Use of Reclaimed Asphalt Pavement and Recycled Asphalt Shingles in Asphalt Mixtures (2016)

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.

43 Specific asphalt availability factors were reported by eight agencies: • 0.85 for RAS asphalt (four agencies). • 0.75 for RAS asphalt (two agencies). • 0.70 for RAS asphalt (one agency). • 0.75 for RAP and RAS (one agency). Additional information provided by the respondents about asphalt availability factors included: • Use a percentage asphalt requirement, although there appears to be very little reactivation of recycled asphalt when using WMA, especially with RAS. • Specification requires adding 0.2% virgin asphalt when RAS is used. • Use effective binder content. • Credit a 75% contribution of the RAP binder. The remain- ing 25% is added to the mix design/determined asphalt content, which is referred to as the “corrected optimum asphalt content (COAC).” We changed from 100% to 75% in 2012 and adjusted again 2014. • Limit the percentage of RAP based on where it is in the pavement structure (i.e., lift-dependent) (three agencies): – 25% in lower lifts; 15% by weight in mixtures in the top 0.2 ft (2.5 in.). – 30% in base lift, 20% in binder lifts, 15% in wear course, but planning on changing to ABR. – Only allow RAS in maintenance mixtures and not in designed mixtures. • Several states indicated that they were interested in using a reduction factor, but had not yet applied one. Five agencies use the ABR equation to establish a mini- mum percentage of virgin asphalt in the total asphalt content (Table 24). One agency noted it set a minimum ABR value based on the lower PG temperature: 70% for PG xx-28 and 80% for PG xx-34. Another agency defines the minimum ABR based on the mixture type: 80% for surface mixtures and 65% for base and binder mixtures. Six agencies use the RAP binder ratio to limit the percent- age of recycled material asphalt that can be used to replace the virgin asphalt (Table 25). Two agencies limit recycled material asphalt from all of the recycled materials in the mix- ture to 23% of the total asphalt content. One agency uses The State Materials Engineer survey (Appendix A) focused on specific practices used when working with recycled materials in the laboratory, test methods, and any modifications needed when designing mixtures with recycled materials. Forty-five responses were received, including agencies that indicated they do not currently use at least 25% RAP or RAS in their mixtures. The main survey topics and the organization of this chapter are as follows: • Total asphalt content • Measuring recycled material asphalt content • Selecting the virgin asphalt grade • Material properties required for volumetric calculations • Sample preparation • Mixture testing. TOTAL ASPHALT CONTENT Accounting for Recycled Asphalt in Mixture The contribution of the recycled asphalt to the total asphalt content can be considered as fully contributing (100% use- ful), partially useful, or not useful (0%; “black rock”). More than 78% of the responding agencies consider that 100% of the RAP asphalt is useful and fewer than 8% that none of the RAP asphalt is useful. Approximately 16% attribute only a portion of the RAP asphalt to the total asphalt content (Table 22). Agencies view the contribution of the RAS asphalt dif- ferently than the RAP asphalt. Agencies are about evenly split between considering RAS asphalt as 100% useful and only partially useful (asphalt availability factor). Fewer than 15% of the respondents believe RAS does not provide any contribution. When mixtures contain a combination of both RAP and RAS, 14 agencies consider the combined RAP and RAS asphalt as 100% available and only three believe none of the recycled asphalt contributes to the total asphalt content. Twenty-seven respondents use the total asphalt content equa- tion to calculate the asphalt content of the mixture (Table 23). Two agencies noted that they determine the asphalt content of 10 RAP samples and then use the average RAP asphalt content in the equation. chapter three STATE MATERIALS ENGINEERS SURVEY

44 Survey Question: For the purposes of mix designs, indicate which “philosophy” is used to establish the contribution of the recycled material asphalt. Materials 100% Available for Mix 0% ("Black Rock") Agency-Assumed Percentage of the Total Recycled Asphalt Content Responses per Row % n % n % n n RAP 25% or Less RAP 83 29 3 1 14 5 35 More Than 25% RAP 77 23 7 2 17 5 30 RAS RAS, Manufacturer Waste 46 12 8 2 46 11 26 RAS, Tear-Offs 39 9 13 3 48 10 23 RAS, any Combination 42 12 7 2 48 9 27 Combination RAP and RAS Combination 42 14 9 3 48 8 33 Not all survey respondents answered all questions. TABLE 22 GENERAL APPROACH FOR CONSIDERING CONTRIBUTION OF RECYCLED MATERIAL ASPHALT CONTENT TO TOTAL ASPHALT CONTENT Survey Question: We use the sum of the new asphalt and recycled asphalt material content: Total asphalt content = (RAP asphalt content) (% of RAP in mix) + (RAS asphalt content) (% of RAS in mix) + (new asphalt content %) Use TAC Equation? % n Yes 84 27 No 16 5 Not all survey respondents answered all questions. n = 32. TABLE 23 USE OF TOTAL ASPHALT CONTENT (TAC) EQUATION Survey Question: We use the asphalt binder ratio (ABR) equation: ABR = (virgin asphalt %)/(Total asphalt content %) Use ABR Equation? % n Yes 21 5 No 78 18 Not all survey respondents answered all questions. n = 23. TABLE 24 USE OF ASPHALT BINDER RATIO EQUATION Survey Question: We use the recycled binder ratio (RBR) RBR = (recycled binder content %)/(Total asphalt content %) Use RBR Equation? % n Yes 26 6 No 74 17 Not all survey respondents answered all questions. n = 23. TABLE 25 USE OF RECYCLED BINDER RATIO EQUATION an ABR limit of 25%. Three agencies define the maximum limit based on where the mixture is located in the pavement structure: • 20% for surface mixture and 35% for base and binder mixtures (one agency). • 40% for surface mixtures and 45% for intermediate and base mixtures (one agency). • 45%, but it varies (no specifics given) (one agency). Total Asphalt Content Section Summary • Agencies use the total asphalt content equation, but vary substantially in the values used for asphalt avail- ability factors. – Eighteen agencies use a value of 1.0 (i.e., 100% availability for both RAP and RAS). NN Eleven agencies use asphalt availability factors < 1.0, but the values used vary. Two of these agen- cies apply an asphalt availability factor to RAP as well as RAS. – Eleven agencies set limits on the percentage of the recycled material asphalt that can contribute to the total asphalt content (i.e., use either ABR or RBR). • One agency noted that RAS does not appear to con- tribute to the total asphalt content when combined with WMA used to lower temperatures. MEASURING RECYCLEd MATERIAL ASPHALT CONTENT How an agency prepares the RAP material before testing can impact the measured asphalt content. One question was included in the survey to determine which RAP fractions

45 are tested and if RAP and RAS are combined prior to deter- mining the asphalt content. A second question was included to determine which sieve size(s) was (were) used to define coarse and fine fractions when the laboratory separates the RAP sample before testing. Sizing Recycled Materials for Testing Twenty-nine agencies routinely determine the RAP asphalt content for unfractionated RAP. Twenty-one of these agen- cies also determine the asphalt content for both coarse and fine fractions (Table 26). Thirteen agencies combine RAP and RAS before testing when the combination of recycled materials is used in the mixture. Although 21 agencies measure the asphalt content of coarse and fine RAP fractions, only 11 fractionate the RAP in their laboratory (Table 27). These 11 agencies use the 4.75-mm (No. 4) sieve size to fractionate coarse and fine RAP. The differences between the number of agencies testing the indi- vidual coarse and fine fractions and those agencies fraction- ating the RAP in their laboratories before testing reflect the variations in the RAP samples submitted for mix designs by the contractors. That is, some contractors separated their RAP supplies into coarse and fine fractions, whereas others maintain RAP gradations with the full range of sieve sizes. Test Methods to determine Asphalt Content There are two test methods that can be used to determine the asphalt content of mixtures as well as asphalt-containing recycled materials. These are the traditional solvent extrac- tion methods (AASHTO T164) and the newer method using an ignition oven (AASHTO T308). Twenty agencies use the ignition oven method for determin- ing the asphalt content of the recycled materials (Table 28). The same 20 agencies also indicated that they conduct solvent extractions; therefore, these agencies have the ability to deter- mine correction factors based on the differences in the mass lost during solvent extraction and the mass lost during ignition oven testing. This method of establishing an ignition oven correction factor is applicable to both aggregates prone to mass loss result- ing from burning as well as to asphalt-containing recycled materials such as shingles with material content that also burns off (e.g., paper backing and roofing felt). Because eight agen- cies noted they no longer use any solvent extraction method in their laboratories, these agencies are not able to determine an ignition oven correction factor using this approach. Comments about determining the ignition oven correction factor for mixtures include: • Not used (four agencies). • Differences in results from solvent extraction (AASHTO T164) and ignition oven (AASHTO T308) are used to Survey Question: The asphalt content of the recycled materials is determined for each material. (Check all that apply.) Value % n RAP (not fractionated) 88 29 Coarse RAP fraction 64 21 Fine RAP fraction 64 21 RAS 58 19 RAP and RAS combined prior to testing 39 13 Not all survey respondents answered all questions. n = 33. TABLE 26 TESTING RECYCLED MATERIALS TO DETERMINE ASPHALT CONTENT Survey Question: If you separate RAP into coarse and fine fractions for testing, please indicate which sieve size is used for “retained on”/“percent passing.” Value % n 4.75-mm (No. 4) 100 11 2.36-m (No. 8) 0 0 Not all survey respondents answered all questions. n = 11. TABLE 27 PREPARATION OF RECYCLED MATERIALS PRIOR TO TESTING Survey Question: Do you use the ignition oven to determine the recycled material asphalt content? (Check all that apply.) Materials Yes No Sometimes, Depending on Aggregate Type Responses per Row % n % n % n N RAP, unfractionated 69 20 28 8 3 1 29 Coarse RAP fraction 61 14 30 7 9 2 23 Fine RAP fraction 61 14 30 7 9 2 23 RAS 50 9 44 8 6 1 18 RAP and RAS combination 50 8 50 8 0 0 16 Not all survey respondents answered all questions. TABLE 28 IGNITION OVEN USAGE

