Aggregate Quality Requirements for Pavements (2018)

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.

29 3.1 Introduction This chapter presents different approaches used by transportation agencies for sampling, testing, and controlling the quality of aggregates used in pavement construction. QA methods have been established by many transportation agencies to ensure that the aggregate materials received on the job site have properties to meet desired criteria for utilizing them in con- struction of a certain pavement layer. Protocols or specifications used by transportation agencies to ensure aggregate quality include (1) method specifications, (2) material speci- fications, and (3) end-result specifications (Dukatz and Marek 1986). Findings from the survey related to different testing types, frequencies, and procedures performed by agen- cies on virgin coarse and fine aggregates, RAP, RCA, SFS, and BFS are summarized in this chapter. Finally, the procedures used to control the quality of blended aggregate products are discussed. Participating agencies identified the responsible authority who they put in charge for test- ing aggregate materials and providing aggregate properties for the design of pavement layers. Figure 3-1 presents the results. Figure 3-1 indicates that 81% of the respondents use in-house geotechnical or material labo- ratories to control the aggregate quality. Additionally, 42% rely on contractor’s laboratory and testing. And about 30% of the agencies use testing results from producers or external geotechni- cal laboratories/private consultants for quality assurance purposes. The alternatives included in Figure 3-1 are considered as the first stage in aggregate quality assurance since in case of a dispute, private testing laboratories or third parties are often hired to confirm the results. Note that multiple option selections were made by survey respondents. 3.2 Aggregate Sampling and Testing Transportation agencies provided information on the frequencies of quality assurance tests conducted on aggregate materials. Results are presented in Figure 3-2. Out of 52 respondents, 24 agencies or 46% check the acceptance of materials once a year while 20 agencies or 39% check the quality on every major construction project. Forty percent of the agencies indicated other approaches or procedures for aggregate material acceptance according to the following general categories: • Based on source history, location, tons of aggregates produced, or volume used in the project; • Depending on what the aggregate is being used for; • During both production and design phase; C H A P T E R 3 Aggregate Sampling, Testing, and Quality Assurance

30 Aggregate Quality Requirements for Pavements • Through producer’s QC program at the quarry and according to reduced frequency verification tests mandated by transportation agency; and • Out-of-state quarries are verified quarterly. However, out-of-country quarries are mandato- rily tested at redistribution terminals. Sampling requirements also varied among different agencies. Most agencies specify sampling techniques for aggregates that comply with a standard, such as AASHTO T 2 (which refers to ASTM D75), Standard Method of Test for Sampling of Aggregates. For example, Washington State DOT specifies AASHTO T 2 for sampling requirements for a source approval and requires limits on aggregate sample cleanliness, sample sizes, and aggregate top sizes based on the aggre- gate type/application, as indicated in Table 3-1 (Washington State DOT 2017). Similarly, North Dakota DOT lists AASHTO T 2, AASHTO R 76, and ND T 248 for aggregate sampling and 81% 32% 6% 30% 41% 17% 43 17 3 16 22 9 0 10 20 30 40 50 0% 20% 40% 60% 80% 100% In-house geotechnical/material laboratories Retained external geotechnical consultant/materials laboratory University laboratory (under research subcontract) Aggregate producer Contractor testing and laboratory Other Number of Responses Percentage of Survey Respondents 53 survey respondents Figure 3-1. Responsibility for testing aggregate materials and providing aggregate properties. 39% 12% 4% 46% 10% 40% 20 6 2 24 5 21 0 10 20 30 40 50 0% 20% 40% 60% 80% 100% Prior to use on every major construction project More than twice every year Twice every year Once a year Less than once a year Other Number of Responses Percentage of Survey Respondents 52 survey respondents Figure 3-2. Frequency of aggregate acceptance checks.

