National Academies Press: OpenBook

Aggregate Quality Requirements for Pavements (2018)

Chapter: Chapter 5 - Aggregate Quality Affecting Pavement Performance

« Previous: Chapter 4 - Procedures for Approving Aggregate Sources
Page 59
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 59
Page 60
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 60
Page 61
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 61
Page 62
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 62
Page 63
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 63
Page 64
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 64
Page 65
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 65
Page 66
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 66
Page 67
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 67
Page 68
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 68
Page 69
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 69
Page 70
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 70
Page 71
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 71
Page 72
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 72
Page 73
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 73
Page 74
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 74
Page 75
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 75
Page 76
Suggested Citation:"Chapter 5 - Aggregate Quality Affecting Pavement Performance." National Academies of Sciences, Engineering, and Medicine. 2018. Aggregate Quality Requirements for Pavements. Washington, DC: The National Academies Press. doi: 10.17226/25205.
×
Page 76

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.

59 5.1 Introduction This chapter focuses on the aggregate-related pavement performance records and how qual- ity may be linked to the behavior of individual pavement layers under traffic loading and envi- ronmental conditions. This includes effects of aggregate quality and source issues on strength, modulus, permanent deformation, and durability characteristics of bound and unbound layers in constructed pavement structures. The common aggregate quality-related pavement distresses reported by transportation agencies will also be presented herein. Finally, environmental and performance-related concerns experienced by transportation agencies regarding the use of recycled (e.g., RAP and RCA) and artificial/byproduct aggregates (e.g., SFS, BFS, and quarry byproducts) will be discussed. Aggregate quality checks are needed to ensure good performance of aggregates in various pave- ment layer applications. Source properties affecting the quality of individual aggregate particles and the ultimate performance of the pavement structure can be related to aggregate’s physical, chemical, and mechanical properties. In NCHRP Report 598, Saeed (2008) summarized the main source properties in recycled aggregates that can influence the performance of different layers of pavement. These properties also apply to virgin aggregate materials used in pavement applications and are summarized in Table 5-1 (after Saeed 2008). Shear strength of aggregates (mainly recycled aggregates) was identified as the main quality performance measure when used in structural layers of flexible and rigid pavements (Saeed 2008). A list of aggregate properties and their influences on the different pavement layer applications is presented in Table 5-2 (after Saeed 2008). 5.2 Pavement Distresses and Aggregate Quality 5.2.1 Flexible Pavements Because aggregate materials constitute more than 90% of flexible pavements, they can highly influence the overall performance, load carrying capacity, and type and rate of distresses in asphalt concrete layers. The two common distresses for asphalt pavements are rutting and fatigue cracking. Several aggregate properties can influence the type and severity of distresses that develop in flexible pavements. NCHRP Report 598 studied some of the main distresses that appear in flexible pavements, the recycled aggregate source properties that influence these distresses, and the related quality measures to check these source properties (Saeed 2008). A summary of the findings is presented in Table 5-3 (after Saeed 2008). NCHRP Report 539 evaluated several aggregate properties that can influence the perfor- mance of HMA mixtures designed by the Superpave method. The aggregate properties that influence performance are (1) coarse aggregate angularity, (2) flat and elongated particles, C H A P T E R 5 Aggregate Quality Affecting Pavement Performance

60 Aggregate Quality Requirements for Pavements Chemical propertiesPhysical properties Mechanical properties Table 5-1. Recycled aggregates source properties that can influence pavement performance. Mass Property of Material Structural layer Construction platform Drainage layer Frost blanket Control pumping Select fill Relevance of Mass Property to the Use of Recycled Material as Table 5-2. Relevance of recycled aggregates source properties to different pavement layer applications. Performance parameter Related aggregate property Test measures Table 5-3. Flexible pavement performance measures and the link to recycled aggregate source properties.

Aggregate Quality Affecting Pavement Performance 61 (3) fine aggregate angularity, and (4) sand equivalent (Prowell et al. 2005). Other source properties tested include abrasion loss by LAA or micro-deval, sulfate soundness, and the quantities of deleterious materials. The aggregate consensus properties are also determined for the aggregate blends used in HMA mixes. NCHRP Report 539 states that coarse aggregate angularity influences the rutting performance of HMA mixes, but it is uncertain if fine aggregate angularity relates to rutting performance of HMA mixes. The sand equivalent test relates to the possibility of moisture damage of asphalt mixtures due to poor adhesion of the binder to the aggregates coated by clay-size particles. The correlation between flat and elongated ratio and fatigue resistance of asphalt mixtures is controversial as mentioned in different studies, while the properties of the fine aggregates in HMA can predominantly control fatigue cracking resistance (Prowell et al. 2005). Aggregate particle angularity or number/percentage of fractured surfaces has been reported as one of the most significant factors affecting rutting performance of asphalt pavements (Parker and Brown 1998). They concluded that mixes with crushed aggregates and highly angular natu- ral sand showed better rutting performance. Similarly, it was identified through field evaluation that using rounded coarse aggregates and uncrushed natural aggregates in asphalt mixtures is one of the main factors contributing to rutting (Button et al. 1990). A state-of-the-art study by Zaumanis and Mallick (2015) for using high RAP contents (> 40%) in asphalt plant mixes proposed putting extra efforts into the design and evaluation of RAP, including proper mixing with the virgin aggregates, the evaluation of the RAP aggregate prop- erties after binder extraction, and checking the dust content of the RAP, to ensure good field performance for rutting, fatigue, and longevity (Zaumanis and Mallick 2015). Abdelrahman et al. (2010) reported a strong relation between the quality of coarse aggregates used in HMA and performance: aggregate chemical composition can affect moisture susceptibility and bonding to the asphalt binder; weak bonding between the aggregates and binder in the presence of excessive deleterious materials such as clay lumps can result in spalling, raveling, and stripping; pavement aggregate abrasion can be highly correlated to magnesium sulfate soundness and micro-deval abrasion (Abdelrahman et al. 2010). The quality for aggregates used in HMA was reported to be influenced by the practices and techniques adopted by the quarries, aggregate plants, asphalt mixing plant operations, and prac- tices during construction, which ultimately influence the performance (Vazquez et al. 2010). MS-16: Asphalt in Pavement Preservation and Maintenance related several asphalt pavement dis- tresses to the quality of aggregates used in HMA mixes: block cracking can result from using absorptive fine aggregates with low asphalt penetration, corrugations or shoving can result from using rounded/smooth aggregates, polished aggregates at the surface can result from using soft aggregates, and aggregate loss from surface treatments can result from using dusty aggregates (Asphalt Institute 2009). Similarly, Adlinge and Gupta (2013) reported that corrugation, raveling, and polishing can result from aggregate quality issues in HMA, including the use of smooth- textured coarse aggregates, loss of adhesion between binder and aggregates, aggregate breakage, and aggregate wear (Adlinge and Gupta 2013). Pan (2006) investigated the effect of morphological shape properties (i.e., angularity, flat and elongated ratio, and surface texture) on the rutting performance of 46 flexible pavement test sections constructed at the National Center for Asphalt Technology test track. Each of the three shape properties was individually correlated to the performance of the different test sections. The University of Illinois Aggregate Image Analyzer (UIAIA) was used to acquire and process images of aggregate particles to identify imaging-based shape properties. All three morpho- logical shape properties individually showed good correlation with the rutting performance of asphalt mixes. Surface texture showed the highest correlation, with coefficient of determination equal to 0.94 and the lowest p-value for a statistical t-test (see Figure 5-1).