46 determine a correction factor (four agencies for RAP, two for RAS). • Varies or depends (no specifics) (two agencies). • Use 0.5% (two agencies). • Based on aggregate type. • As described in the ignition oven test method (AASHTO T308). • Applied only to the total mixture. • Require the contractor to determine the ignition oven correction factor for RAS. Solvent Extraction Methods Twenty-two agencies use one or more solvent extraction meth- ods, with the centrifuge methods being used most frequently (Table 29). Most agencies use the centrifuge extraction method with TCE solvent, although some agencies also use n-propyl bromide (Table 30). Similar responses were given, regardless of the material being tested. Eight agencies com- mented that they do not use any solvent extraction in their laboratories Measuring Recycled Material Asphalt Content— Section Summary • Most agencies use the ignition oven method for deter- mining asphalt content. • About half of these agencies also use a solvent extrac- tion method for determining asphalt content and ignition oven correction factors (when needed). • Nearly 20% of the states noted they are not using any solvent extraction method in their laboratories; there- fore, they do not have the ability to determine correction factors for nonasphalt but burnable materials contained in the recycled materials using this method. RECYCLEd MATERIAL ASPHALT PROPERTIES Twenty-two agencies also recover asphalt from the solvent after extraction (Table 31), but use a range of methods. Abson (eight agencies) and Rotavapor (eight agencies) are the most commonly used methods. Four agencies use the AASHTO T319 combination extraction and recovery method and two agencies state-specific methods. Survey Question: If you use solvent extraction to determine the recycled binder content, indicate which method(s) is (are) used. (Check all that apply.) Materials Centrifuge Reflux Vacuum Extraction Vessel, AASHTO T319 Soaking (nonstandard option) Responses per Row % n % n % n % n % n N RAP, unfractionated 77 17 14 3 9 2 0 0 0 0 22 Coarse RAP fraction 73 11 13 2 13 2 0 0 0 0 15 Fine RAP fraction 73 11 13 2 13 2 0 0 0 0 15 RAS 100 14 14 2 7 1 0 0 0 0 14 RAP and RAS combination 100 10 20 2 20 2 0 0 0 0 10 Not all survey respondents answered all questions. TABLE 29 SOLVENT EXTRACTION TEST METHODS Survey Question: … And indicate which solvent(s) is (are) used. (Check all that apply.) Materials Trichloroethylene (TCE) n-Propyl Bromide (nPB) Toluene Toluene and Ethanol Blend Responses per Row % n % n % n % n n RAP 61 11 28 5 6 1 6 1 18 Coarse RAP fraction 79 11 21 3 0 0 0 0 14 Fine RAP fraction 79 11 21 3 0 0 0 0 14 RAS 62 8 38 5 0 0 0 0 13 RAP and RAS combination 75 9 25 3 0 0 0 0 12 Not all survey respondents answered all questions. TABLE 30 SOLVENTS USED WITH EXTRACTION METHODS

47 Not only do agencies use a range of methods for asphalt recovery, they also evaluate a range of different recycled mate- rial asphalt properties (Table 32). The high temperature DSR shear modulus, G*, is determined as-is after recovery by eight agencies, and four agencies determine the high temperature G* after rolling thin film oven (RTFO) conditioning of the recov- ered asphalt. Another two agencies determine the high tem- perature G* after both RTFO and pressure aging vessel (PAV) conditioning. Additional respondent comments included: • Recovered asphalts are considered as already aged; we only use PAV aging to determine the low tempera- ture grade. • BBR is used to determine the stiffness and m-value for low temperature assessment; no RTFO aging is used before testing. • Only dynamic shear rheometer testing is used to deter- mine failure temperature for recovered asphalt. • Use the Bonaquist method (see chapter two: Literature Review). Recycled Material Asphalt Property— Section Summary • The majority of agencies recovers extracted recycled material asphalt, but uses a range of recovery methods. • No consistent approach is used when determining the high temperature DSR shear modulus, G*, and low temperature BBR stiffness and m-value. – Some agencies determine this value for the as- recovered, and other agencies determine the high temperature G* after RTFO conditioning, or after RTFO and PAV conditioning. • One agency uses the Bonaquist method to evaluate if the recycled material asphalt fully contributes to the total asphalt content properties. RECYCLEd MATERIAL AGGREGATE PROPERTIES Twenty agencies determine recycled material aggregate prop- erties (Table 33). Six other agencies commented that they do not evaluate the individual recycled material aggregate properties because: • Source and quality of RAP is known (two agencies). • All RAP that is required comes from state roads. • RAS aggregate is considered better quality than most of roadway fine aggregate. • RAP and RAS aggregate properties are too variable to classify other than gradation. • Only the mix design blend is tested. Aggregate Test Methods Consensus properties are those that experts consider important to the final pavement properties and include gradation; pres- ence of clay-sized particles; that is, sand equivalent; and coarse and fine aggregate shape (Table 34). Twenty-four agencies Survey Question: Which recovery methods(s) do you use? (Check all that apply.) Value % n Abson (AASHTO T170) 44 8 Rotavapor (ASTM D5404) 44 8 Combination Extraction/Recovery (AASHTO T319) 22 4 Agency-Specific Method 11 2 Not all survey respondents answered all questions. n = 18; more agencies have recovery methods than indicated they recover recycled materials. TABLE 31 METHODS USED FOR ASPHALT RECOVERY Survey Question: Indicate which binder tests you use to determine the true (continuous) recycled binder grade. (Check all that apply.) Testing % n DSR, G*, as-recovered high asphalt temperature 53 8 After Rolling Thin Film Oven (RTFO) Conditioning DSR, G*, high asphalt temperature 33 5 DSR, G*, intermediate asphalt temperature 40 6 BBR, Stiffness, low asphalt temperature 33 5 BBR, m-value, low asphalt temperature 27 4 Testing After RTFO and Pressure Aging Vessel (PAV) Conditioning DSR, G*, intermediate binder temperature 13 2 Stiffness, low binder temperature for stiffness 20 3 m-value, low binder temperature 7 1 Not all survey respondents answered all questions. n = 15. TABLE 32 RECYCLED MATERIAL ASPHALT PROPERTIES EVALUATED AFTER EXTRACTION AND RECOVERY

48 determine RAP gradations (15 agencies after ignition oven and nine after solvent extraction). Slightly fewer agencies deter- mine the percentage passing the 0.075-mm (No. 200). Although the ignition oven method may alter aggregate properties to some degree, nine agencies determine various aggregate shape parameters after ignition oven testing. At most, five agencies evaluate RAP particle shape after solvent extraction, and five determine the sand equivalent for RAP after ignition oven testing. RAS aggregate gradations are the only aggregate property evaluated. Eleven agencies determine the RAS aggregate gradations (six after ignition oven and five after solvent). Respondents provided the following additional comments about recycled material aggregate properties: • Consensus properties are determined for the blended aggregates. • Only the sand equivalent on the composite of virgin aggregates is determined. • Only test RAP aggregate when the source is unknown. • Use AASHTO T335 instead of ASTM D5821 for deter- mining the percentage of fractured faces. • Check properties day to day; reserve the right to check RAP individually if needed. • RAS aggregates considered to have better properties than other aggregates. Eighteen agencies assume that the RAP source properties are acceptable because they came from a state project (Table 35). Only two agencies determine the LA abrasion values for RAP aggregate (after ignition oven) and another two agencies use the micro-Deval (after solvent extraction). One agency evalu- ates RAP sodium sulfate soundness and another evaluates soundness of RAP aggregate using magnesium sulfate. Ten agencies assume that the source properties of the RAS aggregate are acceptable rather than evaluate with testing. Survey Question: Indicate when the aggregate properties of the individual recycled materials need to be determined. Materials % n 15% or less RAP 50 10 25% or less RAP 50 10 More than 25% RAP 40 8 RAS, manufactured waste 20 4 RAS, tear-offs 15 3 RAS, combination 15 3 RAP and RAS combination 30 6 We test aggregates for the mixture (after solvent extraction or ignition oven) rather than individual recycled aggregate properties 30 6 Not all survey respondents answered all questions. n = 20. TABLE 33 WHEN ARE AGGREGATE PROPERTIES DETERMINED FOR RECYCLED MATERIALS Material Particle Size Particle Shape R es po ns es p er R ow Gradation (sieve analysis) Minus 0.075-mm (No. 200) by washing Sand* equivalent (AASHTO T176) Flat and elongated (ASTM D4791) Fractured faces (ASTM D5821) Fine aggregate angularity (AASHTO T304) % n % n % n % n % n % n n After Ignition Oven Testing RAP 100 15 80 12 33 5 53 8 60 9 60 9 15 RAS 100 6 50 3 0 0 0 0 0 0 0 0 6 After Solvent Extraction RAP 100 9 67 6 0 0 33 3 55 5 44 4 9 RAS 100 5 40 2 0 0 0 0 0 0 0 0 5 *Evaluates clay-sized particles. Not all survey respondents answered all questions. Total of 27 individual agencies responded to this question. TABLE 34 CONSENSUS AGGREGATE PROPERTIES DETERMINED FOR RECYCLED MATERIAL AGGREGATE

