the maximum allowable fines contents for aggre- gates with nonplastic and plastic fines. Accordingly, the cur- rent synthesis study gathered information on the state of the practice regarding the maximum amounts of nonplastic and plastic fines allowed in an aggregate matrix. Only one agency (Maryland) currently specifies different threshold limits for the maximum amount of plastic and nonplastic fines allowed in aggregates to be used in pavement construction. It is impor- tant to note that some agencies may not consider differentiat- ing between plastic and nonplastic fines. This is because with adequate production, storage, and construction practices the amount of plastic fines in an aggregate material usually can be controlled. For crushed stones produced from quarry opera- tions, the fines (material passing No. 200 sieve) usually are nonplastic in nature although the nature of fines also depends on the mineralogy of the parent rock. With proper storage, sampling, and transportation practices, contamination of the stockpiles with natural soil and corresponding plastic fines can be controlled adequately. Therefore, aggregate materials used in construction of UAB/subbase courses may primarily comprise nonplastic fines. Accordingly, the lack of differen- tiation between plastic and nonplastic fines in UAB/subbase courses may not always indicate poor practice. Rather, it should be emphasized that in cases in which the aggregate material may comprise plastic fines, it is critical to control the maximum amount of fines allowed in the aggregate matrix because plastic fines in the presence of moisture often lead to significant deterioration in the aggregate shear strength. As discussed, researchers and practitioners in the past have established that unbound aggregate materials contain- ing the optimum amount of fines (material passing No. 200 sieve or finer than 0.075 mm) perform the best as far as shear strength, resilient modulus, and permanent deformation characteristics are concerned. Insufficient fines in an aggre- gate matrix results in unstable matrix behavior because of 1 3 4 2 2 1 1 1 1 0 1 2 3 4 5 N um be r o f A ge nc ie s Maximum Allowable Aggregate Particle Size FIGURE 12 Base layer open-graded drainage: maximum aggregate particle size allowed by agencies. 1 3 4 2 1 1 1 1 1 0 1 2 3 4 5 N um be r o f A ge nc ie s Maximum Allowable Aggregate Particle Size FIGURE 13 Subbase layer open-graded drainage: maximum aggregate particle size allowed by agencies.

21 the excessive movement of the coarse particles with respect to each other. On the other hand, the presence of excessive fines in an aggregate matrix compromises particle interlock through lubricating action at the contact points. This leads to the aggregate material exhibiting lower shear strength and resilient modulus values and accumulating large permanent deformations. Thus, it is critical for transportation agencies to control the amount of fines in aggregates used in pavement applications. Figures 14 to 16 summarize the survey responses con- cerning the maximum amount of fines allowed in different UAB and subbase layer types. There is a wide variation in the maximum allowable fines contents from one agency to the other. Although 33 of 46 respondents restrict the amount of fines (P200) in dense-graded base courses to less than 12%, five agencies reported allowing more than 15% fines. One agency (Georgia) currently specifies different limits for the maximum allowable fines (P200) contents for different aggre- gate types (11% and 15% for silicate and carbonate rocks, respectively). Another agency is in the process of modifying its state specifications to impose a restriction on the maxi- mum allowable fines content. Figures 17 and 18 show the maximum value of PI allowed by agencies for the fines fraction (P200) in dense-graded UAB and subbase courses, respectively. As shown in the two fig- ures, PI = 6 is commonly used as the maximum PI value for the fines fractions in UAB and subbase layers. Moreover, it is important to note that three agencies do not impose any restrictions on the plasticity of fines in the aggregates used for constructing aggregate base and subbase layers. In addi- tion, one transportation agency allows the use of aggregates with fines fraction PI as high as 15. Such high plastic fines, when present in large amounts in an aggregate matrix, may render the aggregate highly moisture susceptible, thus sig- nificantly reducing the aggregate shear strength in the pres- ence of moisture. 2 1 1 1 6 1 8 2 11 1 2 2 1 1 4 0 2 4 6 8 10 12 N um be r o f A ge nc ie s Maximum Allowable Fines Content (%) FIGURE 14 Base layer dense-graded aggregate base: maximum allowable fines content in aggregates used (44 respondents). 2 1 4 1 7 3 1 6 2 1 1 1 5 0 2 4 6 8 6 7 8 9 10 12 13 15 18 20 25 34 Other N um be r o f A ge nc ie s Maximum Allowable Fines Content (%) FIGURE 15 Subbase layer dense-graded aggregate subbase: maximum allowable fines content in aggregates used (35 respondents).

22 1 4 2 4 1 1 1 1 1 0 2 4 6 N um be r o f A ge nc ie s Maximum Allowable Fines Content (%) FIGURE 16 Base/subbase open-graded drainage layers: maximum allowable fines content in aggregates used (17 respondents). *Other: PI = 6 or graded aggregate base and Coquina base; PI = 9 for sand-clay base 5 2 1 1 18 1 1 1 1 1 1 3 0 4 8 12 16 20 N um be r o f A ge nc ie s Maximum Allowable Plasticity Index (PI) FIGURE 17 Maximum value of PI allowed for the P200 fines fraction in dense-graded UAB. 5 1 3 11 1 1 1 1 3 0 3 6 9 12 15 0 4 5 6 8 9 12 15 No Spec N um be r o f A ge nc ie s Maximum Allowable Plasticity Index (PI) FIGURE 18 Maximum value of PI allowed for the P200 fines fraction in dense-graded UAB.

23 SUSTaINaBLE PRODUCTION aND UTILIZaTION OF aGGREGaTES Aggregate crushed stone quarry processes such as blast- ing, crushing, and screening of coarse aggregates produce by-product materials, at approximately 8% of the mined aggregate, commonly known as quarry waste or quarry dust. Quarry wastes typically are less than 0.25 in. (6 mm) in size and consist of coarse, medium, and fine sand-sized particles and a clay/silt-sized fraction, which is less than No. 200 sieve (0.075 mm) in size. Current economic conditions and an increased emphasis in the construction industry on sustain- ability and recycling require production of aggregate gra- dations with lower dust and smaller maximum sizes. These new production limitations have “unbalanced” the aggregates production stream, in part because of the demand for cleaner aggregates with smaller top sizes in increased utilization of finer asphalt concrete mixes, resulting in excessive energy use and increased waste fines. Owing to these increased energy and disposal costs for the aggregate production, re using and recycling of waste products (e.g., RAP, and RCA) may some- times exceed the potential economic benefits. If research can lead toward more beneficial use of the by-product quarry wastes, making more effective use of locally available ma- terials through beneficiation and use of marginal aggregate materials and increasing effective use of recycled materials, aggregate production can become more sustainable through energy conservation and efficient use of aggregate resources. Mineralogical properties of the parent rock and the type of crusher greatly influence the amount of fines produced dur- ing quarry operations. From a report published in the United Kingdom, depending on the type of crusher, limestone quar- rying may produce as much as 25% fines. Filler materials that are less than 0.075 mm in size typically account for 10% to 20% of the total crushed rock aggregate fines produced (Manning and Vetterlein 2004). A more recent review article from Stone, Sand and Gravel Review on utilization of quarry fines for sustainable construction cites a number of successful applications for alternative uses of quarry fines in construc- tion applications (Halmen and Kevern 2010). Among them, the ones linked to pavement applications with the most poten- tial to utilize large amounts of quarry fines are listed in the order of ranking as follows: 1. Pavement working platform, also known as aggregate subgrade/granular subbase construction; 2. Unbound aggregate for road base and as embankment material; 3. Fine aggregate/filler for controlled low-strength materi- als (aka flowable fill); 4. Filler material in hot mix asphalt (HMA) and slurry seals for asphalt pavements; 5. Fine aggregate/filler for PCC, such as use of manufac- tured sand in PCC with higher fines content accord- ing to the ongoing work at the ICAR by Fowler and co-workers (http://aftre.nssga.org/Reports/Project- 102-1.pdf). FHWA also published and updated user guidelines for use of by-products in pavement construction, in which they identified flowable fill as a possible application for the use of quarry fines among many other applications (FHWA 2008). Several studies in the literature report successful utiliza- tion of quarry fines in road base/embankment and flowable fill applications. Preliminary studies conducted at the Uni- versity of Texas indicated that quarry fines could be used economically as flowable fill and in cement-treated pavement subbases (Kumar and Hudson 1992; Hudson et al. 1997). Based on these findings, a recent ICAR research study tested the acceptability of high fines content in aggregate pavement layers and reported that aggregate systems with higher fines benefited considerably from low percentages (1% to 2%) of cement stabilizer (Ashtiani and Little 2007). The study found that with the proper design of fines content, cement content, and moisture, the performance of the stabilized systems with high fines content could perform equivalent to or better than systems with standard fines content. Cement-treated quarry fines also were used as a pavement base material on SH-360 in Arlington, Texas, as part of a research project (Puppala et al. 2008). The study reported that the unconfined compres- sive strength of cement-treated quarry fines was adequate Key Lessons • No common practice exists among transportation agencies as far as testing of aggregates for material quality before their use in unbound base/subbase layers is concerned. • The use of 100% uncrushed aggregates in UABs/ subbase layers is best done with caution, taking into account their substandard strength properties and high rutting damage potentials. • No common practice exists concerning the maximum aggregate particle size allowed in different unbound aggregate pavement layers. To ensure ease of con- struction and adequate compaction, the maximum particle size allowed in UAB, subbase, and drainage layers are best restricted to 1.5 in., 2 in., and 4 in., respectively. • Excessive fines (P200) deteriorate aggregate layer performance, especially in the presence of mois- ture. The maximum amount of fines (P200) allowed in UAB/subbase layers are best restricted to 12% or less. • The presence of plastic fines in an aggregate layer needs to be limited. For instances in which the pres- ence of plastic fines is unavoidable, different threshold limits are best set for the maximum allowable fines content for nonplastic and plastic fines.