Aggregate Sampling, Testing, and Quality Assurance 31 reducing samples of aggregates to testing sizes. Their specification includes standard proce- dures for sampling aggregates from a roadway, flowing stream, conveyor belt, stockpile, or truck to ensure representative samples (North Dakota DOT 2015). The State of Alaska also uses AASHTO T 2 for aggregate sampling (Alaska DOT 2016). The Illinois DOT manual of test procedures for materials includes detailed procedures for sampling aggregates from a belt-stream, bin discharge, truck dumps, and stockpile sampling (Illinois DOT 2015). New Jersey DOT requires obtaining, over a period of several weeks, three samples from each aggregate size for testing before any source approval (New Jersey DOT 2015). Idaho DOT requires that the testing of aggregates is performed by an independent laboratory approved by the agency (Idaho Transportation Department 2014). North Carolina DOT and North Carolina Aggregates Association (NCAA) jointly published a comprehensive aggregate QA program that lists procedures and requirements for sampling at aggregate production and construction sites. The program includes requirements for sampling, record keeping, sampling checks, verification sampling at the source and job site, and AASHTO or ASTM tests that are performed on aggregates (North Carolina DOT and NCAA 2017). Additionally, several agencies provide lists of qualified suppliers/producers, which have been pre-approved as sources for aggregates in pavement applications. Texas DOT’s aggregate quality monitoring program, for example, published a list of qualified sources for aggregates for bituminous and concrete rated applications. Typical source property values are summarized in these catalogs. Examples of these source qualities include LAA, magnesium sulfate soundness, and micro-deval (Texas DOT 2007). Agencies may also require the contractors to perform tests to check the quality of the aggre- gate materials being constructed. For example, Arizona DOT requires several quality control test results from contractors, including the sand equivalent test and fractured coarse aggregate particles for mineral aggregates, as listed in Table 3-2 (Arizona DOT 2008). Transportation agencies reported how they obtained aggregate samples from project sites to perform required tests. Figure 3-3 presents the results. Figure 3-3 shows that 81% of the responding agencies obtain samples and perform quality tests in their own material laboratories. Moreover, 18 agencies (34%) do the sampling and checking at the Table 3-1. Washington State DOT requirements for sampling sizes and sampling locations.

32 Aggregate Quality Requirements for Pavements source (quarry) location while 13 agencies (25%) accept samples shipped from aggregate producers for testing in their agency laboratories. Some agencies assign the sampling to the contractor and then the quality tests are performed by a consultant/contractor. Finally, there were again multiple option selections made by survey respondents. This implies that different practices may be taking place in one agency, depending on what pavement application that aggregate source is used for. 3.2.1 Quality Tests Conducted on Virgin Aggregates Chapter 2 introduced the major tests used to check the quality of virgin, recycled, artificial, and byproduct aggregate materials used in various pavement applications. This section discusses typical test ranges and some DOT specifications and tests required for individual pavement layers. 3.2.1.1 Virgin Coarse Aggregates Out of 51 responding agencies, 96% (42 U.S. DOTs and seven Canadian provinces) reported that they perform at least one test for checking virgin aggregate quality requirements for construction of pavement layers. Those 49 agencies were asked to indicate if they performed any virgin coarse aggregate quality tests for Na2SO4/MgSO4 soundness, deleterious materials, particle angularity, flat and elongated ratio, freeze-thaw, polishing/skid resistance, LAA loss, Type of Test Test Method Sampling Point Minimum Testing Frequency Gradation ARIZ 201 Crusher belt or stockpile 1 per stockpile per day Sand equivalent AASHTO T 176 Crusher belt or stockpile 1 per 2,000 tons of total aggregate Fractured course Aggregate particles ARIZ 212 Uncompacted void content ARIZ 247 Table 3-2. Arizona DOT quality control testing requirements for contractors. 25% 81% 34% 13% 13 43 18 7 0 10 20 30 40 50 0% 20% 40% 60% 80% 100% Samples shipped from aggregate producer and tested in agency (DOT) lab Samples obtained by agency and tested in agency (DOT) lab Samples checked/inspected at the source (quarry) location Other Number of Responses Percentage of Survey Respondents 53 survey respondents Figure 3-3. Procedures for obtaining aggregate samples from project sites for quality tests.