49 Additional respondent comments noted that one agency conducts source tests when investigating polish resistance quality of the aggregate. Another agency noted that the con- tractor has the option to conduct these tests. Recycled Material Aggregate Property— Section Summary • Gradations of the RAP and RAS materials are the most frequently evaluated recycled material aggregate prop- erty after either ignition oven or solvent extraction. • RAP source aggregate properties are typically consid- ered acceptable, usually because RAP was obtained from state projects. • RAS source aggregate properties are also considered acceptable (specific reasons not explored with survey questions). • Only a limited number of agencies measure the recy- cled material aggregate shape, clay-sized particle con- tent, soundness, and toughness. SELECTING THE VIRGIN ASPHALT GRAdE When multiple asphalt grades are routinely used by the agency, it is likely contractors in the state will have multiple asphalt tanks at the plant so that a range of different mixtures can be produced. However, some agencies specify only two asphalt grades that can be used in the state, which implies that the contractors in the state will likely have, at most, two asphalt storage tanks. The percentage and/or type of recycled material that can be used in a given state is limited to the availability of the virgin asphalt needed to produce the desired combined asphalt properties. When a range of asphalt grades is available, the percentage and/or type of the recycled materials can be selected and the asphalt grade selected to meet asphalt specifications. A limited availability of different asphalt grades usually means that the percentage and/or type of recycled material is restricted to what can be used and still meet the asphalt specifications, usually with- out changing the virgin asphalt grade. One question was included in the survey to determine which approach was used by each agency. Ten agencies set the percentage of recycled material to be used and then select the virgin asphalt grade required to meet specification requirements (Table 36). Four agencies specify the virgin asphalt grade then select the percentage of recycled materials. Only one agency noted they use a softener or rejuvenating additive. Additional respondent comments included: • Use an established binder selection table. • Only specify virgin binder grade, then perform AASHTO M320 (Standard Specification for Performance-Graded Asphalt Binder) to verify. • Only PG 64-22 and PG 76-22 binders permitted in our state. • Currently, we do not adjust the grade for mixtures with less than 30% RAP, but will evaluate both blending charts and defined grade bumping in the future. Survey Question: Indicate the aggregate source property tests that are conducted on the recycled material aggregates. (Check all that apply.) Materials LA Abrasion (toughness) Micro- Deval (toughness) Sodium Sulfate Soundness Magnesium Sulfate Soundness Assume Source Properties Are Acceptable R es po ns es pe r R ow % n % n % n % n % n n After Ignition Oven Testing RAP 10 2 0 0 5 1 5 1 95 18 19 RAS 0 0 0 0 0 0 0 0 100 6 6 After Solvent Extraction RAP 0 0 13 2 13 2 8 1 93 14 15 RAS 0 0 0 0 0 0 0 0 100 4 4 Not all survey respondents answered all questions. *Assumption for acceptable RAP provided on the survey was because material came from state project. Total of 34 individual agencies responded to this question. TABLE 35 SOURCE AGGREGATE PROPERTIES FOR RECYCLED MATERIAL AGGREGATES Survey Question: Indicate which approach is used to ensure the blended binder meets the required PG grade. Value % n Establish (select) percent of RAP to be used, then determine the virgin asphalt PG grade needed 67 10 Choose virgin asphalt PG to be used, then determine the percent of recycled material 27 4 Use softening or rejuvenator additive to soften the recycled material binder, then proceed with determining the virgin asphalt PG 7 1 Not all survey respondents answered all questions. n = 15. TABLE 36 VIRGIN ASPHALT GRADE SELECTION APPROACH

50 • Contractor may choose any of these approaches. Most of the time, the contractor selects the RAP percentage and then determines the virgin binder grade they have to use. • No softening or rejuvenating additive is allowed. Eleven agencies “bump” the upper and/or lower virgin asphalt temperature even when not using high RAP (Table 37). Ten agencies “bump” the virgin asphalt grades for high RAP mixtures. Two agencies “bump” the virgin asphalt grade when using a combination of RAP and RAS. At most, four agen- cies extract, recover, and test the recycled material asphalt to determine the true asphalt grade. None of the agencies “bump” the virgin asphalt grade when using RAS. Other respondent comments about selecting the virgin asphalt grade included: • Difficult to determine PG grade because of the high softening point of RAS binder. • If contractor recovers and grades binder, they can use any percent recycled material. Selecting Virgin Asphalt—Section Summary • Various agency approaches are used to select the per- centage of recycled materials used in mixtures: – Most agencies select the percentage of recycled materials and then adjust the virgin PG grade required to meet the binder specification requirements. – Several agencies set the virgin binder PG grade and then select the maximum percentage of recy- cled material that can be used and still meet the specifications. – There appears to be a limited use of softening or reju- venating additives to modify the recycled material asphalt properties. MATERIAL PROPERTIES REQUIREd FOR VOLUMETRIC CALCULATIONS Recycled Material Asphalt Specific Gravities Asphalt binder-specific gravity values are used in various mix design volumetric calculations and the testing of the virgin asphalt-specific gravity is straightforward. However, determining the specific gravity of the recycled material asphalt content requires recovery of the asphalt after solvent extraction and any solvent not completely removed during recovery influences the specific gravity measurements. Other factors that make asphalt recovery and testing more vari- able are the additives used in the original manufacture of the recycled materials such as polymer modifiers, crumb rubber (i.e., contained in some RAP), fibers, fillers, and proprietary additives. Ten agencies use an assumed value for the RAP asphalt- specific gravity and six assume the RAS asphalt-specific grav- ity (Table 38). Assumed values used by the various agencies include: • Same as virgin asphalt (six agencies) • 1.03 (three agencies) • 1.035 • 1.01 • Use supplier data for virgin asphalt binder-specific gravity. Recycled Material Aggregate Specific Gravity The recycled material aggregate specific gravities (RAP, RAS, or a combination of RAP and RAS) are most often determined by measuring the theoretical maximum specific gravity of the recycled material, calculating the effective specific gravity, and finally calculating the recycled material aggregate bulk specific gravity (Table 39). Survey Question: Please indicate how your agency determines the virgin PG grade used in mixtures with recycled materials. (Check all that apply.) Materials Upper PG Temp. Lower PG Temp. Determine True PG with Testing Responses per Row 1 Grade Lower 1 Grade Lower 2 Grades Lower % n % n % n % n n < 25% RAP 73 8 82 9 0 0 9 1 11 >25% RAP 50 5 60 6 10 1 30 3 10 >50% RAP 0 0 0 0 0 0 100 3 3 <5% RAS 0 0 0 0 0 0 0 0 0 >5% RAS 0 0 100 1 0 0 0 0 1 RAP and RAS 33 1 67 2 0 0 33 1 3 Not all survey respondents answered all questions. TABLE 37 METHODS FOR SELECTING THE VIRGIN ASPHALT PG GRADE