24 and that field monitoring indicated low permanent deforma- tion during service. A recent Iowa DOT study also focused on road construction using admixture stabilized limestone fines and found that stabilized fines could perform satisfac- torily as a structural layer in road construction through visual observations (Rupnow et al. 2010). Laboratory compaction, unconfined compression, freezing and thawing, and wet-dry durability test results showed that cement kiln dust (CKD) was not an acceptable stabilizer because of poor durability performance; however, class C fly ash and CKD mixtures were determined to be acceptable. Mishra and Tutumluer (2011) characterized the strength, stiffness, and deformation behavior of different types and qual- ities of aggregates commonly used in Illinois for subgrade replacement and subbase. The project focus was on estab- lishing aggregate cover thickness correlations with aggregate material properties to modify and improve the current IDOT Subgrade Stability Manual thickness requirements through laboratory and field testing. Thick layers of the uncrushed gravel placed over a weak subgrade were observed to undergo internal shear failure owing to high amount of fines and exces- sive movement of the aggregate particles. On the other hand, crushed aggregate layers showed significantly higher resis- tance to internal shear deformation, and test sections con- structed using crushed aggregates failed primarily as a result of subgrade deformation. The influence of compactive effort on aggregate layer performance was clearly apparent; higher relative compaction exhibited better resistance to permanent deformation accumulation. Prolonged exposure to moisture and freeze-thaw effects was found to have a beneficial effect on the crushed dolomite with high amounts (12% to 13%) of nonplastic minus No. 200 fines (Mishra and Tutumluer 2011). Interestingly, carbonate cementation within the fine fraction was identified as the most probable mechanism contributing to “stiffening” of the aggre- gate sections, which resulted in the aggregate layer sustaining a significantly higher number of load applications without undergoing shear failure. Note that upon further loading, the aggregate sections demonstrated “punching” failure into the underlying subgrade (CBR = 3%) similar to the failure of concrete slabs over weak support conditions. Unbound Pavement applications Often different test methods are adopted by transportation agencies to check the adequacy of aggregate materials for use in unbound pavement applications. However, these “recipe- based” test methods are focused primarily on checking the physical and chemical (if applicable) properties of aggre- gates, which often are related to basic geologic origin, miner- alogy, and other properties, such as hardness and durability, and they may not necessarily offer the best means to judge the mechanistic properties and performance of unbound aggre- gate layers. One major disadvantage associated with such physical classification systems is that they could possibly accept unsuitable materials in some cases and reject desir- able materials in other cases, as summarized by Cook and Gourley (2002). Under such a physical classification frame- work, naturally occurring materials could be excluded from use, based on any combination of grading, plasticity, particle hardness, strength, and so forth lying outside the specification- demanded requirements, as outlined in Figure 19. In many areas with a shortage of “standard” or traditional aggregate materials that satisfy normal requirements for road paving, nonstandard local aggregate sources have been suc- cessfully applied in low volume road constructions; typical examples are documented in Table 1. In addition, an early field trial constructed by the Transportation Research Labo- ratory in 1978, in which three marls (local calcareous mate- rials) outside the recommended gradation envelope were substituted for the crushed stone base, indicated that the use of a much wider range of marls, if properly stabilized, is viable technically and economically, as justified by the low values of rut depth and deflection and the high strength of the base (Woodbridge 1999). Bullen (2003) also showed that the use of local aggregate materials in Australia, with appro- priate design, can not only provide the desired pavement performance, but also can promote sustainability in terms of significant cost saving, natural resource conservation, and even environment protection. In the United States, for instance, the taconite aggregate resources in Minnesota, the industrial by-products from iron ore mining, recently have been demonstrated in MnROAD low-volume test section studies to be a promising supply of high-quality, low-cost aggregates for roadway use (Clyne et al. 2010). In Texas, locally available materials (mostly Grade 4), sometimes even with high amount of fines, have been used (with or without stabilization) not only for low- volume roads but also for major roads in some districts. Despite all of the potential benefits and documented suc- cessful applications of local aggregate sources, one major obstacle to their widespread use is the significant engineer- ing uncertainty (or risk) inherent with their long-term per- formance. Such uncertainties cannot be addressed by current physical classification systems to be later considered properly in pavement design. Furthermore, several state transportation agencies are reluctant to relax the traditionally conservative standard specifications. Separate from the physical classification presented pre- viously, the mechanistic classification discerns different qualities of unbound aggregates from mechanical properties that are required as input to the constitutive relationships incorporated into mechanistic-empirical pavement design procedures, as illustrated in Figure 20. It is expected that such mechanistic classification systems, in combination with certain levels of local experiences, have direct relevance or even robust linkage to the actual performance of materials used

FIGURE 19 Nonstandard material groups and their likely problems (Cook and Gourley 2002).

26 Source: Cook and Gourley (2002). TABLE 1 ExAmpLEs of Using nonsTAndArd mATEriALs in Low-VoLUmE sEALEd roAds FIGURE 20 Physical (left) versus mechanical (right) classification for various unbound granular materials (Paute et al. 1994).

27 in pavement layers. The mechanistic nature of the responses of unbound aggregate materials can be characterized by resilient modulus (stiffness), whereas permanent deforma- tion linked to shear strength often relates to rutting damage accumulation. Because both the resilient (recoverable) and permanent deformation/strain components are to be considered simul- taneously for mechanistic-empirical (M-E) evaluation of unbound aggregate behavior, the resistance to permanent deformation under repeated traffic loading relates to rutting damage accumulation in unbound aggregate materials. For example, the Australian Road Association determines both resilient modulus and permanent deformation from repeated load triaxial tests to characterize unbound aggregates and marginal materials (Austroads 2003). Khogali and Mohamed (2007) developed a mechanistic aggregate classification sys- tem based on a test procedure for combined determination of the resilient modulus and permanent deformation poten- tial involving both elastic and plastic responses. Recently, Tao et al. (2010) introduced a mechanistic-based design approach to characterize and compare the behavior of tradi- tional and recycled pavement base materials that employed dissipated energy concept to explain different shakedown responses of materials obtained from laboratory repeated- load triaxial tests and full-scaled accelerated loading tests. It was implied that permanent deformation characteristics of pavement materials provided a better measure for evaluating recycled and marginal materials against traditional unbound aggregates. Shear strength is an important mechanistic property of unbound aggregate materials. The shear resistance of the material mainly contributes to developing a load resistance quality that greatly reduces the stresses transmitted to the underlying layers (Garg and Thompson 1997). Saeed et al. (2001) found under the NCHRP Project 4-23 study that shear strength of unbound aggregates under repeated loading had the most significant influence on pavement performance. Seyhan and Tutumluer (2002) suggested that a limiting value of the shear stress ratio (the level of applied shear stress as a fraction of the shear strength of the material) controlled the permanent deformation behavior of aggregates and that “good” quality aggregates typically had low shear stress ratios in the range of 0.2 to 0.5. To better assess performance and rank different sources of aggregate materials, coupling mechanistic characteristics, including moduli, strength, and permanent strains, under representative ranges of operating environmental conditions is of essential importance from the MEPDG perspective. Development of performance-based material specifications is critical for optimized material use with reduced waste and eventually better utilization of construction dollars. From a MEPDG perspective, determining the best use of different qualities of locally available aggregate materials in road bases/subbases may be a challenge. For example, Lukanen (1980) found early on that certain MnDOT) Class 3 aggregates were even stronger than Class 6 aggregates when placed in pavement granular layers. This was a surpris- ing field evaluation considering that as MnDOT aggregate classes increase, usually better materials, such as a high- quality Class 6, are designated. During MnROAD study, similar contradictory trends were observed in backcalculated base layer moduli from falling weight deflectometer (FWD) testing of flexible pavements (Ovik et al. 2000). For both thin (<15 cm) and thick (>15 cm) asphalt concrete surfac- ing, the backcalculated base moduli of Class 3sp materials often were found to be greater than those of higher material classes (i.e., 4sp, 5sp, and 6sp) (Ovik et al. 2000). In light of these findings, several issues may need to be addressed, such as how to specify material properties based on their end-use performances; where in pavements to place locally available materials (either natural or recycled) of marginal quality; what type of pavements and critical traffic design levels should be determined beyond which no satisfactory pavement performance can be cost-effectively maintained by using marginal materials; and finally, what would be the optimum combination of high- and marginal-quality aggre- gate uses considering certain design features and site factors so that aggregate base and granular subbase materials can be optimized for satisfactory pavement performance. Key Lessons • The use of quarry by-products and other marginal aggregates in UAB/subbase layers can lead to sus- tainable pavement construction practices. • In addition to commonly used tests for evaluating the physical characteristics, the mechanical perfor- mance of such marginal aggregates needs to be carefully studied. • Currently used performance-based tests often fail to adequately evaluate unbound aggregate materials for application in pavement base/subbase layers. Additional research is required toward the develop- ment of new and modification of existing performance- based tests. • Marginal aggregates and quarry by-products may be mixed with high-quality aggregates to develop mate- rial blends with adequate physical, chemical, and mechanical properties. • It is not uncommon for a “weak” rock, such as lime- stone, to show very high resilient modulus values. However, the permanent deformation behavior also is best evaluated before classifying the material as a “good quality” aggregate for use in unbound base/ subbase layers.