Aggregate Sampling, Testing, and Quality Assurance 33 and porosity to also report the testing procedures, for example, ASTM, AASHTO, or agency- developed specification. The number of agencies at each category for individual tests are summarized in Table 3-3. Table 3-3 indicates that 17 agencies do not perform porosity tests on the virgin coarse aggre- gates and that more than 10 agencies do not run polishing/skid resistance and freeze-thaw tests on their coarse virgin aggregate sources. Accordingly, 41 agencies follow the AASHTO procedure to conduct LAA testing, while 22 agencies follow the ASTM test procedure for measuring flat and elongated ratios of virgin coarse aggregate samples. Moreover, 18 agencies have their own pro- cedures for determining deleterious materials. Refer to Appendix C for a detailed compilation of the results given in Table 3-3. Additional tests following other specific test procedures performed by agencies to determine quality of virgin coarse aggregates are listed in Table 3-4. Kansas DOT requires some agency specific test procedures for coarse aggregates in order for an aggregate source to pass their quality requirements. These test procedures are KTMR-21 for soundness, KT-6 for specific gravity and absorption and, if requested, KTMR-28 for acid insolu- ble residue. For coarse aggregates used in concrete applications, Kansas DOT also requires addi- tional tests for durability (KTMR-22) and wetting and drying (KTMR-23) (Kansas DOT 2017). Alaska DOT runs its own test method (ATM 306) for determining the percentage of flat and elongated particles in coarse aggregate (Alaska DOT 2016). The state of Arizona conducts its own test method for fractured coarse aggregate particles, known as ARIZ 212, and ARIZ 251 for the dry and bulk specific gravity (Arizona DOT 2008). Different quality test ranges, as required by Arizona DOT in asphaltic concrete are shown in Table 3-5. Different pavement applications can have different ranges for acceptance by Arizona DOT. The specifications and ranges for a test vary significantly from one agency to another agency. A survey by Prowell et al. (2005) indicated different ranges for acceptance of aggregate materials to be used in Superpave HMA mixtures. As shown in Figure 3-4, most states specify an LAA loss limit of 40% for HMA mixes, with limits varying from 30% to 55%. Some states also specify LAA loss limits based on the class of the road, the equivalent single-axle loads, or the type of aggregates (Prowell et al. 2005). 3.2.1.2 Virgin Fine Aggregates According to NCHRP Report 539: Aggregate Properties and the Performance of Superpave- Designed Hot Mix Asphalt, 85% of the surveyed agencies specified the standard tests for fine aggregates angularity (AASHTO T 304 or ASTM C1252) for Superpave asphalt mixes. However, Oregon DOT discontinued running this test, California DOT has its own procedure of CTM 205, Texas DOT substituted the fine aggregate angularity (FAA) test with running the Hamburg test, and five states do not require the measurement of FAA for Superpave asphalt mixes (Prowell et al. 2005). Quality Test Name AASHTO ASTM Test Procedure by Agency No Test Required Number of Respondents Na2SO4/MgSO4 soundness 31 4 9 4 LAA loss 41 6 5 0 Deleterious materials 19 6 18 5 Particle angularity 25 11 12 4 Flat and elongated ratio 14 22 14 1 Durability: freeze-thaw 13 6 10 11 Polishing/skid resistance 6 4 7 12 Porosity 6 1 4 17 Table 3-3. Number of agencies performing or not performing selected quality tests on virgin coarse aggregates.

34 Aggregate Quality Requirements for Pavements Description of the Test or Property Testing Procedure AASHTO ASTM Agency Specification Degradation - - Idaho IT-15, Alaska Test Method 313, Chapter 8 of Wisconsin DOT Construction and Materials Manual Specific gravity and maximum absorption T 85 Modified - - Micro-deval durability T 327 - - Sand equivalent T 176 - - Organic impurities, coal and lignite T 21 - - Lightweight chert T 19 - - Aggregate durability index T 210 - - Crushed fragments - D5821 Silicon dioxide, insoluble residue - C114 - Nordic abrasion value of coarse aggregates - - Alaska Test Method 312 Clay lumps, shale, soft particles, and friable particles - - Nebraska NDR T 504, Pennsylvania Testing Manual 620 Magnesium oxide/insoluble residue to determine limestone/dolomite and sandstone - - Division 300 and 400 of Illinois DOT Standard Specifications for Road and Bridge Construction Air degrade - - Oregon TM 208 Deleterious materials - - Oklahoma Department Test Methods (OHDL) 9, Pennsylvania Testing Manual 519 Insoluble residue - - Oklahoma Department Test Methods (OHDL) 25 Metallic iron - - Pennsylvania Testing Manual 518 Petrographic number - - General Provisions and Contract Specifications for Highway Construction Manual for Prince Edward Island Note: Hyphens indicate not applicable. Table 3-4. Additional quality tests performed on virgin coarse aggregates. Mineral Aggregate Characteristics Test Method Requirement Combined Bulk Oven Dry Specific Gravity Arizona Test Method 251 2.350–2.850 Combined Water Absorption Arizona Test Method 251 0–2.5% Abrasion AASHTO T 96 100 Rev., Max 9%500 Rev., Max 40% Sand Equivalent AASHTO T 176 (After thoroughly sieving the sample, no additional cleaning of the fines from the Plus No. 4 material is required.) Minimum 55% Fractured Coarse Aggregate Particles Arizona Test Method 212 Minimum 70% (Plus No. 4 material) Source: Arizona DOT 2008. Table 3-5. Arizona DOT quality assurance testing requirements for mineral aggregates used in asphaltic concrete.