51 Some agencies directly measure specific gravities after ignition oven and only a few agencies measure this prop- erty after solvent extraction. The solvent extraction or the ignition oven are known to influence aggregate property test results, but are the only methods currently available for removing asphalt if an agency wants to obtain measure- ments, rather than estimates, of the recycled material spe- cific gravities. Additional respondent comments included: • Extraction method used only during design phase for mixtures with more than 25% RAP; ignition oven method used the rest of the time. • Assume same as virgin aggregate. • Currently introducing bulk specific gravity testing of the RAP aggregate as a design requirement instead of the effective specific gravity (historically used by our agency). • Effective specific gravity is calculated based on the total asphalt content and maximum specific gravity after 2% to 3% virgin asphalt is added. • Asphalt absorption is estimated at 0.3% (for calculation of aggregate specific gravity from theoretical maximum specific gravity). • Do not test; materials too variable. Recycled Material Specific Gravities— Section Summary • Recycled material asphalt-specific gravities are fre- quently assumed. – Assumed values range from 1.01 to 1.035. – Several agencies use the specific gravity of the virgin asphalt for the specific gravity of the recycled material asphalt. • Recycled material aggregate specific gravities are fre- quently calculated from measurements of the recycled material, theoretical maximum specific gravity. Survey Question: Mix design calculations require a number of individual material properties. If your agency assumes, rather than measures, any of the properties in the table below, please enter the typical estimated values in the appropriate text boxes. Materials Recycled Material Asphalt Specific Gravity Virgin Asphalt Specific Gravity Responses per Row % n % n n RAP, unfractionated 90 9 50 5 10 Coarse RAP fraction 87 7 50 4 8 Fine RAP fraction 87 7 50 4 8 RAS 85 6 43 3 7 RAP and RAS combination 80 4 40 2 5 Not all survey respondents answered all questions. n = 10. TABLE 38 NUMBER OF AGENCIES THAT USE ASSUMED VALUES FOR CALCULATIONS Survey Question: Indicate which test methods are used to determine the specific gravities of the recycled materials aggregate. Materials Bulk Specific Gravity Estimate Bulk Specific Gravity Based on Experience Theoretical Maximum Specific Gravity (AASHTO T209) Responses per Row After solvent extraction After ignition oven % n % n % n % n n RAP, unfractionated 24 4 47 8 0 0 65 11 17 Coarse RAP fraction 7 1 36 5 0 0 79 11 14 Fine RAP fraction 7 1 36 5 0 0 79 11 14 RAS 0 0 18 2 9 1 82 9 11 RAP and RAS combination 13 1 38 3 13 1 75 6 8 Not all survey respondents answered all questions. n = 17. TABLE 39 RECYCLED MATERIAL AGGREGATE SPECIFIC GRAVITY

52 SAMPLE PREPARATION The steps for the preparation of materials for fabricating mix design samples are: • Sizing of aggregates and recycled materials, • Drying recycled materials prior to batching samples, • Heating (time and temperature) of materials for mixing, • Order of addition to mixing bowl, • Short-term aging before compaction, and • Compaction of samples. Sizing (Fractionating) Particles Most of the responding agencies use the same procedures for preparing recycled materials for batching their mix design samples, regardless of the type or percentage of recycled materials used in the mixture (Table 40). A range of practices is used to prepare coarse virgin or recy- cled materials for batching (Table 41). Some agencies separate both the virgin and recycled coarse material into individual sizes, whereas other agencies sieve the virgin aggregated into individual sizes, but only use the percent retained on a given sieve size (i.e., 4.75 mm or 2.36 mm). Additional respondent comments showed that: • Five agencies batch materials based on the stockpiles as prepared by the contractor. • Three agencies noted that definitions of “coarse” and “fine” for the recycled materials are established by the contractor. • Four agencies responded that the contractor determines how to fractionate RAP. More than 25% RAP. • One agency does not fractionate for batching. • One agency does fractionate for batching, even if the contractor does not. Survey Question: Does the percent or type of recycled materials used in the mixture change how you fractionate, or don’t fractionate, the materials in the laboratory? Materials Yes No Sometimes Responses per Row % n % n % n n 25% or less RAP 12 3 85 22 4 1 26 More than 25% RAP 15 3 60 12 25 5 20 Shingles, manufacturer waste 0 0 100 15 0 0 15 Shingles, tear-offs 0 0 100 15 0 0 15 Shingles, combination 0 0 100 13 0 0 13 RAP and RAS 13 2 73 11 13% 2 15 Not all survey respondents answered all questions. n = 30. TABLE 40 FRACTIONATING RECYCLED MATERIALS FOR BATCHING Survey Question: Indicate which individual sizes and/or percent retained on a given sieve size are used for batching coarse particles when using various percentages and types of materials in the mixtures. Materials 25 mm (1 in.) 12.5 mm (1/2 in.) 9.5 mm (3/8 in.) 4.75 mm (No. 4) 2.36 mm (No. 8) Retained on 4.75 mm (No. 4) Retained on 2.36 mm (No. 8) Re sp on se s pe r R ow % n % n % n % n % n % n % n n Aggregates 44 7 63 10 63 10 69 11 50 8 56 9 25 4 16 25% or less RAP 15 2 39 5 46 6 39 5 23 3 62 8 15 2 13 25% or more RAP 29 2 43 3 57 4 29 2 14 1 57% 4 0 0 7 Shingles, manufacturer waste 0 0 0 0 67 2 67 2 33 1 0 0 0 0 3 Shingles, tear-offs 0 0 0 0 67 2 67 2 33 1 0 0 0 0 3 Shingles, combination 0 0 0 0 67 2 67 2 33 1 0 0 0 0 3 RAP and RAS 33 1 33 1 67 2 33 1 0 0 33 1 0 0 3 Not all survey respondents answered all questions. n = 21. TABLE 41 FRACTIONATING COARSE RECYCLED MATERIALS

53 • One agency noted they use 19 mm (¾ in.) as a maxi- mum size because we ask that RAP be screened on a 25-mm (1-in.) grizzly before introduction to the plant. A wide range of practices for preparing fine virgin and recycled materials for batching was also reported. Some agencies fractionate fine virgin and RAP materials into the full range of individual fine aggregate sizes (1.18 mm, 0.6 mm, 0.3 mm, 0.15 mm, and 0.075 mm), whereas other agencies just use the materials passing either the 4.75-mm or 2.36-mm sieve size (Table 42). Only one agency separates RAS into individual sieve sizes. Three agencies commented that they batch materials as pre- pared by the contractor and the contractor defines which sieve size is used to define “coarse” and “fine.” One agency noted that fines are defined as passing the 4.75-mm (No. 4) sieve for production, but are defined as passing the 2.36-mm (No. 8) sieve for mix design purposes. The smaller sieve size is used for mix designs to provide better control of the mix design gra- dations. One agency indicated the RAS is batched as a single stockpile with at least 90% passing the 4.75-mm (No. 4) sieve. drying Recycled Materials Nineteen agencies dry recycled materials before using them to prepare mix design samples (Table 43). About two-thirds of these agencies use oven drying, whereas the remaining one- third dry materials under a fan at room temperature (Table 44). Fan drying of either RAP or RAS samples is typically com- pleted overnight (between 16 and 24 hours). Oven tempera- tures used for drying to a constant mass include 100°F, 125°F, and 230°F (38°C, 52°C, and 110°C). No consistent oven tem- perature or time in the oven was evident. Parameters used to identify when materials were dried to a “constant mass” were inconsistent. Several agencies use a maximum change in mass of 0.1% between subsequent weighings, but vary by use of drying method (oven, fan), temperatures, and times between subsequent weighings (e.g., 15 minutes, 30 minutes, 1 hour, and none specified). Addi- tional variations in parameters include using changes in mass of 0.05% or 0.5% over 30 minutes in an oven (300°F or 230°F), respectively. One agency uses a maximum change of 0.5 gram over 30 minutes in an oven at 300°F (149°C). Several agencies use oven drying of either RAP or RAS for a specified time rather than drying to a constant mass. One Survey Question: Indicate what individual sizes and/or percent passing a given sieve are used for batching fine particles when using various percentages and types of materials. Materials Passing 4.75 mm (No. 4) Passing 2.36 mm (No. 8) 1.18 mm (No. 16) 0.6 mm (No. 30) 0.30 mm (No. 50) 0.15 mm (No. 100) 0.075 mm (No. 200) R es po ns es pe r R ow % n % n % n % n % n % n % n n Aggregates 80 12 73 11 20 3 20 3 20 3 20 3 27 4 15 25% or less RAP 80 8 40 4 20 2 20 2 20 2 20 2 20 2 10 More than 25% RAP 67 4 50 3 17 1 17 1 17 1 17 1 17 1 6 Shingles, manufacturer waste 50 2 50 2 25 1 25 1 25 1 25 1 25 1 4 Shingles, tear-offs 50 2 50 2 25 1 25 1 25 1 25 1 25 1 4 Shingles, combination 50 2 50 2 25 1 25 1 25 1 25 1 25 1 4 RAP and RAS 67 2 33 1 33 1 33 1 33 1 33 1 33 1 3 Not all survey respondents answered all questions. n = 21. TABLE 42 FRACTIONATING FINE RECYCLED MATERIALS Survey Question: Is the recycled material dried prior to using to prepare mix design samples? Answer RAP RAS % n % n Yes 91 29 83 19 No 9 3 17 4 Not all survey respondents answered all questions. n = 32. TABLE 43 RECYCLED MATERIAL DRYING PRACTICES Survey Question: What method of drying is used? Method RAP RAS % n % n Under fan at room temperature 34 10 24 7 Oven 66 19 45 13 Not all survey respondents answered all questions. n = 29. TABLE 44 METHODS OF DRYING RECYCLED MATERIALS