28 Best Value Granular Material Concept Continual depletion of available natural aggregate resources has led an increasing number of transportation agencies to haul aggregate material for use in pavement construction from long distances. Such long distance hauling significantly increases the material cost for aggregates. According to NSSGA, transporting aggregates by truck over a distance of 30 to 50 miles can double the material cost for the end user. A 1998 USGS study indicated that for an assumed 56-km (35-mile) transportation distance, the cost of transporting aggregate materials for use in pavement base/subbase layers may exceed the estimated purchase price of the product at the source (Wilburn and Goonan 1998). Thus, more empha- sis is placed by transportation agencies on the utilization of locally available “best value” granular materials, which do not require hauling aggregates from sources farther away from the project locality and incurring significant material hauling and transportation expenses. Best value granular materials are locally available aggre- gate materials (natural or recycled) that can be used in pave- ment construction through slight modification to the design and/construction procedures. Various locally available aggre- gate materials currently are classified as “out of specification” according to traditional “recipe-based” testing techniques and specifications, but still there likely is significant opportunity for better value to be achieved. These materials may not sat- isfy all the requirements specified by transportation agencies for quality assurance of aggregates used in pavement appli- cations. However, through slight modifications to the design and/or construction procedures, these materials can be used in pavement applications and thus will significantly reduce the overall construction cost and energy expenditure. This is particularly true for low-volume road applications for which the traffic volume is sufficiently low to allow the use of these “marginal quality” materials without significantly affecting the pavement performance under loading. This “sustainable” alternative has garnered significant attention from different transportation agencies, and more attention is being paid to better utilization of best value gran- ular materials, the use of which reduces the cost and energy associated with material hauling. A recently completed research study sponsored by MnDOT conducted mechanistic–empirical pavement analyses to evalu- ate the performances of pavement structures with base/subbase layers constructed with locally available aggregate materials (xiao et al. 2011; xiao and Tutumluer 2011). The primary objective was to demonstrate that locally available aggregate materials could be economically efficient in the implementa- tion of available mechanistic-based design procedures. Find- ings from the study indicated that for low-volume roads, base and subbase quality was not significant for achieving 20-year fatigue and rutting performance lives. Thus, for low-volume roads, using locally available and somewhat “marginal” materi als was a significantly cost-effective alternative. How- ever, for traffic volumes greater than 1.5 million equivalent single-axle loads (ESALs), aggregate material quality was crit- ical in governing fatigue and rutting performances. Note that these findings may need to be verified in the field before being implemented into pavement design and construction practices. In addition, the study found that a change in the subbase material quality had a more significant impact on pavement rutting performance than did a similar change in the base material quality. When the base quality was decreased from high to low, its effect on rutting performance was almost neg- ligible for pavements with design traffic levels between 0.6 and 6.0 million ESALs. However, a similar drop in subbase material quality resulted in significant reduction in the rut- ting life. Accordingly, based on the research findings, for a pavement structure comprising “good quality” aggregates in the subbase, locally available “marginal” aggregates may be used in the base layer, while ensuring adequate pavement performance. A high-quality, stiff subbase exhibits almost a bridging effect to better protect the subgrade and offset any detrimental effects of low base stiffness, and as a result, the quality of base materials becomes trivial. Note that this is the same concept as that used in the South-African “inverted pavement” designs, which often use a cement-stabilized sub- base over soft soils to effectively protect the subgrade while providing a stiff underlying layer for the base to enable com- paction of granular base materials, often in excess of 100% of achieved Proctor densities. Figure 21 presents the concept of best value granular materials illustrated as an implementation challenge of recent research study findings (xiao et al. 2011; xiao and Tutumluer 2011). Three components were proposed for incorporation into the MnDOT’s mechanistic–empirical pavement analysis and design program MnPAVE to implement the best-value granular material aggregate selection, utilization, and mechanistic-based design concepts: (1) geographic-information system-based aggregate source management component, (2) aggregate prop- erty selection component for design, and (3) aggregate source selection/utilization component. To accomplish pavement designs, aggregate material source locations are identified with certain aggregate properties to be linked to mechanis- tic MnPAVE pavement analysis property inputs. The quality aspects of the used aggregates are then assessed for cost- effectiveness and unbound aggregate layer design thickness requirements for a sustainable pavement performance. Key Lesson The concept of best-value granular materials involves the use of locally available aggregates (natural or re- cycled) in pavement construction through slight modi- fications to design/construction procedures.

29 RECYCLING AGGREGATES AND RECYCLED GRANULAR MATERIALS Fluctuating oil prices in recent years have magnified the importance of building sustainable pavement systems with stronger and less moisture-susceptible unbound granular lay- ers as the primary load-bearing structural components. High construction demands and accompanying geologic restric- tions imposed by urbanization and environmental concerns have resulted in a scarcity of good-quality aggregate sources in many locations. As a result of this and sustainability issues, increased amounts of recycled or reclaimed aggregates are used to supplement virgin aggregate supplies. The FHWA lists the following recycled aggregate types as being used by different agencies in pavement granular base layer applications (http:// www.fhwa.dot.gov/publications/research/infrastructure/ pavements/97148/004.cfm): (1) blast furnace slag, (2) coal bottom ash, (3) coal boiler slag, (4) mineral processing wastes, (5) municipal solid waste combustor ash, (6) nonferrous slags, (7) reclaimed asphalt pavement, (8) reclaimed concrete, (9) steel slag, and (10) waste glass. According to the USGS, the highway industry (American Concrete Pavement Associa- tion, Construction Materials Recycling Association, FHWA, and National Asphalt Pavement Association) has estimated the quantities of reclaimed and recycled asphalt and concrete materials used in construction at closer to 100 million tons each in 2009; approximately 14 and 18 million tons were used as RCA and RAP aggregate materials, respectively. Of the previously mentioned material types, only the fol- lowing three recycled materials were considered in this syn- thesis study: (1) unbound aggregate materials recycled from old pavement base/subbase layers, and recycled materials: that is, (2) RAP and (3) RCA. Consideration was given in this synthesis to the different test methods used by state transpor- tation agencies to check the adequacy of recycled materials before allowing their use in the construction of UAB and subbase layers. Particular emphasis was given to whether state transportation agencies impose additional requirements for quality assurance of recycled aggregate materials com- pared with those for virgin aggregates. The following properties of recycled aggregates have been identified by NCHRP Report 598 as relevant to their use in unbound pavement layers (Saeed 2008): (1) shear strength, (2) CBR, (3) cohesion and angle of internal friction, (4) resil- ient or compressive modulus, (5) density, (6) permeability, (7) frost resistance, (8) durability index, and (9) resistance to moisture damage. AASHTO specification PP 56-06, Evaluating the Engi- neering and Environmental Suitability of Recycled Materi- als, outlines a framework for assessing the feasibility to use recycled materials in the highway environment by consider- ing issues such as (1) engineering and material properties; (2) environmental, health, and safety properties; (3) imple- mentation aspects; and (4) recycling aspects of the recycled materials. Although the specification recommends a general framework to be adopted before the use of recycled materials, it also clearly recommends the evaluator consider local condi- tions before selecting different criteria and the corresponding threshold values. FIGURE 21 Best value granular material MnPAVE design implementation (Xiao and Tutumluer 2011).