62 Aggregate Quality Requirements for Pavements Moaveni et al. (2014) combined the micro-deval weight loss values with the imaging-based shape property results of two different imaging systems: Enhanced University of Illinois Aggre- gate Image Analyzer and a second-generation Aggregate Imaging System (AIMS-II) for the evaluation of the tendency of aggregates to polishing, breakage, and abrasion. The two systems similarly classified 11 types of aggregates into four categories according to resistance to breakage and abrasion (Moaveni et al. 2014). This tendency to polishing correlates with the frictional properties of the asphalt surface layers for aggregates used in HMA mixes, as reported by Mahmoud and Perales (2015). Mahmoud and Ortiz (2014) developed aggregate polishing curves at different time intervals with micro-deval, which proved that retention characteristics and changes in surface texture characteristics were dependent on aggregates mineralogy. In cold regions, if soft aggregate sources are used in the construction of flexible pavements, depressions would be developed in the wheel paths (wear-induced rutting) due to excessive sur- face abrasion caused by studded tires. The effect of using hard aggregates to reduce the flexible pavement wear caused by studded tires has been studied in Alaska, where high-quality aggregates (a) (b) (c) Figure 5-1. Aggregate shape properties and rutting performance of National Center for Asphalt Technology asphalt mixes showing (a) correlation with surface texture index, (b) correlation with flat and elongated ratio, and (c) correlation with angularity index (Pan 2006).

Aggregate Quality Affecting Pavement Performance 63 are not readily available. According to the cost-effectiveness and performance records collected, pavement performance could be increased by 1.4 to 1.9 times by using harder aggregates con- forming to the Nordic abrasion specification (Frith et al. 2004). Participating transportation agencies listed the aggregate quality-related flexible pavement distresses they have experienced. In total, 29 U.S. DOTs and eight Canadian provinces indicated at least one aggregate quality-related flexible pavement distress. The most commonly reported distresses are listed as follows. Refer to Appendix C for a detailed compilation of the individual distresses reported by each agency. • Stripping, debonding, raveling, popouts/pickouts or breakdown of coarse aggregate • Aggregate polishing • Longitudinal cracking, block cracking, thermal cracking, joint cracking, and fatigue cracking • Joint problems • Rutting and shoving • Aggregate moisture or freeze-thaw damage • Rapid oxidation caused by using high amount of RAP aggregate Stripping and polishing in flexible pavements were found to be the most common aggregate- related distresses reported by agencies. 5.2.2 Rigid Pavements Rigid pavements are very durable if adequately designed and high-quality materials and good construction practices are used. The effect of aggregate on PCC pavement performance has been well documented by many researchers (Delatte 2014). Moreover, service conditions such as temperature, humidity, and induced loads are among the factors that add complexity to design and construction of concrete pavements. PCC pavement performance can be assessed through consideration of each aggregate-related distress type observed in predominant PCC pavement types, including jointed plain concrete, jointed reinforced concrete, and continuously reinforced concrete pavements. Poor performance of PCC pavements can often be attributed to different qualities and properties of the aggregates used to construct the PCC. For example, NCHRP Report 598 listed cracking, pumping, faulting, and roughness as the most common distresses of rigid pavements constructed with recycled aggregates, along with aggregate source properties that influence these distresses as well as the proposed quality measures to check them (Saeed 2008). A summary of the findings is presented in Table 5-4 (after Saeed 2008). Darter et al. (1979) identified low aggregate quality as one of the major causes of distresses in CRCP. Similarly, Hanna (2003) summarized aggregate quality aspects closely linked to predominant Performance parameter Related aggregate property Test measures Table 5-4. Rigid pavement performance measures and the link to recycled aggregate source properties.