54 agency reported using from 125°F to 175°F (52°C to 80°C) for a maximum of 2 hours. Another agency reported using 200°F (93°C) for a maximum of 2 hours. Agencies that only use RAP reported a range of times and temperatures for drying recycled material prior to batching: • 140°F (60°C) overnight. • 230°F (110°C) for a maximum of 1 hour. • 275°F (135°C) for a maximum of 16 hours. • 280°F or 300°F (138°C or 149°C) depending on the required compaction temperature to a constant mass. • Dry RAS using an oven temperature of 125°F for a maxi- mum of 16 hours. Heating Materials for Mixing The length of time and the temperature used to heat the com- ponents can influence the uniformity of the virgin asphalt coating, blending of the recycled asphalt with the virgin asphalt, and the separation of the agglomerated recycled materials. Additional aging of the recycled asphalt may also occur if too high of a temperature or too long of a preheat- ing time at elevated temperatures is used. A series of survey questions explored the heating times and temperatures, the order of the addition of materials into the mixing bowl, and approximate times used by agencies. Eleven agencies heat aggregates, RAP, and RAS separately, and 12 combine these materials and heat together (Table 45). Six agencies using RAS in their mixtures combine the RAS with sand before heating to avoid clumping of the RAS dur- ing heating. Seven agencies combine RAP and RAS materials before heating, when both are used in the same mixture. Additional respondent comments included: • All components are combined before heating for mix- ing and compaction. • Recycled material is kept separate and added 1 hour before mixing in the mix design process in a 350°F oven. • Verify mixtures on plant-produced samples; do not rep- licate mix designs using lab batched samples. • Because the plant heats RAP separately prior to adding to drum mixtures, this is recommended but not required. • RAP and RAS are heated approximately 30 minutes before mixing. • Do not heat the RAP or RAS for mix design purposes. These materials are not preheated in real world produc- tion; preheat virgin aggregate and then add RAP and/or RAS at room temperature before introducing the new asphalt cement. Times and temperatures are grouped by each of the four AASHTO regions (Region 1—Northeastern, Region 2— Southeastern, Region 3—Mid-America, Region 4—Western). Respondents only provided information about what recycled materials are used in their laboratories; therefore, certain data sets such as RAS heating times and temperatures are limited. In general, the times and temperatures used to heat the materials prior to mixing vary considerably within each region (Table 46). Most, but not all, respondents reduce heating tem- peratures for the recycled materials. Some, but not all, agen- cies also reduce the time used to heat the recycled materials. Although the database is small and not complete, the aver- age and standard deviations of the times and temperatures show general trends (Table 47). Average heating times for aggregates vary from 2.5 to 6 hours and from 1.4 to 3.8 hours for RAP. There is a trend for agencies to reduce the heating times for recycled materials regardless of type or AASHTO region. There is also a trend for agencies to use higher temperatures for heat- ing aggregates than for heating recycled materials (Figure 25). Temperatures tend to be slightly higher for heating RAS or combinations of RAP and RAS then when heating just RAP. Heating temperatures for recycled materials are generally more variable than for heating virgin aggregates. Twelve respondents use a temperature probe placed in the aggregate material while it is heated and eight agencies use the Survey Question: Indicate how the materials are, or are not, combined for heating. Materials Heated Separately Combined and Heated Together Responses per Row % n % n n Aggregate and RAP 52 11 57 12 21 Aggregate and RAS 50 6 58 7 12 RAP Fractions 25 2 75 6 8 RAS Fractions 0 0 100 6 6 RAS with Sand (to avoid clumping) 0 0 100 6 6 RAP and RAS 29 2 71 5 7 Not all survey respondents answered all questions. n = 21. TABLE 45 HEATING OF PARTICULATE MATERIALS FOR MIXING

55 time in the oven to confirm the mixing temperature is reached (Table 48). Agencies are inclined to use the time in the oven more often for defining “at temperature” for recycled materials than actual temperature measurements. Order of Addition to Mixing Bowl Agencies begin their mixing process by adding the preheated aggregates to the mixing bowl (Table 49). When the RAP is added to the bowl depends on whether the RAP is heated separately or combined with the virgin aggregate for heating. RAS may be added at the same time as the RAP or added after the aggre- gate and RAP is either dry mixed or mixed with asphalt. There is no consistent order for the addition of RAS to the mixing bowl. Liquids are usually, but not always, added to the bowl after all of the nonliquid materials. If any additive or rejuvenator is used, it is usually preblended with the asphalt before mixing. Additional respondent comments included: • Mix for 5 minutes after adding asphalt. • Chemical WMA mixing temperature and mixing is per- formed according to the manufacturer’s recommendation or appendix in AASHTO R35. Visual inspection is usually used to assess when materials are satisfactorily mixed and only a few agencies use a time frame for mixing (Table 50). Short-Term Aging The short-term aging of the asphalt mixture is used to simulate any heat-induced aging of the binder during pro- duction. Long-term aging of the asphalt mixture is used to simulate aging of the asphalt mixture that occurs over 7 to 10 years of service life. When recycled materials are included in the mixture, the short-term aging provided the elevated temperatures essential to help the blending of the virgin asphalt binder with the recycled asphalt binder. The most frequently used short-term aging time reported by agencies is 2 hours at temperatures between 275°F and 335°F, depending on the virgin asphalt grade (Table 51). Other times used for short-term aging include 1.5 and 4 hours. One agency uses 15 h ± 3 h at a temperature of 140°F. Sample Compaction Compaction levels, controlled by the number of gyrations used for mix designs, are also representative of the position AASHTO Region Heating Time, hours Temperature, oF (oC) Aggregates RAP RAS Combination RAP and RAS Aggregates RAP RAS Combination RAP and RAS Northeastern 1 2 0.5 0.5 — 265 (129) 230 (110) 230 (110) 230 (110) 1 6 2 2 — 310 (154) 290 (143) 290 (143) — 1 — — — — 315 (157) — — 315 (157) 1 3 2 — — 325 (163) 325 (163) — — 1 3.75 1 1 1 350 (176) 350 (176) 350 (176) 350 (176) Southeastern 2 1.5 1.5 1.5 — — — — — 2 — — — — 300 (149) 300 (146) — — 2 — — — — 320 (160) 320 (160) 320 (160) 320 (160) 2 — — — — 325 (163) 295 (146) 295 (163) 295 (163) 2 6 6 — — 335 (168) 335 (168) — — 2 — — — — 340 (171) 245 (118) — — 2 2 — — — 375 (190) — — — Mid-America 3 — — — — 290 (143) 290 (143) — 290 (143) 3 6 1.5 — — 325 (163) 230 (110) — — 3 — — — — 345 (173) 300 (149) 300 (149) 300 (149) 3 6 1 — — 350 (176) 335 (168) — — Western 4 1.5 1.5 1.5 — — — — — 4 — — — — 290 (143) 175 (81) 175 (80) 175 (80) 4 — — — — 290 (143) 290 (143) 290 (143) 290 (143) 4 4 4 4 4 295 (146) 295 (146) 295 (146) 295 (146) 4 2 — — 2 325 (163) — — 325 (163) 4 — — — — 340 (171) — — — — No information provided. n = 22. TABLE 46 MIXING HEATING TIMES AND TEMPERATURES FOR RAP AND/OR RAS MIXTURES

56 AASHTO Region Statistics Aggregates RAP RAS Combination RAP and RAS Heating Time, hours Northeastern Average 3.7 1.4 1.2 1.0 Std. Dev. 1.7 0.8 0.8 — Southeastern Average 3.2 3.8 1.5 — Std. Dev. 2.5 3.2 — — Mid- America Average 6.0 1.3 — — Std. Dev. 0.0 0.4 — — Western Average 2.5 2.8 2.8 3.0 Std. Dev. 1.3 1.8 1.8 1.4 Temperature, oF Northeastern Average 313 299 290 298 Std. Dev. 30.9 52.0 60.0 61.7 Southeastern Average 333 299 308 308 Std. Dev. 25.0 34.2 17.7 17.7 Mid- America Average 328 289 300 295 Std. Dev. 27.2 43.7 — 7.1 Western Average 308 253 253 271 Std. Dev. 23.1 67.9 67.9 66.0 Temperature, oC Northeastern Average 156 148 143 148 Std. Dev. 17.2 28.7 33.0 34.0 Southeastern Average 162 148 162 162 Std. Dev. 8.5 19.0 2.1 2.1 Mid- America Average 164 143 149 146 Std. Dev. 14.9 24.1 — 4.2 Western Average 149 123 123 133 Std. Dev. 9.6 36.7 37.3 36.4 — No data provided or only one value so standard deviation cannot be calculated. n = 22. TABLE 47 AVERAGE HEATING TIMES AND TEMPERATURES FOR MIXING of the mixture in the pavement layer and/or traffic levels. That is, a higher number of gyrations suggest mixture designs for mixtures closer to the surface (i.e., wear courses) or higher traffic levels. Most agency specifications allow for a range of compaction levels, but several states use a single number of gyrations for all mixtures (Table 52). A few states commented that the Marshall mix design method is still used to design larger aggregate size (i.e., bases) and SMA mixtures (typically more gap-graded surface mixtures). Respondent comments included: • Use kneading compactor for mix design. • Do not change the gyrations NDesign when using RAP; do not use RAS at this time, but do not anticipate changing gyration level if RAS is used. • Use different NDesign values based on ESAL Class. Class 2 = 30 million ESALs NDesign = 100. • Use Marshall compaction, AASHTO T245, at 75 blows/ side. Mixtures may be designed using Superpave gyratory compactor at 65 gyrations; OGFC mixtures designed at either 50 Marshall blows or 50 gyrations, but no recycled materials are allowed. • NDesign gyrations are based on the traffic level of the mix design: TL-A = 50, TL-B = 65, TL-C = 75, TL-D and TL-E = 100. • RAP/RAS is not allowed in SMA or surface courses; 12% asphalt replaced by recycled material is allowed in surface course if not SMA and the RAP/RAS Special Provision is used. Our specs are by virgin asphalt type with no distinction by course.