Aggregate Sampling, Testing, and Quality Assurance 35 Furthermore, Kansas DOT requires agency specific test procedures for fine aggregates: KT-6 for specific gravity and absorption, KTMR-26 for mortar strength, and KTMR-28 for acid insoluble residue, if requested (Kansas DOT 2017). As noted in the previous section, survey respondents from 42 U.S. DOTs and seven Canadian provinces reported that they performed at least one test for checking virgin aggregate quality requirements. These agencies were asked to indicate if they performed the following virgin fine aggregate quality tests including Na2SO4/MgSO4 soundness, Atterberg limits, deleterious materials, uncompacted void content, micro-deval for degradation and polishing properties, and sand equiv- alent. The agencies were asked to report the testing procedures that they followed, and the results are presented in Table 3-6. Refer to Appendix C for a detailed compilation of the results given in Table 3-6. Table 3-6 indicates that 16 agencies do not run micro-deval tests on fine aggregates for measur- ing degradation and polishing properties and that 29 agencies follow the AASHTO method for performing sand equivalent test on fine aggregates. Only four agencies use the ASTM procedure. Moreover, similar to the case for virgin coarse aggregates, 13 agencies have their own procedures for running deleterious materials test on fine aggregates. Additional tests following other specific test procedures performed by agencies to determine quality of virgin fine aggregates are listed in Table 3-7. Figure 3-4. Ranges for AASHTO T 96 maximum abrasion loss by different agencies (Prowell et al. 2005). Quality Test Name AASHTO ASTM Test Procedure by Agency No Test Required Number of Respondents Na2SO4/MgSO4 soundness 24 5 7 9 Atterberg limits 25 4 6 7 Deleterious materials 22 6 13 7 Uncompacted void content 27 5 3 8 Micro-deval for degradation and polishing properties 10 4 7 16 Sand equivalent 29 4 10 3 Table 3-6. Number of agencies performing or not performing selected quality tests on virgin fine aggregates.

36 Aggregate Quality Requirements for Pavements 3.2.2 Quality Tests Conducted on RAP According to NCHRP Report 452, the binder content and physical properties of RAP aggregates, including particle shape, texture, and angularity, are required for use in mixture design. These factors also control RAP performance for stiffness, crack resistance, modulus, and deformation characteristics (McDaniel and Anderson 2001). NAPA recommends the following tests on RAP to ensure appropriate quality requirements for asphalt mixtures (West 2015): • Asphalt binder content of the RAP • Bulk specific gravity of the RAP aggregate • Consensus properties of the aggregate recovered from the RAP • RAP asphalt binder properties Common methods for determining the asphalt content of RAP include ignition method (AASHTO T 308 or ASTM D6307) and solvent extraction method (AASHTO T 164). Depend- ing on pavement layer and traffic conditions of the project where RAP is used, additional qual- ity tests might be required to ensure satisfactory performance. For example, if RAP is used in a surface mix exposed to high-speed traffic, the polishing or mineralogical composition of the RAP aggregate might need to be determined. Deniz et al. (2010) evaluated the expansion properties of RAP aggregates, steel slag RAP, and virgin aggregates used in Illinois for use as a base or subbase material. ASTM D2940, Standard Specification for Graded Aggregate Material for Bases or Subbases for Highways or Airports, limits expansion values to no greater than 0.50% at 7 days when tested in accordance with ASTM D4792, Standard Test Method for Potential Expansion of Aggregates from Hydration Reactions. Accord- ing to the findings, stone mastic asphalt RAP, steel slag RAP, surface binder RAP with 60% steel slag, and surface RAP with 92% steel slag can be safely used as base or subbase materials. However, porous and nonporous steel slag should not be used without proper curing that satisfies the limita- tion specified by ASTM D2940 (Deniz et al. 2010). Description of the Test or Property Testing Procedure AASHTO ASTM Agency Specification Dry strength - - Section 5.2 of Alberta Infrastructure and Transportation Standard Specifications for Highway Construction Degradation - - Idaho IT-15 Sand-sized acid insoluble residue content - D3042 modified - Organic impurities T 21, T 71 - - Specific gravity and absorption T 84, T 85 - - Clay lumps Nebraska NDR T504 Hydrometer for clay-sized material, color for organics of natural sands, and mortar strength if color is too high - - Division 300 and 400 of Illinois DOT Standard Specifications for Road and Bridge Construction Lightweight and clay content - - Saskatchewan STP 206-09, STP 206-15 Air degrade - - Oregon TM 208 Deleterious - - - - Oklahoma OHD L 9 California bearing ratio Virginia VTM-7 Note: Hyphens indicate not applicable. Table 3-7. Additional quality tests performed on virgin fine aggregates.