64 Aggregate Quality Requirements for Pavements distresses in PCC pavements. According to Hanna (2003), the quality indicators for aggregates that can significantly affect the performance of concrete pavements were mainly divided into five categories: (1) physical properties, (2) mechanical properties, (3) chemical and petrographic properties, (4) durability properties, and (5) others. The report summarizes common distress types and the aggregate properties related to each distress type for all concrete pavements, specif- ically for jointed plain concrete pavements (corner breaks and faulting) and CRCP (punchouts). The summary of findings is presented in Table 5-5, which lists the aggregate properties that con- tribute to common distresses or performance parameters and the mechanisms leading to those distresses or parameters. A summary of the performance parameters and the corresponding aggregate properties are also presented in Table 5-6. Mineralogy, morphological shape properties, abrasion resistance, coefficient of thermal expansion, and texture are common aggregate properties related to the different performance parameters in concrete pavements. In conclusion, as indicated in Tables 5-5 and 5-6 (both after Hanna 2003), aggregate properties and the quality significantly influence rigid pavement performance. As part of the survey questionnaire, information was collected from participating trans- portation agencies regarding rigid pavement distresses attributed to aggregate quality aspects. Twenty-seven U.S. DOTs and four Canadian provinces indicated that at least one aggregate quality aspect contributed to a rigid pavement distress. The commonly reported distresses are listed as follows. Refer to Appendix C for a detailed compilation of the individual distresses reported by each agency. • ASR and resulting cracking issues • Cracking associated with coefficient of thermal expansion of coarse aggregate • Sulfate attack • Carbonate reactivity of both coarse and fine aggregates • Polishing of fine aggregate • Durability cracking, freeze-thaw expansion, and staining • D-cracking and map cracking • Clay balls/lightweight pieces may affect concrete surface, strength, and durability • Popouts and corner breaks • Joint spalling, joint faulting, and scaling ASR and resulting cracking issues were found to be the common rigid pavement distress linked to aggregate quality reported by agencies. ASR usually results in random cracking at the surface of the concrete and is caused by excessive internal expansion. The most reactive aggre- gate contains opal, chalcedony, and certain forms of chert (opaline and chalcedonic cherts), volcanic glass, or siliceous impurities. Many igneous rock types that are acidic are potentially reactive (Thomas et al. 2013). PCC is unique in the significant degree to which it can be adversely affected by the destructive forces caused by cycles of freezing and thawing. With respect to the important effects of repeated freezing and thawing, the colder winter months of the northern states are not necessarily more unfavorable than the warmer southern winters of the southern regions. While freeze-thaw related damage may be caused by as few as 2 or 3 cycles of freezing per year (Janssen et al. 1986), greater numbers of freeze-thaw cycles generally lead to greater damage. Thus, a milder climate with frequent thaws during the day can be more severe than a colder climate that stays frozen. This can be especially true at higher elevations that thaw daily but re-freeze at night (Barksdale 1991, Vanderhorst and Janssen 1990). Aggregate-related freeze-thaw damage to concrete pave- ments is often manifested by D-cracking that occurs at pavement joints and cracks, which results in the eventual disintegration of the concrete near the joint. Aggregates with D-cracking potential are often identified by testing with AASHTO T 161 (ASTM C666), though this procedure can take 8 weeks or more to complete. A rapid test

Aggregate Quality Affecting Pavement Performance 65 Pavement type Performance parameter Manifestation Aggregate propertiesPCC propertiesMechanisms Table 5-5. Concrete pavement performance parameters and the corresponding aggregate properties.

66 Aggregate Quality Requirements for Pavements procedure called the hydraulic fracture test was developed as part of the Strategic Highway Research Program. This test procedure requires eight working days rather than eight weeks to complete. Though not considered precise enough to use as an aggregate rejection test, the procedure does have merit as a screening test (Janssen and Snyder 1994, Snyder et al. 1996). The identification of some D-cracking aggregates has proved difficult because the aggregates only appear to produce D-cracking after exposure to deicing salt. Some efforts have been made to identify these aggregates by pretreating the aggregates with sodium chloride prior to making AASHTO T 161 concrete specimens (Dubberke and Marks 1985). Desta et al. (2015) described the redevelopment of the hydraulic fracture test equipment with major renovations, sample preparation procedures, and analysis of results to produce a reliable 8-day test that predicts the results of the 90-day AASHTO T 161 test with significant accuracy. Indiana DOT has developed a new testing specification based on the new hydraulic fracture test (Desta et al. 2015). The new hydraulic fracture test is a simpler, quicker test using equipment that is less expensive to buy and maintain than equipment used for AASHTO T 161, which can lead to both cost reduction and cost avoidance. With a quicker test, aggregate quality can be ensured, especially if the aggregate source needs to be changed during construction or for sources with highly variable aggregate quality, making these sources a more viable option. Note: AAR = alkali-aggregate reactivity. * Because roughness is affected by the presence of distresses, any aggregate properties that influence the development of those distresses will also influence the development if roughness. ** Surface friction is mainly affected by the polish resistance of fine aggregates because of the presence of the mortar-rich layer at the top surface of PCC pavements. Table 5-6. Aggregate properties determining quality linked to corresponding concrete pavement performance parameters.