30 In addition to the general evaluation framework listed in AASHTO PP 56-06, agencies may sometimes adopt the tox- icity characteristic leaching procedure (TCLP) to chemically evaluate the potential harmful effects of leaching through an UAB/subbase layer constructed using recycled materials. Designated as Method 1311 by the EPA (http://www.epa. gov/osw/hazard/testmethods/sw846/pdfs/1311.pdf), TCLP determines the mobility of both organic and inorganic haz- ardous materials in recycled and waste materials. For exam- ple, the Georgia Department of Transportation (GDOT) lists the TCLP as a required “acceptance” test for recycled con- crete base aggregates for sources of RCA that are not from GDOT projects or GDOT pavements. The ongoing Transportation Pooled Fund Study TPF- 5(129), Recycled Unbound Pavement Materials (http://www. pooledfund.org/Details/Study/361), has the objective of mon- itoring the performance of several test cells at MnROAD con- structed using recycled materials in the granular base layers, including blended with virgin materials and 100% recycled asphalt and concrete pavement materials. Issues that are being considered include variability in material properties, purity of material, and how to identify and control material quality. The project findings will include laboratory studies, examination of existing field sites, and evaluation of data from MnROAD test sections. Anticipated results from this project include a suite of tests and/or protocols that may be used to identify the critical characteristics of these recycled materials, as well as optimum design criteria and best construction practices needed for a durable base that meets the properties proposed for layer design. Reclaimed asphalt Pavement Since a principal constituent of RAP is its mineral aggregates, the overall chemical composition of RAP is similar to that of the mineral aggregates. Asphalt cements constitute only a minor percentage of RAP. The principal elements in asphalt cement molecules are carbon and hydrogen. Other materi- als, such as sulfur, nitrogen, and oxygen, usually are pres- ent in very small amounts. Asphalt cements are made up of asphaltenes, resins, and oils. Upon oxidation, the oils con- vert to resins and asphaltenes, in which the resins convert to asphaltene-type molecules, resulting in age hardening and a higher viscosity binder (Roberts et al. 1996). This change in the chemical composition would influence the unbound layer stiffness and shear strength and, consequently, its perfor- mance parameters, such as rutting and fatigue cracking. RAP can be used as granular base or subbase material in pavement structures (e.g., Garg and Thompson 1996; Maher and Popp 1997; Bennert et al. 2000; Chini et al. 2001). Garg and Thompson (1996) conducted a field testing research pro- gram to investigate the potential of using RAP as a pavement base. This study demonstrated that the performance of the RAP base was comparable to that of a crushed stone base. According to Stroup-Gardiner and Wattenberg-Komas (2013), when RAP is used as an aggregate in an unbound appli- cation, the volume of asphalt in the RAP reduces the specific gravity, and the presence of asphalt seals most of the surface area of the particles. These characteristics result in a lower unit weight and a reduced amount of water needed to achieve the desired compaction level. A study by Taha et al. (1999) recom- mends blending granular RAP with virgin aggregates to attain the proper bearing strengths because the RAP-bearing capacity usually is lower than that of conventional granular aggregate bases. As conventional granular aggregate content increased, dry density and CBR values increased (Taha et al. 1999). There- fore, it is important to characterize and quantify the expected range of RAP properties before application. Findings from the ongoing Transportation Pooled Study TPF-5(129) have indi- cated that although RAP materials may show high resilient modulus values, aggregate layers constructed using 100% RAP materials often accumulate high permanent deformation values. The degree of expansion for the RAP materials is not well known. Expansion of the RAP material is particularly critical when the RAP contains expansive components, such as steel slag, which may not be commonly allowed in pavement base/ subbase layers. Note that steel slag aggregates often are used in HMA surface courses where their high frictional charac- teristics are particularly useful. Therefore, any RAP material obtained from these surface courses with steel slag aggregates potentially may lead to expansion and resulting pavement heave when used in UAB/subbase courses. Recent experi- ences with volume changes of 10% or more have been attrib- utable to hydration of the calcium and magnesium oxides in the recycled steel slag aggregate when water was encountered in the pavement base layer (Collins and Ciesielski 1994). The free lime hydrates rapidly and can cause large volume changes over a relatively short period of time (weeks), whereas magne- sia hydrates much more slowly and contributes to long-term expansion that may take years to develop. The potential expan- sion depends on the origin of the slag, grain size and gradation, and the age of the stockpile (Rohde et al. 2003). Deniz et al. (2010) studied the expansive properties of 17 RAP materials, including recycled steel slag aggregates, with respect to those of the virgin aggregates in the laboratory following the ASTM D4792 Potential Expansion of Aggregates from Hydration Reactions test method. The RAP materials had much lower tendencies to expand than did the virgin steel slag aggregates, most likely owing to an effective asphalt coating around the aggregate that prevents any significant ingress of water into the aggregate. Depending upon the level of expansion and the material gradation, dense-graded aggregate base applications with steel slag under pavements and structures may have to be avoided. Recycled Concrete aggregate Kuo et al. (2001) investigated the feasibility of using RCA as a base course material in asphalt pavements. Through literature

31 review, laboratory testing, accelerated performance testing and pavement distresses monitoring, FWD testing, and theo- retical analysis of pavements, Kuo et al. developed the fol- lowing specifications for use of RCA in Florida (see Table 2). According to the American Concrete Pavement Association (2008), RCA typically is highly angular and has a higher water absorption capacity, lower specific gravity, lower strength, and lower abrasion resistance than do conventional construc- tion aggregates. Recommendations provided for the design, construction, and QC of RCA bases and subbases in pavement applications were reviewed by Stroup-Gardiner and Wattenberg- Komas (2013). Pavement design needed to consider the stiffer RCA layer properties compared with unstabilized base materi- als, which was a function of the additional hydration associated with the RCA materials. The stiffening effect was enhanced when using dense gradations with high RCA fines (minus 4.75 mm size) contents. Properties that influenced the per- formance of RCA base materials for pavements included aggregate toughness, frost susceptibility, shear strength, and stiffness. Recommendations for QC/QA testing include Micro-Deval (AASHTO T327), tube suction, static triaxial (AASHTO T234), repeated load testing, and resilient modulus. Unbound RCA bases might limit the fines (minus 4.75 mm size) content to prevent clogging drainage features and might be used below the drainage systems. Stabilizing the RCA could bind excess fines. Type of Test Proposed Specifications Gradation test Sieve No. 50 mm 37.5 mm 19 mm 9.5 mm #4 #10 #50 #200 Gradation limits (90% confidence interval) 100 98–100 65–100 40–83 27–63 20–49 8–24 2–6 Limerock bearing ratio Test Minimum 120 LA abrasion loss <48% Sodium sulfate test <5% Sand equivalent >70% Heavy metals 5 ppm Asbestos Free of asbestos Optimum moisture content 10%–12% Maximum dry unit weight 108–120 pcf Permeability 0.10–1.40 (ft/day) Impurities <2.0% by weight Structural coefficient 0.30 Thickness requirement Minimum 8.0 in. (20.4 cm) Thickness equivalency 1.0 in. (2.54 cm) Source: Kuo et al. (2001). TABLE 2 PROPOSED SPECIFICATIONS FOR USE OF RCA IN UNBOUND AGGREGATE BASE LAYERS Key Lessons • Extensive testing of locally available natural and recycled aggregates to characterize their shear strength, resilient modulus, and permanent defor- mation behaviors will enable optimum use of these materials and limit material hauling and transporta- tion costs. • RAP materials are best tested in the laboratory for resilient modulus and permanent deformation behav- ior before being used in UAB/subbase layers. Sev- eral studies have reported high resilient modulus Recently, Ooi et al. (2011) recommended the following practices when using RCA as a base course: (1) allow only uncrushed concrete that can be visually inspected to be used as RCA, (2) accept RCA only from suppliers who can guar- antee the quality, (3) RCA from unknown sources should not be accepted unless certified by a qualified engineer or sci- entist as being free of deleterious materials (such as alumi- num), (4) avoid using building demolition RCA, (5) require a paper trail to document the RCA source, (6) use a nonferrous metal detector to determine whether aluminum is present and inspect the RCA visually before use.

32 POTENTIaL ENVIRONMENTaL IMPaCTS FROM USING RECYCLED MaTERIaLS The potential environmental impact from using recycled materials in UAB and subbase layers remains a concern for transportation agencies. Although the environmental con- cerns regarding the use of RAP and RCA in unbound aggre- gate layers are not as pronounced as are those associated with the use of some other recycled materials, such as fly ash and silica, several transportation agencies require these materials to meet environmental quality requirements (Saeed 2008). There is a concern that RAP and RCA used in unbound aggregate pavement layers, when subjected to intermittent wetting and drying, may leach contaminants into the ground- water. This is particularly true for RCA because it presents a potential for leaching of residual hydroxyl (OH-) ions from the cement paste, thus raising the pH of groundwater. More- over, concrete that already has been subjected to alkali-silica reaction and sulfate attacks may involve deleterious expan- sion that adversely affects the performance of UAB and sub- base layers (Rathje et al. 2002). RAP also has certain environmental implications with respect to the potential for contamination of ground and sur- face water systems. The aromatic compound is an organic compound of asphalt cement that has become a great con- cern because the levels of this aromatic hydrocarbon pres- ent in asphalt cement could exceed published soil clean-up standards available in several states. Concerns about the use of RAP are also addressed in NCHRP Report 443: Primer Environmental Impact of Construction and Repair Materi- als on Surface and Ground Waters, which was prepared from NCHRP project 25-09. Most binder treatment, required for mechanical reasons, significantly modifies the leaching behav- ior of a recycled material. Sadecki et al. (1996) analyzed the leachate water from three experimental stockpiles made from coarse concrete, fine concrete, and salvaged bituminous ma- terial obtained from pavement millings, respectively. Follow- ing EPA-approved methods for analyzing the leachate quality, they observe that the pH value exceeded the Minnesota stan- dards, whereas the chromium content may have exceeded the standard sometimes. The polynuclear aromatic hydrocarbon concentrations from the bituminous piles often were near or below detectable limits. Measured parameters, such as alka- linity, chloride, sodium, potassium, total solids contents, and the conductivity were, in general, higher for the concrete piles than for the bituminous pile. They suggested that the potential impact of stockpile runoff on groundwater can be controlled with proper management of stockpile locations. A study by Hill et al. (2001) has shown that the binder treatment may dilute or amend leachable levels, alter the pH, and reduce the permeability. However, the addition of alkali binder could introduce some contaminants, such as calcium. The study by Hill et al. recommends determining optimum binder treatment with regard to type of recycled materials, especially when there is a lack of local moisture- and performance-related material properties. The environmental impacts to soils or groundwater need to be evaluated when RAP is stockpiled or used as an unbound granular material. Although RAP is usually free from damaging chemical compounds, RAP obtained from milling pavements subjected to deicing salts may contain some hazardous chemicals. Cosentino et al. (2003) analyzed RAP for heavy metal (silver, cadmium, chromium, lead, and selenium) concentrations and concluded that the levels were well below the limits speci- fied by the EPA. Cosentino et al. reported that the strength- deformation characteristics of field sites constructed using RAP improved over the 8-week study period, as reflected from field CBR, FWD, Clegg impact hammer, and soil stiff- ness gauge test results. Based on the results, the authors also recommended permitting the use of RAP as a subbase below rigid pavements. Snyder and Bruinsma (1996) conducted an extensive lit- erature review evaluating the effects of RCA usage in UAB layers on pavement drainage. Later, using thermodynamic techniques, Bruinsma et al. (1997) observed that the Port- landite [Ca(OH)2] present in RCA can be dissolved in water and subsequently lead to the precipitation of calcite (CaCO3) upon coming in contact with atmospheric carbon dioxide (CO2). Such precipitation often can lead to a reduction in the permittivity of pavement subdrainage systems. They con- curred with Muethel (1989) and Tamirisa (1993) in stating that the calcite precipitation can be controlled by reducing the amount of Portlandite [Ca(OH)2] readily available for dissolution, which may be accomplished through reduction in the amount of concrete cement fines. According to Stroup-Gardiner and Wattenberg-Komas (2013), crushing concrete reveals previously unexposed surfaces that contain some calcium hydroxide and partially unreacted cement grain and that react with air to form cal- cium carbonate precipitate. High levels of sodium chloride have been found in RCA produced from pavements subjected to deicing salts over years of service and may cause corro- values for RAP accompanied by significantly high permanent deformation accumulations. • The expansive properties of RAP materials contain- ing expansive components such as steel slag are best carefully evaluated before their application in UAB/subbase layers. • Recycled crushed concrete often can be adequately used in UAB/subbase layers. • Care is to be taken while blending two different recy- cled aggregate types to ensure that the resulting blend possesses adequate physical, chemical, and mechanical properties.