Aggregate Sampling, Testing, and Quality Assurance 37 Transportation agencies were asked about tests they perform for checking RAP quality require- ments for construction of pavement layers. In total, 28 U.S. DOTs and five Canadian provinces reported that they perform tests for checking RAP quality requirements. These agencies were asked to identify if they performed any of the RAP quality tests including residual asphalt binder content, micro-deval for polishing and degradation properties, deleterious materials, expansion properties, flat and elongated ratio, and freeze-thaw tests. The agencies were also asked to report the test procedures, for example, ASTM, AASHTO, and agency-developed test procedure that they followed. The results are summarized in Table 3-8. Refer to Appendix C for a detailed compilation of the results given in Table 3-8. Table 3-8 presents data to indicate that 13 agencies do not evaluate freeze-thaw and expan- sion properties for their RAP sources, 21 agencies measure residual asphalt content in RAP as a quality indicator following the AASHTO test procedure, while only three agencies use the ASTM method. Also, 10 agencies mentioned that they have their own procedures for identifying resid- ual asphalt content in RAP. Additional tests following other specific test procedures performed by agencies to determine quality of RAP aggregates are listed in Table 3-9. 3.2.3 Quality Tests Conducted on RCA The quality tests conducted on RCA are dependent on the application in pavement construction. For using RCA in concrete pavements, ACI 555R-01 lists quality checks for aggregate density, water absorption, LAA loss, sulfate soundness, and contaminants. For LAA, ACI 555R-01 lists ASTM C33 standard procedure for aggregates in concrete with abrasion loss values lower than 50% for general construction and lower than 40% for RCA used in crushed stone layers beneath pavements. These LAA loss values are the same standard values for virgin aggregates used in concrete. Regarding RCA absorption, ACI 555R-01 specifies that the absorption be limited to 10% (ACI 2001). For the allowable quantities of contaminants such as plaster, asphalt, wood, soil, gypsum, plastic, and rubber in RCA aggregates used in concrete pavement applications, ACI 555R-01 lists some limits based on research conducted on recycled aggregates (ACI 2001 and Hansen 1986). For both Note: Hyphens indicate not applicable. Quality Test Name AASHTO ASTM Test Procedure by Agency No Test Required Number of Respondents Residual asphalt binder content 21 3 10 13 Micro-deval for polishing and degradation properties 4 - 1 12 Deleterious materials 3 - 4 0 Expansion properties 1 - - 13 Flat and elongated ratio 5 1 1 10 Durability: freeze-thaw 2 - - 13 Table 3-8. Number of agencies performing or not performing selected quality tests on RAP. Note: Hyphens indicate not applicable. Description of the Test or Property Testing Procedure AASHTO ASTM Agency Specification Degradation - - Idaho IT-15 Consensus properties M 323 - - Specific gravity and absorption T 84, T 85 - Saskatchewan STP 204-09 Table 3-9. Additional quality tests performed on RAP.