Aggregate Quality Affecting Pavement Performance 67 5.2.3 Unsurfaced Pavements The Massachusetts Unpaved Roads Best Management Practices Manual: A Guidebook on How to Improve Water Quality While Addressing Common Problems describes the main distresses in unpaved roads as surface deteriorations and surface deformations. Deterioration can be caused by losing fine materials to the air as dust or losing coarse aggregates from raveling. Deformation distresses include rutting, depressions, corrugations, and the formation of potholes, caused by poor gradations or excessive water due to high moisture content or drainage issues (Berkshire Regional Planning Commission 2001). The FHWA’s gravel roads construction and maintenance guide notes that the surface gravel used for the construction of unsurfaced pavements (gravel roads) has higher percentages of sand-sized particles that will facilitate drainage and are easier to bind compared with conventional base courses that have larger top size particles (FHWA 2015). Due to the criticality of drainage and binding characteristics, aggregate quality checks are required for constructing good performing gravel roads. Among the quality tests for aggregates, the percentages of materials passing the No. 200 (0.075-mm) sieve with a commonly required value to fall between 4% and 15% and the Atterberg limits test were listed (FHWA 2015). Participating transportation agencies identified the aggregate quality-related surface treat- ment or unpaved road distresses they have encountered. In total, 15 U.S. DOTs and six Canadian provinces indicated at least one aggregate quality-related distress. Those commonly reported distresses are listed as follows: • Aggregate breakdown/popouts under traffic (too soft/deleterious), and dirty aggregate • Bleeding • Aggregate polishing and loss of frictional properties (most common) • Chip seals debonding from roadway • Raveling and delamination • Rutting • Surface cracking • Oxidation • Stripping of chert gravels Refer to Appendix C for a detailed compilation of the individual distresses reported by each agency. 5.3 Aggregate Quality Indicators Linked to Pavement Performance 5.3.1 Aggregate Quality Indicators Physical characteristics of the particles affecting load dissipation and interlocking aspects often differentiate “good quality” from “poor quality” aggregates used in pavement construction. Chemical properties of the aggregates controlling their durability and soundness are also critical to ensure long-lasting pavement structures (Tutumluer 2013). Participating transportation agencies indicated whether they link aggregate quality to the per- formance of a certain pavement layer. The results showed that 32 U.S. DOTs and seven Canadian provinces do not collect information regarding how the quality of aggregate used in construction may affect pavement end performance. Whereas 11 U.S. DOTs and New Brunswick province indicated that they record or link aggregate quality to the performance and reported certain specific aggregate quality or source deficiency issues causing poor performance of a certain pave- ment layer. Accordingly, Table 5-7 lists the number of transportation agencies that linked an aggregate quality deficiency to a specific performance issue in a given pavement layer. The total

68 Aggregate Quality Requirements for Pavements is 12 respondents. Refer to Appendix C for a detailed compilation related to individual responses reported by each agency. Under repeated traffic loading and environmental conditions, aggregate particles are continu- ously subjected to degradation through attrition, impact, and grinding- and polishing-type mechanisms. Physical and chemical properties of aggregates can affect strength, durability, modulus, and deformation behavior in unbound and bound pavement layers (Tutumluer 2013). A summary on the physical properties affecting aggregate quality is presented in the following text. NCHRP Report 405: Aggregate Tests Related to Asphalt Concrete Performance in Pavements (Kandhal and Parker 1998) identified several aggregate properties that affect the performance of HMAs. These properties are (1) gradation and size; (2) particle shape, angularity, and surface texture; (3) porosity and absorption; (4) cleanliness and deleterious materials; (5) toughness and abrasion; (6) durability and soundness; (7) expansive characteristics; (8) polish resistance and frictional characteristics; (9) mineralogy and petrography; and (10) chemical properties. The study investigated several aggregate properties such as particle shape, angularity, and surface texture, toughness and abrasion, durability and soundness, as well as the properties of fine materials pass- ing No. 200 (0.075-mm) sieve. According to NCHRP Report 405, the following nine properties of aggregates are linked to the performance of HMA pavement layers: (1) sieve analysis/gradation is related to fatigue and permanent deformation; (2) uncompacted voids content of coarse aggregates is related to fatigue and permanent deformation; (3) flat and elongated ratio of coarse aggregate is related to fatigue and permanent deformation; (4) uncompacted voids content of fine aggregates is related to fatigue and permanent deformation; (5) Methylene blue test is related to stripping and permanent deformation; (6) D60 and D10 gradation properties are related to stripping and permanent deformation; (7) Methylene blue test of materials passing sieve No. 200 (0.075 mm) Aggregate Quality Issue Pavement Layer ASC ABC PCC BC SBC DR FI S_BC S_SBC ST Source deficiency 6 2 5 2 - - - - - 3 Blending 1 - - - - - - - - - RAP use 6 5 - - - - - - - 2 RCA use - - 1 1 1 - - - - - SBS use 2 1 1 - - - - - - 1 BFS use 1 - - - - - - - - - Weathering soundness 3 2 1 - - - - - - 1 Degradation resistance 2 1 2 1 - - - - - 2 Polishing resistance 8 - 3 - - - - - - 4 Plasticity of fines 3 2 1 1 - - - - - 1 Mineralogical composition 1 1 1 - - - - - - 1 Clay content 2 1 2 1 - - - - - 1 Particle shape 1 - - 1 1 1 1 1 1 1 Durability: Freeze-thaw 2 1 5 - - - - - - 1 ASR - - 6 - - - - - - - Note: Hyphens indicate zero (0) transportation agency. ASC = asphalt surface course, ABC = asphalt base course, BC = base course, DR = drainage layer, FI = filter layer, SBC = subbase course, S_BC = stabilized base course, S_SBC = stabilized subbase course, and ST = surface treatment. Table 5-7. Number of transportation agencies that reported a link between an aggregate quality issue and poor performance of a given pavement layer.