33 sion concerns if used in new PCC with steel. The alkalinity decreased rapidly when diluted with low pH water and expo- sure of the dissolved calcium hydroxide with CO2. The runoff could also be highly alkaline because of leaching of calcium hydroxide from freshly crushed concrete. Precipitate could clog drain pipes and filter fabrics, but washing the crushed concrete helped minimize some of these problems. AASHTO specification M 319-02, Reclaimed Concrete Aggregate for Unbound Soil-Aggregate Base Course, clearly identifies the high likelihood of increased pH values for water percolating through UAB layers constructed using RCA. The AASHTO specification further recommends setting appro- priate limits on the proximity of such layers to ground water and surface waters. Moreover, such layers should not be used in the vicinity of metal culverts susceptible to corrosion under such high-alkaline environments. Finally, precipita- tion of soluble minerals from the water percolating through base course layers containing RCA may lead to the clogging of drainage layers or other pavement drainage features. Thus, it is important to closely monitor and regulate the use of RCA in close proximity to such components. In some instances, recycled aggregates may not satisfy agency QC requirements for use as unbound aggregate pave- ment layers. In such cases, slight adjustments may be made to the QC specifications accompanied by design modifica- tions to allow the use of such “marginal” recycled materials. However, in cases in which the recycled material properties deviate significantly from agency specifications, the use of those materials in unbound aggregate pavement layers is best prohibited. The survey of state and Canadian provin- cial transportation agencies indicated the presence of envi- ronmental concerns in several agencies regarding the use of RAP and RCA in UAB and subbase layers. One respondent (Indiana DOT) mentioned an experience with loss of vegeta- tion caused by leaching from an unbound aggregate layer constructed using RCA. Indiana DOT has since prohibited the use of recycled materials in UAB and subbase layers, particularly for pavements with underdrain systems, because precipitation of leachates potentially may lead to clogging of the underdrain system. A recent report by FHWA summarized the experience of several state transportation agencies concerning the use of RCA in transportation applications. Based on the experi- ence of MnDOT, RCA could be used up to 100% as a filter/ separation layer under a permeable aggregate base drainage layer in accordance with the applicable drainage specifica- tions. In the presence of drainage layers and/or perforated drainage pipes, a blend of RCA with new aggregate could be used as subgrade when at least 95% of the RCA was retained on the 4.75-mm sieve. Alkaline effluent from RCA layer was not a significant issue when RCA was kept a sufficient dis- tance from the drainage outlets. A blend of open-graded RCA with new aggregate could be used for improved stability and density (FHWA 2004). Recycling of Unbound aggregate Material from Existing Pavements The survey of state and Canadian provincial transportation agencies collected information on agency policies regarding recycling of unbound aggregate materials from base and sub- base layers of existing pavements. Twenty-four of 46 (52.2%) respondents indicated that recycling of unbound aggregate materials from existing pavement base and subbase layers was a common practice in their respective states. Moreover, seven responded that such recycling was done occasionally in their states. Twenty-one of 46 respondents indicated that the use of recycled aggregates from existing base and subbase courses was incorporated into their state specifications, whereas 22 states did not allow the inclusion of such materials into specifications. Only two states allow contractors to use locally available “marginal” or “out of specification” aggregates for UAB and subbase applications. Six states allow the use of such “marginal” materials occasionally because of economic issues. Most agencies have a stricter material quality require- ment for UAB layers than for subbase layers. Therefore, these agencies sometimes allow the use of marginal aggregates in subbase layers while prohibiting their use in base layers. Some states also indicated that marginal materials occasionally were blended with virgin aggregates to lower the cost associated with material procurement and transportation. Commonly Used Recycled Materials in Unbound aggregate Base and Subbase Layers The survey of state and Canadian provincial transportation agencies indicated that RCA and RAP are the two most com- monly used recycled materials in UAB and subbase layers. Some agencies also reported the use of less commonly avail- able materials, such as air-cooled blast furnace slag (three agencies), glass cullets (seven agencies), aggregates blended with oil field waste (one agency), and so forth. Figure 22 shows the relative distribution of state transportation agen- cies using different recycled aggregate materials in UAB and subbase layer construction. Current State of the Practice Regarding Testing of Recycled Materials In the survey of state and Canadian provincial transportation agencies conducted under the scope of this synthesis study, information was gathered on the current state of practice regarding the testing of recycled materials for quality accep- tance. Agency responses indicated that sieve analysis, abra- sion tests such as Los Angeles abrasion or Micro-Deval, and percent deleterious materials are the most commonly used tests for evaluating recycled granular material quality (see Fig- ure 23). Four agencies also indicated the use of Atterberg limit tests on recycled granular materials. Note that soundness tests using sodium or magnesium sulfate may result in RCA being susceptible to sulfate attack, therefore resulting in high loss values. Thus, AASHTO specification M 319-02 recommends the use of soundness tests using sulfate solutions only when

34 local experience has found these methods to be satisfactory. In lieu of the sulfate soundness tests, agencies may opt to waive the soundness requirements or adopt one of the following alter- native test methods: • AASHTO T 103: Soundness of Aggregates by Freezing and Thawing; • New York State DOT Test Method NY 703-08: Resis- tance of Coarse Aggregate to Freezing and Thawing; or • Ontario Ministry of Transportation (MOT), Test Method LS-614: Freezing and Thawing of Coarse Aggregate. When asked about environmental concerns regarding the use of recycled granular materials in pavement unbound aggre- gate layers, 68% (17 of 25 respondents) indicated no such con- cerns. This indicates a potential gap in knowledge concerning phenomena such as leaching from RCA and the resulting con- 67% (31) 80% (37) 22% (10) 4% (2) 0 10 20 30 40 0% 20% 40% 60% 80% 100% Reclaimed Asphalt Pavement (RAP) Recycled Concrete Aggregates (RCA) Other (please indicate) None of the above Number of Responses Percentage of Respondents 46 survey respondents 32% (8) 48% (12) 88% (22) 44% (11) 12% (3) 0 5 10 15 20 25 0% 20% 40% 60% 80% 100% Na2SO4 / MgSO4 Soundness Test Los Angeles Abrasion and/or Micro Deval Test Sieve Analysis Percent Deleterious Materials Other (sand equivalent) Number of Responses Percentage of Respondents 25 survey respondents FIGURE 22 Use of different recycled aggregate materials by state transportation agencies in unbound aggregate pavement base and subbase applications (46 respondents). FIGURE 23 Different tests used by agencies for evaluating the material quality of recycled granular materials. Key Lessons • Test recycled materials for potential environmental impacts before use in UAB/subbase layers. tamination of groundwater. For the agencies that did report environmental concerns with recycled granular materials, leaching and resulting change in the pH level of groundwater were reported to be the primary concerns. Sixty-four percent of the responding agencies do not require any strength, defor- mation, or modulus characterization of recycled materials such as RCA and RAP before their use in unbound aggregate pavement layers. For the agencies that require such tests to be conducted on recycled materials, the quality requirements are the same as those for virgin aggregates.