38 Aggregate Quality Requirements for Pavements recycled coarse and fine aggregates, these limits are 10 kg/m3 for impurities such as plasters, clay lumps and other impurities of densities less than 1,950 kg/m3, and 2 kg/m3 for asphalt, plastics, paints, cloth, paper, wood, and similar material particles retained on 1.2 mm sieve, as well as other impurities of densities less than 1,200 kg/m3 (ACI 2001, Hansen 1986). Illinois DOT considers RAP as a contaminant for RCA and limits its quantity to 5% for Class C crushed concrete aggregates used for embankments, subbases, stabilized subbases, bases, gravel roads, and shoulders, and to 2% for Type B crushed concrete aggregates used for wedge shoulders (Illinois DOT 2017). FHWA’s Technical Advisory Circular T5040.37 restricts RCA to have not more than 1% asphalt and requires 90% of all the aggregates to be “cement paste and aggregate” (FHWA 2007). The Technical Advisory Circular also requires the RCA to be free from chlorides and reactive materials unless mitigation measures are taken to prevent recurrence of materials related to distresses in the new concrete. Chloride content is limited to 0.04 kg chloride ion/m3 (0.06 lb chloride ion/yd3) for the RCA to be used in new continuously reinforced concrete pavement or jointed reinforced concrete pavement to slow the corrosion of steel (FHWA 2007). The use of higher chloride con- tents, however, is permitted for jointed plain concrete pavements with epoxy-coated dowel bars. AASHTO MP 16-13 (AASHTO 2015) also divides RCA for concrete pavement applications into three classes based on the maximum allowable quantities of deleterious materials; these quantities are summarized in Table 3-10 (AASHTO 2015, Reza and Wilde 2017). In Table 3-10, class designations refer to the weathering exposure categories: Severe (A), Moderate (B), and Negligible (C). AASHTO MP 16-13 (2015) lists ranges and levels of different quality tests for RCA used in concrete applications. These ranges are a maximum LAA loss of 50%; a maximum fines content [passing sieve No. 200 (0.075 mm) of 1.5%]; a maximum chloride content of 0.6 lb/yd3; and a maximum sodium sulfate soundness loss of 12% if this test is applicable by local experience (AASHTO 2015, Reza and Wilde 2017). In total, 19 U.S. DOTs and British Columbia province indicated that they perform at least one test for checking RCA quality requirements for construction of pavement layers. Those 20 agencies were asked to identify which of the following RCA quality tests they performed, including LAA loss, absorption, deleterious materials, ASR, micro-deval for polishing and deg- radation properties, freeze-thaw, and specific gravity. The agencies were also asked to indicate which testing procedures they followed. Table 3-11 summarizes the results. Only Pennsylvania DOT reported that it performs similar tests on both RCA and virgin coarse aggregates, and five agencies reported they do not run micro-deval, ASR, deleterious, and absorption tests on their RCA sources. Additionally, 14 agencies perform LAA tests following the AASHTO test Source: AASHTO 2015, Reza and Wilde 2017. Note: sp gr SSD = specific gravity saturated surface dry. Table 3-10. Maximum allowable percentages of impurities for RCA in concrete pavements.

Aggregate Sampling, Testing, and Quality Assurance 39 procedure, while two agencies run LAA tests following the ASTM method and three other agen- cies follow their own specifications. Although the ASR test has been reported in the literature as an important quality indicator of RCA materials and as one of the predominant deterioration means of concrete (Thomas et al. 2013), only seven agencies measure this property for RCA sources using different test methods. Florida DOT reported that they do not perform sound- ness on RCA because of a possible chemical reaction with the paste. Refer to Appendix C for a detailed compilation of the results given in Table 3-11. 3.2.4 Quality Tests Conducted on Artificial/Byproduct Aggregates NCHRP Synthesis 435: Volume 5: Slag Byproducts lists some of the quality test ranges for common types of SFS and BFS based on the research findings by John Yzenas and the Recycled Materials Resource Center (RMRC 2008, Stroup-Gardiner and Wattenberg-Komas 2013a). These physical, quality, and engineering properties are summarized in Tables 3-12 and 3-13 (Stroup-Gardiner and Wattenberg-Komas 2013a). Table 3-11. Number of agencies performing or not performing selected quality tests on RCA. Quality Test Name AASHTO ASTM Test Procedure by Agency No Test Required Number of Respondents LAA loss 14 2 3 1 Absorption 6 2 1 5 Deleterious materials 4 3 5 5 ASR 3 3 1 5 Micro-deval for polishing and degradation properties 2 1 3 5 Durability: freeze-thaw 2 2 3 4 Specific gravity 6 3 3 4 Table 3-12. Engineering properties for iron and steel slags. Physical Property Blast Furnace Slag Steel Furnace Slag Test Method BOF EAFACBFS Open Graded Blast furnace slag is from the first furnace; steel furnace slag is from the second furnace.