Aggregate Quality Affecting Pavement Performance 69 is related to stripping and permanent deformation; (8) micro-deval test is related to potholes, raveling, and popouts; and (9) magnesium and sulfate soundness are related to potholes, ravel- ing, and popouts (Kandhal and Parker 1998). A summary of the aggregate performance tests and correlation values is given in Table 5-8 (Kandhal and Parker 1998). In this table, the correlation coefficient (R) ranges between –1 and 1 and refers to the linear dependence or goodness of fit between a specific test and the performance of the HMA layer. The significance level (P) refers to the probability of rejecting the null hypoth- esis when it is true. Higher absolute values of R and lower P values indicate a better correlation between the prospective test and the performance of the HMA layer. The research in NCHRP Report 557: Aggregate Tests for Hot-Mix Asphalt Mixtures Used in Pavements (White et al. 2006) investigated several aggregate performance tests for aggregates used in asphalt pavements. The study concluded that micro-deval and magnesium sulfate soundness were important aggregate performance tests for all materials, traffic conditions, and climates. Accordingly, White et al. (2006) proposed maximum limits of 15% and 20% for micro- deval and magnesium sulfate soundness tests, respectively. 5.3.2 Mineralogy Mineral composition of aggregates has a significant effect on the physical and chemical char- acteristics that ultimately govern the performance of unbound aggregate base or subbase layers under loading. This is particularly true as far as degradation and polishing due to interparticle friction is concerned. Calcareous aggregates like limestone and dolomite show significantly lower resistance to particle degradation and polishing. Therefore, unbound aggregate base or subbase Table 5-8. Aggregate performance tests and correlation coefficients related to HMA pavement layer performance.

70 Aggregate Quality Requirements for Pavements layers constructed using these aggregates are likely to undergo significant changes in gradation during compaction and subsequently under traffic loading (Tutumluer 2013). 5.3.3 Plasticity of Fines Increasing the percentage of fines passing the No. 200 sieve (0.075 mm) in a mix reduces the permanent deformation resistance (Barksdale 1972, 1991, Thom and Brown 1988). The type of fines (nonplastic or plastic) in an aggregate layer has also been found to significantly affect the performance. The results from the Illinois DOT field study, Experimental Feature IL 03-01, indicated that increased aggregate fines had a significant effect on their performance in working platform applications (Illinois DOT 2005). Tutumluer et al. (2009) investigated the effect of the plasticity of fines on the performance of three different types of unbound aggregates (limestone, dolomite, and uncrushed gravel) under repeated loading (resilient modulus and permanent deformation tests) and the effect of the plasticity of fines on moisture susceptibility and strength. The study found that (1) the presence of low amounts of nonplastic fines in an aggregate layer may not adversely affect the pavement performance. For plastic fines, however, the values of the immediate bearing value showed rapid decrease with moisture contents even at low fines contents, thus indicating increased moisture sensitivity; (2) increasing the amount of fines did not result in significant decreases in aggregate modulus and strength behavior in the case of nonplastic fines but had a drastic effect in the case of plastic fines; (3) plasticity of fines (with PI > 10) is the second most important property that affected aggregate behavior (after particles angularity). High amounts of plastic fines at wet of optimum quickly destruct the aggregate load transfer matrix and result in excessive permanent deformations; and (4) for low percentages of nonplastic fines, moisture content did not have a significant effect on aggregate performance and often aggregate type or angularity was the governing factor. However, for aggregates with plastic fines, moisture becomes the most governing factor for aggregate behavior (Tutumluer et al. 2009) 5.3.4 Particle Shape, Surface Texture, and Angularity The morphological or shape properties of aggregate particles significantly affect the perfor- mance of unbound/bound layers used as highway pavements. Such constructed aggregate layers were evaluated in terms of shear strength, modulus, and permanent deformation characteris- tics (Kandhal and Parker 1998, Meininger 1998, Tutumluer and Pan 2008). Road safety from the perspective of frictional resistance is a function of pavement microtexture, which is highly influenced by the magnitude of aggregate surface texture and angularity (Dunford 2013, Forster 1989, Luce et al. 2007). Increasing particle angularity and roughness was found to increase the resilient modulus and decrease the Poisson’s ratio (Hicks and Monismith 1971, Allen and Thompson 1974), Thom (1988, Thom and Brown 1988), Allen (1973) and Barksdale and Itani (1989) found that angular materials resisted permanent deformation better than rounded particles because of the improved particle interlock and higher angle of shear resistance between particles. Thom and Brown (1988) observed that permanent deformation was primarily affected by visible roughness of particles. Barksdale and Itani (1989) also concluded that blade-shaped, crushed particles are slightly more susceptible to rutting than other types of crushed aggregate and that cube-shaped, rounded river gravel with smooth surfaces is more susceptible to rutting than crushed aggregates. Saeed et al. (2001) showed a linkage between aggregate properties and unbound layer performance and that aggregate particle angularity and surface texture mostly affected shear strength and stiffness. Angularity index values measured with UIAIA for 50-50 blends of the crushed stone and gravel correlated well with the shear strength properties from triaxial tests (Rao et al. 2002).