FIGURE 25 Statistics for times and temperatures used for mixing. Survey Question: Indicate how you know the material is at required temperature for mixing. Materials Based on Time in Oven Temperature Probe in the Material While it Is in the Oven Temperature Measured Immediately after Removing from Oven Responses per Row % n % n % n n Aggregates 38 8 57 12 19 4 21 RAP 61 11 44 8 17 3 18 RAS 55 6 46 5 18 2 11 Combined RAP and RAS 44 4 56 5 22 2 9 Not all survey respondents answered all questions. n = 22. TABLE 48 DETERMINING REQUIRED MIXTURE TEMPERATURES Materials Survey Question: Indicate which materials are added to the mixing bowl and in what order. Order of Addition of Materials to Mixing Bowl 1st 2nd 3rd 4th % n % n % n % n Aggregates Aggregates, all fraction (sieve sizes) 95 19 5 1 0 0 0 0 Recycled Materials RAP, coarse 35 7 55 11 5 1 0 0 RAP, fine 35 7 45 9 10 2 0 0 RAS 25 5 15 3 10 2 0 0 Liquids Asphalt 10 2 25 5 45 9 10 2 Rejuvenator 0 0 0 0 0 0 0 0 Asphalt and rejuvenator preblended prior to start of mix design sample prep 5 1 10 2 10 2 5 1 Not all survey respondents answered all questions. n = 20. TABLE 49 ORDER OF ADDITION OF MATERIALS INTO MIXING BOWL

58 Survey Question: Indicate how long materials are mixed (mixing time). Value Mixing Times After Each Material(s) Are Added to the Mixing Bowl Added 1st Added 2nd Added 3rd Added 4th % n % n % n % n 1 minute 0 0 0 0 0 0 0 0 2 minutes 8 1 17 2 8 1 0 0 Based on visual inspection of uniformity 83 10 92 11 100 12 25 3 Not all survey respondents answered all questions. n = 12. TABLE 50 MIXING TIMES AFTER THE ADDITION OF EACH GROUP OF MATERIALS TO THE MIXING BOWL Survey Question: If used, enter time and temperatures used for short-term aging of the compacted samples. Mix with RAP Mix with RAP and RAS Mix with RAS Comments Time, h Temp., oF Time, h Temp., oF Time, h Temp., oF 2 300 2 300 2 300 Temperatures estimated; use AASHTO R30, which requires conditioning for 2 h ± 5 min at the required compaction temperature ±3°C. Mixing and compaction temperatures are provided by the asphalt supplier. 2 275 2 275 2 275 Compaction temperature is specific to virgin binder — 300 300 — 300 Chemical WMA aged according to AASHTO R30. — — — — — — 2 hours at compaction temp. 2 — 2 — 2 — Mixing and compaction temperatures. These temperatures are specified by the contractor and listed on the mix design. 2 275–280 — — 2 275–280 Curing temps are based on suppliers recommended compaction temperatures for a binder. 2 300 2 300 2 300 — 2 275 — — — — — 2 compaction — — 2 — 2 — 2 — Held at compaction temperatures; depends on binder grade. 1 10°C above mixing temp — — — — Follow ASTM D6926 and AASHTO T312 15 ± 3 140 — — — — — AASHTO Spec — — — — — — 2 — 2 265/ 300 2 — Use 265°F for PG 64-22 conditioning and compaction. Use 300°F for PG 76-22 4 — 4 295–335 4 — Aging per R30. Compaction temperatures vary based on virgin binder grade. 1.5 315 — — — — — 2 280/300 — — — — — 4 295 4 295 4 295 — 2 300 ± 25 2 300 ± 25 2 300 ± 25 300°F–325°F depending on PG binder grade of 67-22 or 76-22. 2 300 ± 20 2 300 ± 20 2 300 ± 20 Guidelines say to age for 2 hours in a forced air draft oven at compaction temperature, which is about 280°F–320°F, which is typical JMF temperature. Not all survey respondents answered all questions. — No answers were provided. TABLE 51 SHORT-TERM AGING PRACTICES

59 Survey Question: Enter the typical number of gyration(s), NDesign, which is (are) used to compact recycled material mixtures in the text boxes. Agency Number Wear Course Binder Course Base Course 25% or less RAP More than 25% RAP RAS RAP and RAS 25% or less RAP More than 25% RAP RAS RAP and RAS 25% or less RAP More than 25% RAP RAS RAP and RAS 1 75 — — — 75 — — — 75 — — — 2 50 to 125 50 to 125 — — — — — — — — — — 3 100 80 — — — — — — 100 80 — — 4 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 50, 65, 80 5 50, 75 50, 75 — — 50, 75 50, 75 — — 50, 75 50, 75 — — 6 40, 60, 90 40, 60, 90 40, 60, 90 40, 60, 90 40, 60, 90 40, 60, 90 40,60,90 40,60,90 40, 60, 90 40, 60, 90 40, 60, 90 40, 60, 90 7 — — — — — — — — — — — — 8 50 50 — — 50 50 — — 50 50 — — 9 75, 100 — — — — — — — — — — — 10 65 65 — — 65 65 65 — 70 blow 70 blow 70 blow 70 blow 11 80 — — — 80 — — — 80 — — — 12 65 65 65 — 65 65 65 — 65 65 65 — 13 50 — — — 50 — — — 50 — — — 14 85 — — — 85 — — — 50 50 — — 15 75, 100 — — — — — — — — — — — 16 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 17 60 60 60 60 60 60 60 60 60 60 60 60 18 65 65 65 65 65 65 65 65 65 65 65 65 Agency Number SMA Pervious/Permeable Comments 25% or less RAP More than 25% RAP RAS RAP and RAS 25% or less RAP More than 25% RAP RAS RAP and RAS 1 — — — — — — — — — 2 — — — — — — — — Gyrations depend on ESALs. Typically, binder and base course arethe same as the wearing course. 3 — — — — — — — — 100 gyration mixtures used in severe duty applications; stricter limits on allowable recycled materials in 100 gyration wearing courses. RAS equally likely to be in our 100, 80, or 65 gyration mixtures. 4 50 50 50 50 50 50 — — RAP/RAS not allowed in SMA or surface course. 12% binder replaced is allowed in surface course if not SMA and the RAP/RAS Special Provision is used. 5 — — — — — — — — — 6 — — — — — — — — 40, 60, and 90 gyrations typically used levels 7 — — — — 50 — — — — 8 60–75 60–75 — — — — — — — 9 — — — — — — — — — 10 65 — — — — — — — Base mixture is 6 inch Marshall 11 — — — — — — — — Gyration requirements based on traffic load 12 — — — 50 50 — — — 13 50 — — — 50 — — — — 14 75 — — — 50 — — — Number of gyrations depends on traffic. 15 100 — — — — — — — — 16 50, 75, 100 50, 75, 100 50, 75, 100 50, 75, 100 — — — — — 17 60 60 60 60 — — — — — 18 35 35 35 65 — — — — Use Marshall hammer for pervious mixtures. Marshall hammer may be also used for SMA mixtures. Not all survey respondents answered all questions. — No data provided. TABLE 52 NUMBER OF DESIGN GYRATIONS (Ndesign)