40 Aggregate Quality Requirements for Pavements Property Iron Slag ACBFS Steel Slag (type not identified) Table 3-13. Typical quality and engineering properties for iron and steel slag materials. SFS is obtained during the steel-making process. SFS is primarily composed of lime, iron oxide, and silica. The mechanical properties of SFS such as EAF steel slag satisfy (and normally exceed) the aggregate requirements for granular aggregate subbase or base for permanent deformation, stiffness modulus, and fatigue (Pasetto and Baldo 2010). Additionally, Yi et al. (2012) indicated that using SFS as a coarse aggregate in HMA enhances the mechanical properties and using SFS in cold in-place recycling mixes improves Marshall stability, resilient modulus, and tensile strength, and reduces the susceptibility to moisture damage and permanent deformation. Using SFS in refractory concrete mixes was also found to have comparable mechanical properties to virgin aggregates when the SFS is heated to a high temperature before use (Yi et al. 2012). BFS can be considered similar to conventional aggregate for application in pavement construction. The high stability of BFS aggregates and their ability to lock up can be useful in construction over weak subgrade. Additionally, the compacted unit weight of BFS aggregates is relatively smaller than the natural aggregates, and the material characteristics produce the greater volume for the same weight (Chesner et al. 1998). In total, 15 U.S. DOTs reported that they perform at least one test for checking SFS or BSF quality requirements for construction of pavement layers. These agencies were asked which of the following quality tests they performed for SFS/BFS, including chemical composition, mineralogi- cal properties, specific gravity, micro-deval for polishing and degradation properties, expansion properties, and freeze-thaw. Additionally, they were asked to report which testing procedures they followed. The results are summarized in Tables 3-14 and 3-15. Refer to Appendix C for a detailed compilation of the results given in Tables 3-14 and 3-15. The survey findings showed that four agencies do not check expansion properties of SFS and BFS. Eight agencies perform specific gravity tests on SFS following the AASHTO procedure and two agencies follow the ASTM method. Additionally, four agencies reported that they have developed their own procedures for measuring specific gravity of SFS sources. Moreover, seven Quality Test Name AASHTO ASTM Test Procedure by Agency No Test Required Number of Respondents Chemical composition 2 2 2 2 Mineralogical properties - 2 - 3 Specific gravity 8 2 4 - Micro-deval for polishing and degradation properties 1 - 2 3 Expansion properties 2 1 3 4 Durability: freeze-thaw 2 1 2 3 Note: Hyphens indicate zero (0) number of respondents. Table 3-14. Number of agencies performing or not performing selected quality tests on SFS.

Aggregate Sampling, Testing, and Quality Assurance 41 agencies perform specific gravity tests on BFS following the AASHTO procedure and only one agency follows the ASTM method. Additionally, three agencies have developed their own pro- cedures for measuring specific gravity of BFS. Although mineralogical and chemical properties need to be considered as important quality indicators for both SFS and BFS aggregate sources (Lewis 1982), few agencies reported that they check this property. Also, it was found that not many agencies evaluate the freeze-thaw properties of SFS and BFS aggregates. Arkansas DOT indicated that the DOT performs LAA testing on SFS and BFS aggregate sources. Pennsylvania DOT checks the quality of SFS and BFS aggregate sources with similar tests per- formed on natural coarse aggregates. They have also developed the PMT 130 test procedure for evaluating expansion characteristics of SFS aggregates. 3.2.5 Quality Tests Conducted on Blended/Stabilized Aggregates NCHRP Report 598: Performance-Related Tests of Recycled Aggregates for Use in Unbound Pave- ment Layers evaluated the engineering performances of different mixes of RCA, RAP, and virgin aggregates (limestone, granite, and gravel). The blended materials evaluated included RCA with limestone from Illinois (RCA-LS-IL), RCA with gravel from Louisiana (RCA-GV-LA), RCA with granite from South Carolina (RCA-GR-SC), RAP with limestone from Mississippi (RAP-LS-MS), RAP with granite from Colorado (RAP-GR-CO), and RAP with gravel from Louisiana (RAP-GV-LA). Quality tests evaluated included frost susceptibility by tube suction test and index method, micro-deval toughness, and Canadian freeze-thaw for durability. Typical 50:50 blended mixes of recycled and virgin materials showed lower LAA loss than RCA and RAP individually. Additionally, 50:50 blends of RCA and virgin aggregates and RCA exhibited higher material loss than other blends due to higher percentages of fines in the as-received condition (Saeed 2008). Kazmee et al. (2012) evaluated the effectiveness of RCA, derived from crushing an old airport concrete pavement, when blended with virgin crushed dolomite aggregates for pavement base or subbase application. RCA specimens consistently gave the highest strength and modulus and the lowest permanent deformation accumulation. Crushed dolomite showed the highest degree of moisture sensitivity due to its 13% fines content when compared with the RCA with 8% fines. The research findings showed that the single source materials, either RCA or virgin aggregates alone, performed consistently better than the aggregate blends by achieving higher strength and modulus and lower permanent deformation (Kazmee et al. 2012). For asphalt concrete mixes, the Bailey method can be used to ensure conformance of blended aggregate mix to required quality. The Bailey method is a method for mix design that ensures the aggregate interlock and optimum packing (controlled by changing the fine aggregate coarse fraction ratio and the fine aggregate fine fraction ratio), to meet the required volumetric design Note: Hyphens indicate zero (0) number of respondents. Quality Test Name AASHTO ASTM Test Procedure by Agency No Test Required Number of Respondents Chemical composition 1 1 1 3 Mineralogical properties - 2 - 3 Specific gravity 7 1 3 1 Micro-deval for polishing and degradation properties 1 - 2 3 Expansion properties 2 2 1 4 Durability: freeze-thaw 2 1 2 2 Table 3-15. Number of agencies performing or not performing selected quality tests on BFS.