Aggregate Quality Affecting Pavement Performance 71 Furthermore, Pan et al. (2006) studied the effects of particle angularity and surface texture for 21 blends of uncrushed and crushed aggregate sources on the resilient modulus and permanent deformation behavior of unbound granular materials and concluded that both angularity and surface texture were closely linked to modulus and deformation. The effect of aggregate shape, texture, and gradation on the performance of fresh concrete has also been investigated by Quiroga and Fowler (2004), who observed that aggregate blends with cubical shape and rounded and smooth particles required less paste at a given slump as opposed to blends with flat, elongated, angular, and rough-surfaced particles. The effects of shape prop- erties of aggregates quantified by digital image processing has been studied on the compressive strength properties of cement concrete (Polat et al. 2013). The study concluded that spherical particles were desirable for increased compressive strength, unit weight and slump. The research in NCHRP Synthesis 539 focused on aggregate properties and the performance of Superpave-designed HMA (Prowell et al. 2005). The study showed that increasing coarse aggregate fractured faces or angularity increased rutting resistance. Additionally, increased particle index value or uncompacted voids in coarse aggregates also provided increased rutting resistance. The latter combines the effect of form (particle shape and often 3-D geometry), angularity, and surface texture associated with each particle. The relationship between aggregate surface texture and asphalt pavement skid resistance was studied by Masad et al. (2007) in Final Report for Highway IDEA Project 114 by using a second generation of aggregate imaging system (AIMS-II). The findings verified that skid resistance not only was related to the average aggregate surface texture but also to the texture distribution within an aggregate sample. Accordingly, aggregate surface texture distribution could be considered in developing accurate performance models to predict asphalt pavement skid resistance. Frictional resistance or skid resistance is an indicator for the performance of asphalt pavements (Mahmoud and Perales 2015). The frictional resistance is a function of the macrotexture and microtexture of the pavement surface (Dahir 1979, Mahmoud and Perales 2015). The macro- texture is primarily influenced by the aggregate shape properties of the mix, and the polishing resistance of aggregates needs to be tested to ensure the quality of the aggregates that are being used for the surface coarse HMA mixes. Mahmoud and Perales (2015) concluded that surface texture is a better indicator of aggregates quality to use for ranking aggregate materials for performance when compared with angularity. 5.3.5 Resistance to Degradation The most important factors that cause in-service aggregate degradation include (1) material source properties including mineralogy and petrography, (2) shape and angularity of aggregate particles, (3) initial size distribution and arrangement of particles, (4) force concentration on particle surfaces, and (5) aggregate layer maintenance operations and environmental conditions (Tolppanen 2001). Saeed et al. (2001) reported in NCHRP Report 453 that LAA and micro-deval testing were among the most important tests that influence the performance of aggregates used in unbound pavement layers. Several abrasion/degradation testing procedures for correlating the laboratory testing results with the field performance of asphalt pavement (Wu et al. 1998) were compared. The results showed that micro-deval and magnesium sulfate soundness tests provide the best correlations with field performance. Several studies in the literature investigated the characterization of aggregate degradation and its effects on the bearing capacity of unbound/bound layer from the perspectives of change in

72 Aggregate Quality Requirements for Pavements size distribution or decrease in coarse-to-fine fraction ratio (Gatchalian et al. 2006, Lynn et al. 2007, Pintner et al. 1987). It is important to note that, on the one hand, aggregate degrada- tion can cause abrasion, which results in particles losing their angularity and surface texture or becoming more rounded and spherical. Aggregate degradation changes the void ratio or packing properties and ultimately influences the performance. On the other hand, aggregate particles after breakage possess fresh surfaces that might have sharp edges and rougher surface texture (Moaveni 2015). Several studies also focused on comparing LAA wear, micro-deval, and impact and crushing wear. Brandes and Robinson (2006) reported that micro-deval correlated best with magne- sium and sodium soundness degradation for aggregates used in HMA when compared with LAA and another dry test, such as impact and crushing, which showed a lower correlation. Test results for degradation of coarse aggregates used in the state of Virginia showed less vari- ability and more repeatability with micro-deval as compared with soundness test. Micro-deval showed better ability to differentiate the quality of good-performing and poor-performing aggregates (Hossain et al. 2008a). Another study for the durability of fine aggregates in Virginia proposed micro-deval over other conventional durability tests due to higher pre- cision and correlation to performance from 10 aggregate sources (Hossain et al. 2008b). Similarly, Rangaraju and Edlinski (2008) reported a high correlation between aggregate per- formance and micro-deval with 18% permissible loss, while sulfate soundness and LAA did not correlate well with performance given South Carolina DOT’s specifications of 15% and 55% loss, respectively, for these tests. The study also reported higher micro-deval loss for aggregates with finer gradations (finer micro-deval-C versus micro-deval-B/micro-deval-A gradations were investigated) (Rangaraju and Edlinski 2008). Similar effect of grain size dis- tribution on micro-deval loss was also reported for aggregates from 72 sources in the state of Maine (Nener-Plante 2013). Erichsen et al. (2011), on the other hand, concluded that a suitable durability test is dependent on aggregate quality and the method of degradation in the field: LAA test correlates better with degradation by fragmentation while micro-deval and Nordic abrasion tests correlate better to degradation by wearing, in which a more poorly graded gradation curve is obtained after the tests (Erichsen et al. 2011). Similarly, Cooley and James (2003) tested aggregates from 72 sources in eight states with micro-deval and LAA and reported a poor correlation between micro-deval, sodium sulfate soundness, and LAA for the same aggregates, indicating that these tests are essen- tially measuring different properties, depending on the aggregate source. A study for aggregates used in Alaska concluded better correlation between performance of aggregates in HMA and micro-deval test when compared with the Washington degradation test that is currently used in the state (Liu et al. 2017). 5.3.6 Aggregate Performance Tests Shear strength tests (triaxial tests conducted on wet and dry samples) as well as stiffness tests (resilient modulus conducted on wet and dry samples) have been reported as the most relevant tests for characterizing the strength, modulus, and permanent deformation behavior of aggre- gates used in pavement layers (Saeed et al. 2001). Participating transportation agencies identified as the performance-related laboratory tests that they performed on aggregate sources before use in pavement construction, as shown in Figure 5-2. Out of the seven agencies that marked other alternatives to evaluate the field performance of aggregates, two agencies did not report any test, while Alaska DOT reported that it performs AASHTO T 307 only for research purposes. The other four agencies mentioned the specific tests listed in Table 5-9.