35 SUMMaRY This chapter presents an overview of different types of aggre- gate materials available as natural resources in the United States. Geologic phenomena responsible for the formation of different rock types are discussed, and the distribution of different rock types in the conterminous United States is pre- sented. Important aggregate properties that affect the perfor- mance of UAB and subbase layers are discussed. Accordingly, different test procedures commonly used by transportation agencies to check the quality of aggregates before including them in UAB and subbase layers are listed. At this stage, the state of the practice in aggregate material selection and qual- ity check is discussed by presenting the information gathered through survey of state and Canadian provincial transporta- tion agencies. The concept of best value granular material utilization was introduced for pavement projects with the potential to save energy and material hauling costs. The sus- tainable production and use of aggregate and recycled granu- lar materials in base and subbase layers are discussed, as are different tests recommended by researchers to check the qual- ity of recycled aggregate materi als in base and subbase layer applications. REFERENCES Adu-Osei, A., D.N. Little, E. Tutumluer, and R.L. Lytton, “Effect of Compaction on Orthotropic Properties of Aggregates,” In Proceedings of the European Work- shop on Compaction of Soils and Granular Materials, ISSMGE, Paris, France, May 19, 2000. Allen, J., The Effect of Non-constant Lateral Pressures of the Resilient Response of Granular Materials, Ph.D. thesis, University of Illinois at Urbana–Champaign, Urbana, 1973. Allen, J.J. and M.R. Thompson, “Resilient Response of Granular Materials Subjected to Time Dependent Lateral Stresses,” Transportation Research Record 510, Trans- portation Research Board, National Research Council, Washington, D.C., 1974, pp. 1–13. American Concrete Pavement Association (ACPA), Recy- cled Concrete in Subbases: A Sustainable Choice, Technical Series, TS204.9P, ACPA, Washington, D.C., 2008 [Online]. Available: http://www.pavement.com/ Concrete_Pavement/Technical/Downloads/TSs.asp. Ashtiani, R.S. and D. Little, Acceptability Criteria for High Fines Content Aggregate Pavement Layers, Report No ICAR 401131, International Center for Aggregates Research (ICAR), College Station, Tex., 2007, 100 pp. Austroads, Development of Performance-based Specifica- tions for Unbound Granular Materials. Part A: Issues and Recommendations, AP-T29/03, Sydney, Australia, 2003. Barksdale, R.D., “Laboratory Evaluation of Rutting in Base Course Materials,” Proceedings of the 3rd International Conference on Structural Design of Asphalt Pavements, 1972, pp. 161–174. Barksdale, R.D., “Crushed Stone Base Performance,” Trans- portation Research Record 954, Transportation Research Board, National Research Council, Washington, D.C., 1984, pp. 78–87. Barksdale, R.D. and S.Y. Itani, “Influence of Aggregate Shape on Base Behavior,” Transportation Research Record 1227, Transportation Research Board, National Research Council, Washington, D.C., 1989, pp. 173–182. Barksdale, R.D., S.F. Brown, and F. Chan, NCHRP Report 315: Potential Benefits of Geosynthetics in Flexible Pave- ments, Transportation Research Board, National Research Council, Washington, D.C., 1989. Barksdale, R.D., The Aggregate Handbook, National Stone Association, Washington, D.C., 1991. Barksdale, R.D., C.O. Pollard, T. Siegel, and S. Moeller, Eval- uation of the Effect of Aggregate on Rutting and Fatigue of Asphalt, Technical Report FHWA-AG-92-8812, Geor- gia Department of Transportation, Atlanta, 1992. Bennert T., W.J. Papp, A. Maher, and N. Gucunski, “Utili- zation of Construction and Demolition Debris Under Traffic-Type Loading in Base and Subbase Applications,” Transportation Research Record: Journal of the Trans- portation Research Board, No. 1714, Transportation Research Board of the National Academies, Washington, D.C., 2000, pp. 33–39. Bilodeau, J.-P., G. Dore, and P. Pierre, “Erosion Susceptibil- ity of Granular Pavement Materials,” International Jour- nal of Pavement Engineering, Vol. 8, No. 1, Mar. 2007, pp. 55–66. Bilodeau, J.-P., G. Dore, and P. Pierre, “Gradation Influ- ence on Frost Susceptibility of Base Granular Materials,” International Journal of Pavement Engineering, Vol. 9, No. 6, Dec. 2008, pp. 397–411. Bilodeau, J.-P., G. Dore, and P. Pierre, “Pavement Base Unbound Granular Materials Gradation Optimization,” In Proceedings of the 8th International Conference on the Bearing Capacity of Roads, Railways and Airfields, Vol. 1, Taylor & Francis Group, London, U.K., 2009, pp. 145–154. Bolen, W.P., 2010 Minerals Yearbook: Sand and Gravel, Con- struction [advance release], U.S. Geological Survey, U.S. Department of the Interior, Washington, D.C., June 2012. Brown, S.F. and F.W.K. Chan, “Reduced Rutting in Unbound Granular Pavement Layers through Improved • Evaluate the use of recycled materials in pavements on a project basis, rather than by following generic guidelines. For example, the use of RCA in UAB/ subbase layers may or may not be allowed depend- ing on whether the pavement has an underdrain system or not. This is particularly important for RCA obtained from sources that have not been tested pre- viously for material quality. Some of the quality test- ing requirements may be relaxed for “properly tested and documented” RCA sources.

36 Grading Design,” Proceedings of the Institution of Civil Engineers, Transport, Vol. 117, No. 1, 1996, pp. 40–49. Bruinsma, J.E., K.R. Peterson, and M.B. Snyder, “Chemi- cal Approach to Formation of Calcite Precipitate from Recycled Concrete Aggregate Base Layers,” Transpor- tation Research Record: Journal of the Transportation Research Board, No. 1577, Transportation Research Board of the National Academies, Washington, D.C., 1997, pp. 10–17. Bullen F., “Design and Construction of Low-Cost, Low- Volume Roads in Australia,” Paper No. LVR8-1116, TRR 1819, School of Engineering, University of Tasmania, Hobart, Australia, 2003. Chini, A., S. Kuo, J. Armaghani, and J. Duxbury, “Test for Recycled Concrete Aggregate in Accelerated Test Track,” Journal of Transportation Engineering, ASCE, Vol. 127, No. 6, 2001, pp. 486–492. Clyne, T.R., E.N. Johnson, and B.J. Worel, Use of Taconite Aggregates in Pavement Applications, Final Report MN/ RC-2010-24, Minnesota Department of Transportation, St. Paul, June 2010. Collins, R.J. and S.K. Ciesielski, NCHRP Synthesis of High- way Practice 199: Recycling and Use of Waste Materials and By-Products in Highway Construction, Transporta- tion Research Board, National Research Council, Wash- ington, D.C., 1994. Cook, J.R. and C.S. Gourley, “A Framework for the Appropri- ate Use of Marginal Materials,” World Road Association (PIARC), Technical Committee C12 Seminar, Mongolia, 2002. Cosentino, P.J., et al., Developing Specifications for Using Recycled Asphalt Pavement as Base, Subbase or Gen- eral Fill Materials, Phase II, Report No. FL/DOT/RMC/ 06650-7754, Florida Department of Transportation, Tal- lahassee, 2003. Dawson, A.R., N.H. Thom, and J.L. Paute, “Mechanical Characteristics of Unbound Granular Materials as a Func- tion of Condition,” Flexible Pavements, Proceedings of European Symposium Euroflex 1993, A.G. Correia, ed., Balkema, Rotterdam, the Netherlands, 1996, pp. 35–44. Deniz, D., E. Tutumluer, and J.S. Popovics, “Evaluation of the Expansive Characteristics of Reclaimed Asphalt Pavement (RAP) and Virgin Aggregate Used As Base Materials,” Transportation Research Record: Journal of the Transportation Research Board, No. 2167, Transpor- tation Research Board of the National Academies, Wash- ington, D.C., 2010, pp. 10–17. Dempsey, B.J., “Laboratory and Field Studies of Channel- ing and Pumping,” Transportation Research Record 849, Transportation Research Board, National Research Coun- cil, Washington, D.C., 1982. Elliott, R.P. and S.I. Thornton, “Resilient Modulus and AASHTO Pavement Design,” Transportation Research Record 1196, Transportation Research Board, National Research Council, Washington, D.C., 1988, pp. 116–124. Federal Highway Administration (FHWA), Transportation Applications of Recycled Concrete Aggregate, FHWA State of the Practice National Review, FHWA, Washing- ton, D.C., Sep. 2004. Federal Highway Administration (FHWA), User Guidelines for By-product and Secondary Use Materials in Pave- ment Construction, FHWA, Washington, D.C., 2008 [Online]. Available: http://www.fhwa.dot.gov/pavement/ pub_details.cfm?id=384. Garg, N. and M.R. Thompson, “Lincoln Avenue Reclaimed Asphalt Pavement Base Project,” Transportation Research Record 1547, Transportation Research Board, National Research Council, Washington, D.C., 1996, pp. 89–95. Garg, N. and M.R. Thompson, “Triaxial Characterization of Minnesota Road Research Project Granular Materials,” Transportation Research Record 1577, Transportation Research Board, National Research Council, Washing- ton, D.C., 1997. Gates, L., E. Masad, R. Pyle, and D. Bushee, Aggregate Imaging System 2 (AIMS2): Final Report, Report FHWA- HIF-11-030, Highway for Life Program Office, Office of Infrastructure, Federal Highway Administration, U.S. Department of Transportation, Washington, D.C., 2011. Gray, J.E., (1962). “Characteristics of Graded Base Course Aggregates Determined by Triaxial Tests,” Engineering Research Bulletin No. 12, National Crushed Stone Asso- ciation, Alexandria, Va., 1962. Halmen, C.K. and J.T. Kevern, “Utilization of Quarry Fines for Sustainable Construction,” Stone, Sand & Gravel Review, July/Aug. 2010, pp. 26–27 [Online]. Available: http://www.nssga.org/ssgReview/. Hicks, R.G. and C.L. Monismith, “Factors Influencing the Resilient Properties of Granular Materials,” Transpor- tation Research Record 345, Transportation Research Board, National Research Council, Washington, D.C., 1971, pp. 15–31. Hill, A.R., A.R. Dawson, and M. Mundy, “Utilization of Aggregate Materials in Road Construction and Bulk Fill,” Journal of Resources, Conservation and Recycling, Vol. 32, No. 3-4, July 2001, pp. 305–320. Holtz, R.D., “Compaction Concepts,” Chapter 3, In Guide to Earthwork Compaction, State of the Art Report 8, Trans- portation Research Board, National Research Council, Washington, D.C., 1990. Holubec, I., Cyclic Creep of Granular Materials, Report No. RR147, Department of Highways, Ontario, Canada, 1969. Hudson, W.R., D. Little, A.M. Razmi, V. Anderson, and A. Weismann, An Investigation of the Status of By-Product Fines in the United States, Research Report 101-1, Inter- national Center for Aggregate Research, Austin, Tex., Nov. 1997. Illinois Department of Transportation (IDOT), “Subgrade Stability Manual,” Bureau of Bridges and Structures, IDOT, Springfield, May 1, 2005, 27 pp. [Online]. Avail-