42 Aggregate Quality Requirements for Pavements criteria of Superpave, while the method also ensures good performance (Vavrik et al. 2002). For Portland cement concrete applications, NCHRP Research Results Digest 281: Aggregate Tests for Portland Cement Concrete Pavements: Review and Recommendations lists both direct and indirect tests to be used for aggregate blends in concrete. Indirect tests are those in which concrete specimens are tested. Tests are also divided into Level I tests (essential) and Level II tests (optional, based on the outcome of Level I tests). Level I tests include absorption, gradation, properties of fines passing No. 200 sieve (quantity and plasticity), shape, angularity and texture, thermal expansion of aggregates, aggregate abrasion, elastic modulus, polishing, aggregate strength and mineralogy, and ASR as well as freezing and thawing resistance (D-cracking). Level II tests include Methylene blue test for clays, XRD, XRF, and TGA for aggregate mineralogy as well as other alternative tests for ASR and D-cracking (Hanna 2003). Participating transportation agencies were asked to indicate whether they blend aggregate to improve quality (e.g., virgin + marginal, virgin + quarry byproduct, etc.). It was found that 36% (13 U.S. DOTs and three Canadian provinces) blend different aggregate types to improve the quality. Subsequently, those 16 agencies were requested to report if they had certain proce- dures to control the quality of the blended product. The descriptions of the different procedures reported by the agencies are listed in Table 3-16. 3.3 Quality Assurance Practices Different agencies have developed guidelines for statistical analyses of the results and ranges obtained from laboratory quality testing on aggregate sources. Quality assurance is required to address accuracy, precision, and capability. Accuracy means the average of all measurements falls relatively close to an understood point (or target). Precision indicates that all of the measure- ments over time are very close together. Capability is achieved if the process is both accurate and precise such that the process remains within specification or other predetermined limits with a high degree of confidence. As an example, Indiana DOT defines aggregate quality control as the Alabama All aggregates must be approved before blending. This will allow for confidence that all aggregates meet the minimums of the standard specifications. Illinois Mechanical blending through interlocked feeders at the production site. Kansas, Montana, and Virginia Blend must meet specification requirements for intended use. Maine Periodic agency tests on material sampled during construction. New York Both components of the blend must be tested individually as well as the final product. The producer must also address the blending process in their operations plan as well as adhere to individual requirements determined by the specific reason for blending material. Ohio The product must be blended by pugmill to a consistent ratio and tested according to that specific blend. Aggregates are blended and stockpiled prior to delivery to an asphalt plant. Oregon and South Dakota Blended sample must pass all quality tests as unblended sample would. Pennsylvania and Vermont Quality control plan is submitted and approved/disapproved by the district material’s engineer/manager. Wyoming Natural pit run filler from other sources may be required to bring volumetric properties into specification. There is a specification for the requirements of pit run filler. Agency Procedures Table 3-16. Procedures to control the quality of the blended aggregate product.

Aggregate Sampling, Testing, and Quality Assurance 43 prediction of product performance within pre-established limits for a desired portion of the out- put (Indiana DOT 2017b) and proposes two principles included in an aggregate quality control program: making sure that the correct target quality values are understood and achievable, and controlling variability within pre-established limits. Indiana DOT illustrates the process of managing and implementing an aggregate quality con- trol program into an agency, as shown in Figure 3-5. EMPLOYEE INPUT VISUAL OBSERVATIONS ESTABLISH STANDARDS MEASUREMENTS OVER TIME PROCESS CONTROL NEW CAPABLE PROCESS IDENTIFY KEY PROCESS VARIABLES CURRENT PROCESS ELIMINATE SPECIAL CAUSES OF VARIABILITY COMMUNICATE WITH CUSTOMERS MAKE MEASUREMENTS AND ASSESSMENTS REDUCE COMMON CAUSES OF VARIABILITY ESTABLISH PRODUCT TARGETS AND LIMITS NEW STABLE PROCESS DECISION MAKING Figure 3-5. Stages for managing and implementing an aggregate quality program in an agency.