Aggregate Quality Affecting Pavement Performance 73 5.4 Performance Concerns of Recycled and Artificial/Byproduct Aggregates Several limitations exist for using QB as aggregates in pavement applications, mainly due to performance concerns and other logistical factors. NCHRP Synthesis 435: Volume 4 lists some of the performance concerns for using QB: (1) consistency of the composition of QB and pos- sible variabilities are not well studied or understood, (2) guidelines for best practice of using QB are not readily available, (3) physical and mineralogical information of QB can be limited or unavailable, (4) knowledge about local sources can be limited, and (5) higher absorption values than other aggregates can lead to excessive Portland cement or binder requirements depending on the application (Stroup-Gardiner and Wattenberg-Komas 2013b). For using RAP as a recycled aggregate in pavement applications, the quality of the binder and aggregates can be questioned. This issue may concern many agencies and cause them to strictly use RAP coming from their own projects (Hoppe et al. 2015). Some agencies limit the quantity of RAP used in asphalt to control the dust content. One limitation of using RAP in asphalt mixtures is the quantity of dust associated with the milling process of RAP. Agen- cies can also specify limited quantities of RAP in asphalt mixtures based on the total binder content required and the percentage of binder replacement from the RAP due to possible changes in the performance grade with the use of aged binder from RAP (National Asphalt Pavement Association 2015). For example, the brittle and stiff behavior of mixes contain- ing RAP and adjustment requirement to the virgin binder was indicated by New York DOT (Wagoner et al. 2005). The higher quantities of fines passing the No. 200 (0.075-mm) sieve associated with RAP was also reported as one of the main limitations to Superpave mixes (Stroup-Gardiner and Wattenberg-Komas 2013c). 57% 29% 43% 10% 33% 12 6 9 2 7 0 10 20 0% 20% 40% 60% 80% 100% Skid resistance tests, e.g., British Pendulum or similar Triaxial shear strength tests Repeated load triaxial resilient modulus test (AASHTO T 307, *NCHRP 1-28) Repeated load triaxial permanent deformation test Other Number of Responses Percentage of Survey Respondents 21 survey respondents Figure 5-2. Performance-related laboratory tests performed on aggregate sources before use in pavement construction (*MR test procedure). Aggregate Performance-Related Test British pendulum only on limestone aggregates AASHTO T 283, Resistance of Compacted Bituminous Mixture to Moisture Induced Damage for Superpave Insoluble residue test as a measure of skid (polishing) resistance in asphalt surface layer Agency Alabama New Mexico Oklahoma Wyoming R-value testing Table 5-9. Alternative aggregate performance-related tests considered by participating transportation agencies.

74 Aggregate Quality Requirements for Pavements Findings from the Transportation Pooled Study TPF-5(129) (Edil et al. 2012) showed that although RAP materials could show very high resilient modulus values, unbound aggregate specimens tested in the laboratory and base/subbase layers constructed in the field using 100% RAP materials often accumulated high permanent deformations. A similar finding was also reported for using 100% RAP in construction platforms by researchers at the Illinois Center for Transportation (Kazmee and Tutumluer 2015, Kazmee et al. 2017). Expansion of RAP is particularly critical when it contains expansive components like steel slag. Steel slag aggregates are often used in HMA surface courses because of their high fric- tional characteristics. Therefore, any RAP material obtained from these surface courses with steel slag aggregates may potentially lead to expansion and resulting pavement heave when used in unbound aggregate base/subbase courses. Recent experiences with volume changes of up to 10% or more have been attributable 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 within a few weeks, while magnesium oxide hydrates more slowly and contributes to long-term expan- sion that may take years. The potential expansion 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) reported that RAP materials had much lower tendencies to expand when compared with the expansion potentials of virgin steel slag aggregates, likely due to the asphalt coating around the aggregates that prevents any significant ingress of water into them. Depend- ing on the level of expansion and the material gradation, dense-graded aggregate base applica- tions with steel slag may have to be avoided (Deniz et al. 2010). Ooi et al. (2011) listed the following criteria for using RCA as a base course. The criteria are to (1) allow only uncrushed concrete that can be visually inspected for use as RCA, (2) accept RCA only from suppliers that can guarantee the quality, (3) not accept RCA from unknown sources unless certified by a qualified engineer or scientist as being free from deleterious materials (such as aluminum), (4) avoid using building demolition RCA, (5) require a paper trail to document the RCA source, and (6) use a nonferrous metal detector to determine whether aluminum is present and also to inspect the RCA visually before use. Reza and Wilde (2017) mentioned some quality concerns regarding the use of RCA in Portland cement applications. These pertained to concerns over the source of the RCA and possible dis- tresses in the RCA, exhibited due to material quality, such as D-cracking and ASR. Other con- cerns included issues with workability and product quality related to the use of RCA in concrete pavements. High absorption, durability concerns, higher drying shrinkage, and lower modulus of elasticity are also among the quality concerns related with the use of RCA in PCC (Reza and Wilde 2017). Several studies also reported leaching to groundwater and high pH levels as issues with RCA (Engelsen et al. 2012, Reza and Wilde 2017, Sani et al. 2005). NCHRP Synthesis 435: Volume 5 identified several limitations for using iron and steel slags in pavement applications including use in HMA and PCC layers, as summarized in Table 5-10, for BFS, SFS, and other types of slags. Table 5-10 lists slag availability, slag material properties, and regulations as the most frequent barriers for using combustion byproducts in pavement applications. Undesirable material properties include high specific gravity, high absorption, and issues with freeze thaw. One limitation for steel slags is that it can lead to decreased compatibility and workability and high hauling costs due to its high specific gravity and unit weight (Stroup- Gardiner and Wattenberg-Komas 2013a). Due to leaching, expansion, and other past issues experienced by agencies with the use of steel slag, the 17 agencies that use steel slag aggregates, as listed in NCHRP Synthesis 435: Volume 5, adopted different criteria for acceptance in applications other than HMA. Some agencies specified an