37 able: http://www.dot.il.gov/bridges/pdf/subgrade stability manual.exe. Janoo, V.C., Quantification of Shape, Angularity, and Sur- face Texture of Base Course Materials, U.S. Army Corps of Engineers Special Report 98-1, Cold Regions Research & Engineering Laboratory, Hanover, N.H., 1998. Jorenby, B.N. and R.G. Hicks, “Base Course Contami- nation Limits,” Transportation Research Record 1095, Transportation Research Board, National Research Coun- cil, Washington, D.C., 1986, pp. 86–101. Kaya, Z., A. Cetin, B. Cetin, and A.H. Aydilek, “Effect of Compaction Method on Mechanical Behavior of Graded Aggregate Base Materials,” In Proceedings of the 2012 GeoCongress, Oakland, California, Mar. 25–29, 2012, pp. 1486–1494. Kamal, M.A., A.R. Dawson, O.T. Farouki, D.A.B. Hughes, and A.A. Sha’at, “Field and Laboratory Evaluation of the Mechanical Behavior of Unbound Granular Materials in Pavements,” Transportation Research Record 1095, Transportation Research Board, National Research Coun- cil, Washington, D.C., 1993, pp. 88–97. Khogali, W.E.I. and E.H.H. Mohamed, “Mechanistic Classi- fication of Unbound Materials,” Transportation Research Record: Journal of the Transportation Research Board, 2016, Transportation Research Board of the National Academies, Washington, D.C., 2007, pp. 39–48. Knutson, R.M. and M.R. Thompson, “Resilient Response of Railway Ballast,” Transportation Research Record 651, Transportation Research Board, National Research Coun- cil, Washington, D.C., 1977, pp. 31–39. Kolisoja, P., T. Saarenketo, H. Peltoniemi, and N. Vuorimies, “Laboratory Testing of Suction and Deformation Proper- ties of Base Course Aggregates,” Transportation Research Record: Journal of the Transportation Research Board, 1787, Transportation Research Board of the National Academies, Washington, D.C., 2002, pp. 83–89. Kumar, D.S. and W.R. Hudson, Use of Quarry Fines for Engineering and Environmental Applications, Special Research Report for the National Stone Association, Center for Transportation Research, The University of Texas-Austin, Oct. 1992, 130 pp. Kuo, S.S., H.S. Mahgoub, and J.E. Ortega, Use of Recycled Concrete Made with Florida Limestone Aggregate for a Base Course in Flexible Pavement, Florida Department of Transportation Final Report for Contract BC 409, State Materials Office, Gainesville, 2001, 212 pp. Langer, W.H., Aggregate Resource Availability in the Conterminous United States, Including Suggestions for Addressing Shortages, Quality, and Environmental Con- cerns, Open File Report 2011-1119, U.S. Geological Survey, U.S. Department of the Interior, Washington, D.C., 2011. Lekarp, F., U. Isacsson, and A. Dawson, “State of the Art. I: Resilient Response of Unbound Aggregates,” Journal of Transportation Engineering, Vol. 126, No. 1, 2000a, pp. 66–75. Lekarp, F., U. Isacsson, and A. Dawson, “State of the Art. II: Permanent Strain Response of Unbound Aggregates,” Journal of Transportation Engineering, Vol. 126, No. 1, 2000b, pp. 76–83.1. Lukanen, E.O., Application of AASHO Road Test Results to Design of Flexible Pavements in Minnesota, Final Report FHWA/MN-80/09, Office of Materials and Research, Min- nesota Department of Transportation, St. Paul, 1980, 83 pp. Maher, M.H. and W. Popp, Recycled Asphalt Pavement as a Base and Subbase Material, ASTM STP 1275, Ameri- can Society of Testing and Materials, New Orleans, 1997, pp. 42–53. Manning, D. and J. Vetterlein, Exploitation and Use of Quarry Fines, Mineral Solutions, Ltd., Manchester, U.K., 2004, 55 p. Maree, J.H., C.R. Freeme, N.J.W. van Zyl, and P.F. Savage, “The Permanent Deformation of Pavements with Untreated Crushed-Stone Bases as Measured in Heavy Vehicle Simulator Tests,” Proceedings of the 11th Austra- lian Road Research Board Conference, Melbourne, 1982. Masad, E., T. Al-Rousan, J. Button, D.N. Little, and E. Tutumluer, NCHRP Report 555: Test Methods for Char- acterizing Aggregate Shape, Texture, and Angularity, Proj- ect 4-30A, Transportation Research Board of the National Academies, Washington, D.C., 2007, 93 pp. Meininger, R.C. and S.J. Stokowski, “Wherefore Art Thou Aggregate Resources for Highways,” FHWA-HRT-11-006, Public Roads, Vol. 75, No. 2, Sep./Oct. 2011 [Online]. Available: http://www.fhwa.dot.gov/publications/public roads/11septoct/06.cfm. Mishra, D., E. Tutumluer, and A.A. Butt, “Quantifying Effects of Particle Shape and Type and Amount of Fines on Unbound Aggregate Performance Through Controlled Gradation,” Transportation Research Record: Journal of the Transportation Research Board, No. 2167, Transpor- tation Research Board of the National Academies, Wash- ington, D.C., 2010a, pp. 61–71. Mishra, D., E. Tutumluer, and Y. xiao, “Particle Shape, Type and Amount of Fines, and Moisture Affecting Resilient Modulus Behavior of Unbound Aggregates,” In ASCE Geotechnical Special Publication No. 203, Pav- ing Materials and Pavement Analysis, Proceedings Book of the ASCE GeoShanghai 2010 International Confer- ence, B. Huang, E. Tutumluer, I.L. Al-Qadi, J. Prozzi, and x. Shu, Eds., Shanghai, China, June 35, 2010, 2010b, pp. 279–287. Mishra, D. and E. Tutumluer, “Field Performance Evalua- tions of Illinois Aggregates for Subgrade Replacement & Subbase—Phase II,” Draft Final Report, Illinois Center for Transportation (ICT) R27-81 Project, University of Illinois Urbana–Champaign, Nov. 2011, 62 pp. Mishra, D., Aggregate Characteristics Affecting Response and Performance of Unsurfaced Pavements on Weak Sub- grades, Ph.D. Thesis, University of Illinois at Urbana– Champaign, 2012.