60 Sample Preparation—Section Summary Sizing of Aggregates and Recycled Materials • If the contractors submit RAP samples from fraction- ated stockpiles, a number of agencies appear to use what is submitted as individual fractions for batching. • Definitions of coarse and fine RAP fractions vary. The most frequently used screens for retained on/passing are the 4.75-mm (No. 4) and 2.36-mm (No. 8). The sieve size used for designation may be specified by the agency or determined by the contractor. • Definitions of coarse and fines for production testing may be different from those used for design purposes. • A few agencies fractionate aggregates and RAP into the full range of individual fine aggregate sieve sizes (1.18 mm, 0.6 mm, 0.3 mm, 0.15 mm, and 0.075 mm) for batching. • Comments received indicate an agency may fractionate RAP for batching, even if the contractor does not fraction- ate for production or an agency may not fractionate RAP. Drying of Recycled Materials Prior to Batching • There is no standardized method for drying recycled materials prior to batching. • There is a high level of variability in what is “constant mass” (0.05%, 0.1%, 0.5% maximum change in mass; change of 5 grams maximum), how to dry the material (fan, oven), at what temperature [from room tempera- ture to 300°F (149°C)], and for how long (from 1 hour to overnight). • Particulate sizes used for batching mix design samples are dependent on how each contractor manages its asphalt plant stockpiles for both aggregates and recy- cled materials. • Additional sizing (fractionating) may be used in the labo- ratory to improve laboratory control over the gradation. Heating (Times and Temperatures) of Materials Prior to Mixing • Some agencies heat virgin aggregates, RAP, and RAS separately, whereas others combined aggregated and RAP before heating. – Agencies that use a combination of RAP and RAS typ- ically combine these recycled materials before heating. – RAS may be combined with sand before heating to avoid RAS clumping. • Although heating times and temperatures vary widely, there appear to be a few general trends: – Virgin aggregates tend to be preheated at higher tem- peratures than recycled materials when these materi- als are not combined before heating. – Preheating temperatures for recycled materials are generally lower than those used for virgin aggregates. – There appear to be more variability in preheating temperatures for recycled materials than for preheat- ing virgin aggregates. – Some agencies do not preheat the recycled materials. Order of Addition to Mixing Bowl • Virgin aggregates, followed by the recycled materi- als (if not already mixed with the aggregate), are usu- ally added to the mixing bowl followed by the asphalt binder and any liquid additives. • Materials are typically mixed until they appear to be uniformly coated, although a few agencies use specified times to achieve adequate mixing. Short-Term Aging Prior to Compaction • Short-term aging time is most frequently 2 hours at temperatures between 275°F and 335°F, depending on the virgin asphalt grade. Other practices include 1.5 h, 4 h, and 15 h ± 3 h at a temperature of 140°F. Compaction of Samples • Gyratory compaction is fixed by mixture type (which reflects different traffic levels) by a number of agencies, regardless of the percentage or type of recycled material. • Other agencies use the traffic (ESALs) to define NDesign. • A few agencies still use Hveem or Marshall mix designs for base mixtures, SMA, OGFC, and other specialty mixtures. MIXTURE TESTING A number of questions were included in the agency survey to document the impact of high RAP percentages, RAS materi- als and/or a combination of RAP and RAS recycled materi- als have on the required mix design volumetrics anticipated mixture performance from laboratory testing and perceptions of pavement performance. Volumetrics Survey respondents were asked about their perception of any changes in the mix design volumetrics resulting from the addition or percentage of recycled materials compared with similar mixtures without recycled materials (Table 53). Between 18 and 22 agencies consistently answered specific questions about preparing, mixing, and compacting recycled material asphalt mixtures. However, at most, 10 agencies indicated that it can be difficult to obtain the required mixture volumetrics, which suggests that at least half of the agen- cies responding to these questions do not consider that the recycled materials adversely impact mixture volumetrics.

61 The dust-to-asphalt ratio, air voids, and VMA criteria are more difficult to meet when there is more than 25% RAP, RAS, and a combination of RAP and RAS in the mixtures. Lower percentages of RAP can make it difficult to meet these requirements, but not as frequently. Performance-Based Mixture Testing Performance-based mixture testing is used to determine if the asphalt mixtures, as designed, can achieve the desired service life and successfully resist showing evidence of key pavement distress(es). Long-Term Aging Samples used for evaluating performance-based mixture prop- erties may be subjected to long-term aging to simulate the heat and oxidation hardening of asphalt binders that occurs during 7 to 10 years of in-service use. Only two agencies indicated they use long-term aging (Table 54), whereas a number of agencies commented they do not use long-term aging. Performance-Based Mixture Testing Survey respondents were asked which test methods and test temperatures are used to evaluate mixture rutting, stiffness, traffic-related cracking, and thermal cracking. Twenty-five agencies indicated they use some type of testing to evaluate the rutting potential of asphalt mixtures (Table 55). The most frequently used devices are the APA and Hamburg loaded wheel devices, which are used during mix designs and/or for approving material changes during construction. Nine agen- cies are exploring the use of the AMPT device for studies that are investigating the dynamic modulus frequency sweeps over a range of temperatures and for determining creep char- acteristics (i.e., flow number and flow time) of the mixtures at warmer temperatures. Two agencies noted that they are using the AMPT device for research purposes only at this time. One agency uses the Hveem Stabilometer. Additional respondent comments noted that they evaluate the rut resistance of the mixture using the: • Hamburg device for every 20,000 tons produced, • APA when questions arise about a submitted design, or • Only for high gyration mixtures (NDesign of 100 and 125). Eight agencies evaluate the mixture stiffness by measur- ing the dynamic modulus over a range of temperatures or at a single temperature for research purposes (Table 56). Only two agencies evaluate mixture stiffness during the mix design process. Another agency noted it will get an AMPT test device for dynamic modulus testing at the end of its fatigue research project. One agency commented that it does not conduct performance mixture testing in the laboratory. Survey Question: Check the box if it is more difficult to obtain acceptable properties when compared to similar mixtures without any recycled material content. Materials Air Voids, % VMA, % VFA, % Dust-to-Asphalt Ratio % n % n % n % n 25% or less RAP 50 5 40 4 20 2 50 5 More than 25% RAP 90 9 100 10 60 6 100 10 RAS mixtures 60 6 70 7 30 3 60 6 RAP and RAS combination mixtures 70 7 70 7 30 3 80 8 Not all survey respondents answered all questions. n = 10. TABLE 53 INFLUENCE OF RECYCLED MATERIAL ON ACHIEVING DESIRED MIX DESIGN VOLUMETRICS Survey Question: If used, enter time and temperatures used for long-term aging of the compacted samples. Mix with RAP Mix with RAP and RAS Mix with RAS Comments Time, h Temp., oF Time, h Temp., oF Time, h Temp., oF 4 300 4 300 4 300 For chemical WMA, use AASHTO R30. 120 185 120 185 120 185 — Not all survey respondents answered all questions. n = 2. TABLE 54 LONG-TERM AGING PRACTICES

62 At this time, only three agencies investigate traffic- related cracking of mixtures in their state (Table 57). Two agencies use the disc-shaped compact tension for research purposes (DCT). One agency uses the SCB test results at intermediate temperatures as a means of allowing the contractor to use higher percentages of fractionated RAP. Two agencies indicated that the DCT test is currently in development or is only used for information. Another two agencies are considering selecting a fatigue test or evaluat- ing the SCB test. Four agencies currently use the IDT test to evaluate the thermal cracking potential of their mixtures for research pur- poses (Table 58). Two of these agencies use this test method to approve changes in materials during construction and one agency uses the test method during mix designs. Only one agency uses the SCB test method for approving changes in materials during construction. Three agencies are currently researching the SCB and DCT methods. Volumetric and Performance-Based Mixture Testing—Section Summary Mix Design Volumetrics • Mix design volumetrics are perceived to be more dif- ficult to obtain for mixtures with recycled materials. • When the percentage of RAP increases above 25%, the likelihood of having difficulty in achieving the desired mix design volumetrics increases. Performance-Based Mixture Testing • Rutting potential is the most frequently evaluated per- formance characteristic during mix designs, to approve material changes during construction, and for research studies. A single agency may have more than one device for different applications. A range of devices are currently used. Survey Question: Rutting: If you evaluate the rutting potential of mixtures in your lab, please indicate which method(s) you use. (Choose all that apply.) Test Method Used Routinely for Our Mix Designs Use When Approving Changes in Materials During Construction Use for Research Studies Response per Row % n % n % n n Asphalt Pavement Analyzer (APA) 63 8 15 2 62 8 13 Hamburg Rut Tester 63 5 13 1 88 7 8 Wet rut testing to determine stripping inflection point 67 2 0 0 100 3 3 Asphalt Mixture Performance Test (AMPT) 0 0 0 0 100 9 9 Dynamic modulus 0 0 0 0 100 9 9 Flow number 0 0 0 0 100 8 8 Flow time 0 0 0 0 100 4 4 Hveem Stabilometer 100 1 100 1 0 0 1 Not all survey respondents answered all questions. n = 25. TABLE 55 TEST METHODS FOR EVALUATING RUTTING POTENTIAL Survey Question: Mixture Stiffness: If you evaluate mixture stiffness in your lab, please indicate which method(s) you use. (Choose all that apply) Materials Used Routinely for Our Mix Designs Use When Approving Changes in Materials During Construction Use for Research Studies % n % n % n Resilient modulus at a single temperature 13 1 13 1 38 3 Resilient modulus at several temperatures 0 0 0 0 38 3 Dynamic modulus at a single temperature 13 1 13 1 75 6 Dynamic modulus over a range of temperatures to develop a master curve 0 0 0 0 100 8 Indirect tensile strength 13 1 13 1 13 1 Not all survey respondents answered all questions. n = 8. TABLE 56 TEST METHODS USED TO EVALUATE MIXTURE STIFFNESS