Aggregate Quality Affecting Pavement Performance 75 expansion test either by the DOT or by the supplier (Illinois, Indiana, Minnesota, Ohio, South Carolina, and West Virginia) and some states required moisture curing prior to use. For example, Ohio requires 1-month moist curing, Missouri requires 3 months, while South Carolina and West Virginia require 6 months. None of the states listed above permit using steel slag aggregates in PCC (Washington State DOT 2015). For slag use in concrete pavements, it has been reported that dissolution of calcium sulfide in air-cooled BFS might lead to paste expansion and cracking, but the extent is not well known. Concrete toughness, loss of workability, and maintenance costs (which can be twice as high compared with using natural aggregates from the experience of Michigan DOT) have been cited as issues for using BFS (Morian et al. 2012). Leachate from BFS poses a potential risk to the environment and remains an aesthetic concern. This requires checking the odor and discoloration of water and determining the properties of leachate from BFS such as pH and redox conditions. The Recycled Materials Resource Center lists several quality assurance measures to reduce leachate issues, including stockpiling the air- cooled BFS for a minimum of 1 month before usage and using air-cooled BFS above grade only and in areas were drainage is not problematic and the BFS is not immersed in water (RMRC 2017). For example, bucket test for leachate determination can be performed prior to construction, which is the current practice by Illinois DOT (Illinois DOT 2015, RMRC 2017). Additionally, the hardness of BFS can be measured by Moh’s scale ranges from 5 to 6, which is similar to that of durable igneous rock. Nevertheless, BFS is brittle and easy to breakdown when subjected to impact loading (RMRC 2017). No correlation between LAA test for BFS and its degradation performance in the field has been reported (Chesner et al. 1998). Therefore, some of the states do not consider the degradation requirements based on LAA testing for utilization of BFS. On the other hand, the EPA studied the leachate results of SFS along with the toxicity characteristic leaching procedure regulatory limits. The results showed that the average values of metals under consideration did not violate the limits set in toxicity characteristic leaching procedure regulation levels (Stroup-Gardiner and Wattenberg-Komas 2013a). Therefore, it was Barrier category Reasons for classification as barrier States with barrierresponses Question: Comment on barriers to the use of combustion byproducts in highway applications that have been either overcome or still exist Table 5-10. Limitations for using combustion byproducts in pavement applications.

76 Aggregate Quality Requirements for Pavements concluded that the use of SFS was safe and did not induce any contamination even in areas where the groundwater table was shallow. Participating transportation agencies indicated whether they have any environmental (e.g., leaching) or performance (e.g., cracking) concerns regarding the use of recycled aggregate (RAP and RCA) or artificial/byproduct aggregates (SFS and BFS) in pavement construction. It was found that 57% (26 U.S. DOTs and three Canadian provinces) have experienced at least one environmental or performance-related concern. Those 29 agencies described any environmental or performance issues of their concern, and their concerns are listed in Table 5-11. Using RAP and RCA in pavement applications was linked to the two most commonly encountered concerns among different agencies. High percentages of RAP were reported to create premature asphalt cracking while using RCA involved higher pH and leaching concerns. Refer to Appendix C for a detailed compilation related to individual responses reported by each agency. Description of Environmental and/or Performance Concern Concerned with potential leaching with RCA. Concerned with RCA in rigid or flexible pavement due to amount of fines from crushing. Too high of a RAP content (50% range) resulted in premature cracking. Any recycled materials used in base courses must be tested for environmental compliance by a contractor prior to delivery. Not allowing RCA in concrete due to the fact it may have ASR in it. Higher pH of RCA if used in same area as metallized pipe. Leaching of steel furnace slag. RCA cannot be used in ephemeral drainages or high-water table conditions. Bituminous mixtures replacing 20% of the binder with RAP experiencing raveling and early oxidation making the mix brittle and prone to cracking. Leaching. However, mostly large stockpile of RAP stored at quarry as by product of milling. Leaching of high pH water leaving the right-of-way from the use of RCA and BFS. The formation of tufa in underdrains due to the use of RCA. Hot Mix: Polishing of coarse aggregates caused by the contractor using a higher percentage and less quality RAP in our premium surface courses than is indicated in the mix design. Unbound Base Layers: Leaching of blast furnace slag. Concerned with loss of fatigue resistance in surface courses when using RAP. Agency Alberta, British Columbia, Kentucky, Maine, Pennsylvania, Tennessee, and Washington State Arkansas British Columbia, Idaho, Kansas, Kentucky, New York, and Utah Connecticut Michigan, Montana, and South Dakota Missouri New Jersey Delaware Florida and Maryland Illinois and Indiana Ohio Ontario Rhode Island Texas Need to meet requirements included in Specification: DMS-11000: Evaluating and Using Nonhazardous Recyclable Materials Guidelines. Table 5-11. Environmental or performance issues reported by agencies using RAP, RCA, or artificial/byproduct aggregates (SFS and BFS) in pavement construction.

Next: Chapter 6 - Conclusions »
Aggregate Quality Requirements for Pavements Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB's National Cooperative Highway Research Program (NCHRP) Synthesis 524: Aggregate Quality Requirements for Pavements Aggregate Quality Requirements for Pavements documents transportation agency requirements for the quality of aggregates for various pavement types. Constructing and maintaining pavements requires an abundant and dependable supply of quality aggregates. Aggregate comes from a wide range of materials, including quarried rock, sand, and gravel, and materials such as slag, reclaimed asphalt pavement, and recycled concrete aggregate. While all transportation agencies have specifications for aggregate quality, there is wide variation in what different agencies consider suitable aggregates for specific applications.

The report is accompanied by the following appendices:

  • Appendix A: Survey Questionnaire
  • Appendix B: Survey Respondent Information
  • Appendix C: Compilation of Survey Responses Provided by Agency Respondents
  • Appendix D: Links to Approved Aggregate Lists and Specifications Published by Agencies
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!