38 Muethel, R.W., Calcium Carbonate Precipitate from Crushed Concrete, Michigan Department of Transportation, Lan- sing, 1989. Ooi, P.S.K., F. Rajabipour, A. Shafaatian, and S. Joo, “Foren- sic Investigation of Distressed Pavement Supported on a Base Course Containing Recycled Concrete Aggregate,” Transportation Research Record: Journal of the Trans- portation Research Board, No. 2253, Transportation Research Board of the National Academies, Washington, D.C., 2011, pp. 22–31. Ovik, J.M., B. Birgisson, and D.E. Newcomb, Characteriz- ing Seasonal Variations in Pavement Material Properties for Use in a Mechanistic-Empirical Design Procedure, Final Report MN-RC-2000-35, Minnesota Department of Transportation, St. Paul, Dec. 2000, 191 pp. Pan, T., E. Tutumluer, and J. Anochie-Boateng, “Aggregate Morphology Affecting Resilient Behavior of Unbound Granular Materials,” Transportation Research Record: Journal of the Transportation Research Board, 1952, Transportation Research Board of the National Acad- emies, Washington, D.C., 2006, pp. 12–20. Paute, J.L., P. Hornych, and J.P. Benaben, “Comportement mécanique des graves non traitées,” Bulletin de Liai- son des Laboratoires des Ponts et Chaussées 190, 1994, pp. 27–38. Puppala, A.J., S. Saride, S.K. Sirigiripet, and R. Williammee, “Evaluation of Cemented Quarry Fines as a Pavement Base Material,” ASCE Geotechnical Special Publica- tion 177, Geo-Congress, New Orleans, La., Feb. 2008, pp. 312–319. Rao, C., E. Tutumluer, and I.T. Kim, “Quantification of Coarse Aggregate Angularity Based on Image Analy- sis,” Transportation Research Record: Journal of the Transportation Research Board, 1787, Transportation Research Board of the National Academies, Washington, D.C., 2002, pp. 17–24. Rathje E.M., et al., Recycled Asphalt Pavement and Crushed Concrete Backfill: Results from Initial Durability and Geotechnical Tests, Center for Transportation Research Report 4177-2, The University of Texas at Austin, 2002, 70 pp. Roberts, F.L., P.S. Kandhal, E.R. Brown, D.-Y. Lee, and T.W. Kennedy, Hot Mix Asphalt Materials, Mixture Design, and Construction, 2nd ed., NAPA Education Foundation, Maryland, 1996, pp. 102–114. Rohde, L., W.P. Nunez, and J.A.P. Ceratti, “Electric Arc Fur- nace Steel Slag Base Material for Low-Volume Roads,” Transportation Research Record: Journal of the Trans- portation Research Board, 1819, Transportation Research Board of the National Academies, Washington, D.C., 2003, pp. 201–207. Rowshanzamir, M.A., Resilient Cross-Anisotropic Behavior of Granular Base Materials Under Repetitive Loading, Ph.D. Thesis, School of Civil Engineering, University of New South Wales, Australia, 1995. Rupnow, T.D., V.R. Schaefer, and D.J. White, “An Investi- gation of the Use of Limestone Screenings in Roadway Construction,” CD-ROM, Proceedings of 89th Annual Meeting of the Transportation Research Board, Jan. 10–14, 2010, Washington, D.C. Sadecki, R.W., G.P. Busacker, K.L. Moxness, K.C. Faruq, and L.G. Allen, An Investigation of Water Quality in Run- off from Stockpiles of Salvaged Concrete and Bituminous Paving, Report No. MN/PR-96/31, Minnesota Depart- ment of Transportation, St. Paul, 1996. Saeed, A., J.W. Hall, Jr., and W. Barker, NCHRP Report 453: Performance-Related Tests of Aggregates for Use in Unbound Pavement Layers, Transportation Research Board, National Research Council, Washington, D.C., 2001, 55 pp. Saeed, A., NCHRP Report 598: Performance-Related Tests of Recycled Aggregates for Use in Unbound Pavement Layers, Transportation Research Board of the National Academies, Washington, D.C., 2008, 53 pp. Seyhan, U. and E. Tutumluer, “Anisotropic Modular Ratios as Unbound Aggregate Performance Indicators,” Journal of Materials in Civil Engineering, Vol. 14, No. 5, 2002, pp. 409–416. Snyder, M.B. and J.E. Bruinsma, “Review of Studies Concern- ing Effects of Unbound Crushed Concrete Bases on PCC Pavement Drainage,” Transportation Research Record 1519, Transportation Research Board, National Research Council, Washington, D.C., 1996, pp. 51–58. Stroup-Gardiner, M. and T. Wattenberg-Komas, NCHRP Synthesis 435: Recycled Materials and Byproducts in Highway Applications, Volume 6: Reclaimed Asphalt Pavement, Recycled Concrete Aggregate, and Construc- tion Demolition Waste, Transportation Research Board of the National Academies, Washington, D.C., 2013. Taha, R., G. Ali, A. Basma, and O. Al-Turk, “Evaluation of Reclaimed Asphalt Pavement Aggregate in Road Bases and Subbases,” Transportation Research Record 1652, Transportation Research Board, National Research Coun- cil, Washington, D.C., 1999, pp. 264–269. Tamirisa, R., Study of Highway Base/Subbase Aggregates That Cause Depositions of Calcareous “Tufa” in Drains, Master’s thesis, Department of Civil Engineering, Uni- versity of Toledo, Ohio, 1993. Tao, M., M.N. Louay, N.D. Munir, Z. Zhang, and Z. Wu, “Application of Shakedown Theory in Characterizing Tra- ditional and Recycled Pavement Base Materials,” Jour- nal of Transportation Engineering, Mar. 2010, Vol. 136, No. 3, pp. 214–222. Thom, N.H., Design of Road Foundations, Ph.D. Thesis, University of Nottingham, Department of Civil Engineer- ing, May 1988, 307 pp. Thom, N.H. and S.F. Brown, “Effect of Moisture on the Structural Performance of a Crushed-Limestone Road Base,” Transportation Research Record 1121, Trans- portation Research Board, National Research Council, Washington, D.C., 1987, pp. 50–56. Thom, N.H. and S.F. Brown, “The Effect of Grading and Density on the Mechanical Properties of a Crushed Dolo- mitic Limestone,” Proceedings of the 14th ARRB Confer- ence, Part 7, 1988, pp. 94–100.

39 Thompson, M.R. and Q.L. Robnett, “Resilient Properties of Subgrade Soils,” Transportation Engineering Journal, ASCE, Vol. 105, No. TE1, 1979. Thompson, M.R., and T. LaGrow, “A Proposed Conventional Flexible Pavement Thickness Design Procedure,” Trans- portation Engineering Series No. 55, Civil Engineering Studies, University of Illinois at Urbana–Champaign, 1988. Tutumluer, E. and U. Seyhan, “Neural Network Modeling of Anisotropic Aggregate Behavior from Repeated Load Triaxial Tests,” Transportation Research Record 1615, Transportation Research Board, National Research Coun- cil, Washington, D.C., 1998, pp. 86–93. Tutumluer, E. and U. Seyhan, “Effects of Fines Content on the Anisotropic Response and Characterization of Unbound Aggregate Bases,” Proceedings of the 5th International on Unbound Aggregates in Roads (UNBAR5), Unbound Aggregates in Road Construction, A.R. Dawson, Ed., A.A. Balkema, University of Nottingham, U.K., 2000, pp. 153–161. Tutumluer, E. and T. Pan, “Aggregate Morphology Affect- ing Strength and Permanent Deformation Behavior of Unbound Aggregate Materials,” Journal of Materials in Civil Engineering, ASCE, Vol. 20, No. 9, Sep. 2008, pp. 1–11. Tutumluer, E., D. Mishra, and A. Butt, Characterization of Illinois Aggregates for Subgrade Replacement and Sub- base, Final Report, Illinois Center for Transportation (ICT) R27-1 Project, University of Illinois, Urbana–Champaign, Sep. 2009, 90 pp. Uthus, L., E. Tutumluer, I. Horvli, and I. Hoff, “Influence of Grain Shape and Surface Texture on the Deformation Properties of Unbound Aggregates in Pavements,” Inter- national Journal of Pavements, Vol. 6, No. 1–3, Jan.–May– Sep. 2007, pp. 75–88. Van Niekerk, A.A., Mechanical Behavior and Performance of Granular Bases and Sub-Bases in Pavements, Ph.D. Thesis, Delft University of Technology, Delft, the Nether- lands, 2002. Willett, J.C., 2008 Stone (Crushed). U.S. Geological Survey Mineral Commodity Summaries, pp. 158–159 [Online] Available: http://minerals.usgs.gov/minerals/pubs/ commodity/stone_crushed/mcs-2008-stonc.pdf [accessed Jan. 8, 2013]. Willett, J.C., 2010 Minerals Yearbook: Stone, Crushed [advance release], U.S. Geological Survey, U.S. Depart- ment of the Interior, Washington, D.C., Nov. 2011. Wilburn, D.R. and T.G. Goonan, “Aggregates from Natural and Recycled Sources: Economic Assessments for Con- struction Applications-A Materials Flow Analysis,” U.S. Geological Survey Circular 1176, U.S. Department of the Interior, Washington, D.C., 1998. Woodbridge, M.E., “Use of Soft Limestone for Road-Base Construction in Belize,” Seventh International Confer- ence on Low-Volume Roads, Baton Rouge, La., 1999. Woolf, D.O., “Results of Physical Tests of Road Build- ing Aggregates: To January 1951,” Physical Research Branch, Bureau of Public Roads, 1952. Wu, Y., F. Parker, Jr., and K. Kandhal, Aggregate Tough- ness/Abrasion Resistance and Durability/Soundness Tests Related to Asphalt Concrete Performance in Pavements, NCAT Report No. 98-4, National Center for Asphalt Tech- nology, Auburn, Ala., 1998. xiao, Y., E. Tutumluer, and J. Siekmeier, “Mechanistic- Empirical Evaluation of Aggregate Base and Granular Subbase Quality Affecting Flexible Pavement Perfor- mance in Minnesota,” Transportation Research Record: Journal of the Transportation Research Board, No. 2227, Transportation Research Board of the National Acad- emies, Washington, D.C., 2011, pp. 97–106. xiao, Y. and E. Tutumluer, Best Value Granular Material for Road Foundations, Final Report, Submitted to Minnesota Department of Transportation, St. Paul, Dec. 2011, 179 pp.