63 • Mixture stiffness is evaluated using either resilient modu- lus or dynamic modulus, primarily for research purposes. • Cracking potential is evaluated primarily for research purposes at this time using one or more methods (DCT and SCB at low temperatures, SCB at intermediate temperatures). • One agency allows fractionated RAP to be used at higher percentages as long as SCB testing is conducted. PERCEIVEd INFLUENCE OF RECYCLEd MATERIALS ON PAVEMENT PERFORMANCE Respondents were asked to indicate their level of agreement or disagreement with a series of statements about the impact of recycled materials on pavement performance based on their experiences. Most either agree or strongly agree that rutting resistance is improved when using RAS or a combination of RAS and RAP, and increasing percentages of RAP (Table 59). Additional comments provided by the respondents included: • Recycled materials improve rutting resistance because of stiffer materials. • Although everyone understands that recycled material typically increases stiffness and decreases rutting poten- tial, because recycled asphalt does not homogenously mix with virgin binder, the use of softer grades required by very high RAP mixtures could lead to rutting when mixing is incomplete. • The answer given is based on today’s specs. If we would increase the required film thickness, or VMA or asphalt content required in design, my answer will likely change. • There are many factors that can increase a mixture’s rut- ting potential. Increasing the percentage of RAP alone is not one of them. Agency responses are about evenly divided between believ- ing that the types and percentages of recycled material do not noticeably influencing mixture moisture sensitivity and noting they believe the moisture sensitivity increases (Table 60). Sim- ilar responses were received with regard to changes in the mix- ture IDT. Respondents are about evenly split between noting little change is expected and the tensile strength is expected to increase with increasing percentage and/or type of recycled materials. Additional respondent comments received about no significant problems with moisture sensitivity included: • Have not experienced any moisture sensitivity issues regardless of the percentage of RAP used. Survey Question: Cracking (Nonthermal): If you evaluate cracking potential of mixtures in your lab, please indicate which method(s) you use. (Choose all that apply.) Materials Used Routinely for Our Mix Designs Use When Approving Changes in Materials During Construction Use for Research Studies % n % n % n Fatigue cracking, bending beam (AASHTO T321) 0 0 0 0 0 0 Overlay tester 0 0 0 0 0 0 Disc-shaped compact (DCT) tension test (ASTM D7313) 0 0 0 0 67 2 Semi-circular bend (SCB) test 0 0 0 0 33 1 Not all survey respondents answered all questions. n = 3. TABLE 57 EVALUATING NONTHERMAL CRACKING POTENTIAL Survey Question: Thermal Cracking: If you evaluate the thermal cracking potential of mixtures in your lab, please indicate which method(s) you use. (Choose all that apply.) Materials Used Routinely for Our Mix Designs Use When Approving Changes in Materials During Construction Use for Research Studies % n % n % n Indirect tensile strength (AASHTO T322) 20 1 40 2 80 4 Semi-circular bend (SCB) test 0 0 20 1 60 3 Disc-shaped compact (DCT) tension test (ASTM D7313) 0 0 0 0 60 3 Not all survey respondents answered all questions. n = 5. TABLE 58 EVALUATING THERMAL CRACKING POTENTIAL

64 • Virtually all of our agency mixtures employ a liquid anti- stripping agent with some hydrated lime used as well. • Using the proper asphalt binder and/or other additives can counteract the negative effects of increased RAP percentages. Respondent comments about why they experience mois- ture sensitivity problems included: • RAS, if from tear-offs, have the potential for greater issues with tensile strength ratio (TSR) testing than do manufactured waste shingles. • Observed increased dust with recycled materials, less active or soft binder to promote coating, and high strength resulting from increased stiffness. Other respondent comments about moisture sensitivity included: • Test Hamburg wet and require the TSR (Lottman) to be 80% retained after freeze/thaw cycle. • Differentiate between wet or dry strength. • Use Immersion Compression test for moisture sus- ceptibility. The majority of respondents agree or strongly agree that the percentage or type of recycled materials increases the traffic- related cracking potential (Table 61). Additional respondent comments about traffic-related cracking potential included: • Successful implementation of mix design parameters that include crack testing may mitigate my concern in this area. • Depends on where the mixtures are used within the pavement structure. • This is why we do not allow RAP in surface courses. • Answered neutral here because an increase in the per- centage of RAP could increase the potential for non- thermal cracking if the proper virgin binder is not used to rejuvenate the RAP binder. The majority of respondents agrees or strongly agrees that the percentage or type of recycled material increases the ther- mal cracking potential (Table 62). Additional comments by respondents included: Survey Question: Mixture Durability Potential: Based on your experience, indicate your level of agreement with the following statements. Statement Strongly Agree Agree Neutral Disagree Strongly Disagree Responses per Row % n % n % n % n % n n Moisture Sensitivity Moisture sensitivity is increased with increasing percentages of recycled RAP 4 1 30 7 57 13 9 2 0 0 23 Moisture sensitivity is increased with increasing percentages of recycled shingles 12 2 35 6 47 8 6 1 0 0 17 Moisture sensitivity is increased with a combination of RAP and RAS 12 2 35 6 47 8 6 1 0 0 17 Mixture Strength Indirect tensile strength is increased with increasing percentages of recycled RAP 10 2 45 9 40 8 5 1 0 0 20 Indirect tensile strength is increased with increasing percentages of recycled shingles 12 2 47 8 35 6 6 1 0 0 17 Indirect tensile strength is increased with a combination of RAP and RAS 12 2 41 7 41 7 6 1 0 0 17 Not all survey respondents answered all questions. n = 23. TABLE 60 PERCEIVED INFLUENCE OF RECYCLED MATERIAL ON MIXTURE DURABILITY POTENTIAL Survey Question: Rutting Potential: Based on your experience, indicate your level of agreement with the following statements. Statement Strongly Agree Agree Neutral Disagree Strongly Disagree Responses per Row % n % n % n % n % n n Rutting potential is increased with increasing percentages of recycled RAP 0 0 4 1 13 3 46 11 38 9 24 Rutting potential is increased with increasing percentages of recycled shingles 0 0 0 0 28 5 28 5 44 8 18 Rutting potential is increased with a combination of RAP and RAS 0 0 0 0 18 3 35 6 47 8 17 Not all survey respondents answered all questions. n = 24. TABLE 59 PERCEIVED INFLUENCE OF RECYCLED MATERIAL ON RUTTING POTENTIAL

65 Survey Question: Cracking Potential (Nonthermal Cracking): Based on your experience, indicate your level of agreement with the following statements. Statement Strongly Agree Agree Neutral Disagree Strongly Disagree Responses per Row % n % n % n % n % n n Cracking potential is increased with increasing percentages of recycled RAP 21 5 50 12 29 7 0 0 0 0 24 Cracking potential is increased with increasing percentages of recycled shingles 50 9 39 7 11 2 0 0 0 0 18 Cracking potential is increased with a combination of RAP and RAS 35 6 53 9 12 2 0 0 0 0 17 Not all survey respondents answered all questions. n = 24. TABLE 61 PERCEIVED INFLUENCE OF RECYCLED MATERIAL ON NONTHERMAL CRACKING POTENTIAL • Do not have a problem with this as our specified asphalt grade is PG xx-22 and our low temp design requirement is closer to a PG xx-16. • Depends on where the mixtures are used within the pave- ment structure. PERCEPTIONS OF INFLUENCE OF RECYCLEd MATERIALS ON PAVEMENT PERFORMANCE— SECTION SUMMARY Based on respondent’s experience and perceptions it is likely that: • Rut resistance can decrease with increasing percentages of any of the recycled material. • Moisture sensitivity may increase with increasing per- centages of recycled materials; however, almost half the respondents believe that the recycled materials may not have any influence one way or the other (i.e., neutral position). • Nonthermal cracking increases with increasing per- centages of RAP and combinations of RAP and RAS. – Respondents are more likely to strongly agree that increasing RAS increases nonthermal types of cracking. • Thermal cracking potential increases as the percentage of RAP increases. – Respondents are more likely to strongly agree that increasing RAS or combinations of RAP and RAS percentages increases thermal cracking. Survey Question: Thermal Cracking Potential: Based on your experience, indicate your level of agreement with the following statements. Statement Strongly Agree Agree Neutral Disagree Strongly Disagree Responses per Row % n % n % n % n % n n Thermal cracking potential is increased with increasing percentages of recycled RAP 5 1 77 17 18 4 0 0 0 0 22 Thermal cracking potential is increased with increasing percentages of recycled shingles 25 4 63 10 13 2 0 0 0 0 16 Thermal cracking potential is increased with a combination of RAP and RAS 20 3 60 9 20 3 0 0 0 0 15 Not all survey respondents answered all questions. n = 22. TABLE 62 PERCEIVED INFLUENCE OF RECYCLED MATERIAL ON THERMAL CRACKING POTENTIAL