When incorporated RAP is greater than 20%, suggested testing both the RAP and virgin binder (bitumen) using a blending chart. At a higher percentage the change in RAP mixture characteristics becomes evident; the effect of RAP in mix designs varies with the quantity of RAP in the mix. This combination must meet gradation specifications and Superpave consensus properties. Puttaguanta and others, (1996), found no significant difference when 25-50% of RAP is incorporated into the mixes. In summary, observations from literatures reviewed showed that low percentages (5-20%) RAP can be incorporated in superpave mix designs. Flow charts to design of RAP mixes are shown in Figure 2.2
(a) Step 1
(b) Step 2
(c) Step 3
Figure 2.2 RAP mix design procedure
2.3. Aggregate Tests
The specific aggregate tests are discussed in the subsequent section.
2.3.1 Gradation of RAP Aggregate
The gradation investigation is used to determine aggregate particle size distribution. The aggregate gradation is one of the most important properties related with the control of HMA mixes (Pavement Interactive, 2011). The rutting feature of pavements is controlled by aggregate gradation. The standard gradation and sieve analysis can be conducted according to AASHTO T27 and ASTM C136 “Sieve Analysis of Fine and Coarse Aggregate”. Millings are material fashioned by the milling method, removed from the existing pavement. Milling and crushing leads to RAP degradation, Milled RAP is finer than crushed RAP. RAP aggregate normally meets the ASTM specifications based on ASTM D692. The particle size distribution of a milled pavement varies depending on aggregate and equipment type. A representative of RAP size distribution is shown Table 2.2.
Table 2.2 Characteristic range of particle size distribution for reclaimed RAP
Sieve Size Percent Finer after Processing or Milling
37.5 mm (1.5 in) 100
25 mm (1.0 in) 95 – 100
19 mm (3/4 in)
84 – 100
12.5 mm (1/2 in) 70 – 100
9.5 mm (3/8 in) 58 – 95
7.5 mm (No. 4) 38 – 75
2.36 mm (No. 8) 25 – 60
1.18 mm (No. 16) 17 – 40
0.60 mm (No. 30) 10 – 35
0.30 mm (No. 50) 5 – 25
0.15 mm (No. 100) 3 – 20
0.075 mm (No. 200) 2 – 15
2.3.2 Specific Gravity
HMA volumetric examination is achieved through specific gravity test. Results of the test performed by William (2007), suggested that the traditional saturated surface dry method display the least extent of variation. The model for determining specific gravity of coarse and fine aggregates is described in ASTM C127 and ASTM C128, correspondingly. Specific gravity of RAP aggregates are determined by:
Technique 1: Specific gravity tests are carried out on sieved fractions according to ASTM C127 and ASTM C128 after obtaining aggregates either by solvent or ignition oven extraction process. The extraction method could modify the aggregate specific gravity. Mallick and others (1998) demonstrate the differences that exist in the specific gravity of virgin and aggregates obtained from ignition extraction method.
Technique 2, calculation based on the theoretical specific gravity
Gse = 100 – Pb
100 – Pb
Gse = aggregate effective specific gravity
Gmm= theoretical maximum specific gravity (AASHTO T209)
Pb = bitumen content (percent by total mass of mixture); and Gb= specific gravity of bitumen.
Gsb = Gse
Pba . Gse
100 . Gb + 1
Where: (El Sayed, 2012)
Pba= absorbed bitumen (percent by weight Gsb of aggregate); Gse= aggregate effective specific gravity; Gsb= aggregate bulk specific gravity; and Gb = RAP binder specific gravity.
The absorbed binder content in the aggregate is only assumed as the accurate binder content is difficult to determine (El Sayed, 2012).
2.3.3 Los Angeles Abrasion Test
RAP must be hard enough to withstand breakdown when subjected to traffic tonnes. Aggregates lacking enough hardness may result in construction failure (Hameed, 2009). About 94 percent of the states in the U.S prefer the Los Angeles abrasion test (Pavement Interactive, 2011). Ahmad and others (2004) performed research on degradation and abrasion of RAP aggregate in Malaysia. They focused on aggregates extracted from RAP from both full-depth recovery and milling. They concluded that the aggregate clearly degraded by further refinement of aggregate size, retaining considerable strength to defy wear and abrasion. This study also showed that millings are finer and denser than virgin aggregate.
2.3.4 Aggregate Crushing Value Test
Aggregates should have an acceptable resistance to crushing and enough strength under loads (Han and others, 2011). Low aggregate crushing value is preferable
ACV = B x 100
Where, B = Weight of the fraction passing 2.36mm sieve
A = Total weight of dry sample
The mean of the two tests equals the aggregate crushing value (Al kourd and Hammad, 2009). The standard specification is < 30%
2.3.5 Aggregate impact value Test
Aggregate impact value is a measure of the toughness of the material. The aggregates should therefore have enough toughness to defy disintegration.
Weight of dry sample (W1 gm)
Weight of fraction passing 2.36 mm sieve (W2 gm)
Aggregate impact Value (percent) = W2 / W1 X 100 (Transportation Engineering Lab Manual, 2013; The constructor, 2015).
Aggregate Impact Value Toughness properties
<20% Extraordinarily tough
10 – 20% Very tough
20-30% Satisfactory for surface pavement
>30% Weak for surface pavement
2.3.6 Flakiness and Elongation Index (Shape Test)
Proportion of flaky and elongated particle in a mixture determine aggregate shapes and arrangements, flaky and elongated aggregates are damaging as they may cause intrinsic flaw under heavy tonnes. An excessive amount of these materials in the HMA mix may lead to construction failures.
Aggregates are classified flaky if: Width > 2.0
Aggregate are classified elongated if: Width > 2.5
2.4 RAP Variability
One of the main worry of incorporating a large percentage of RAP in HMA mixes is variability. Opus of RAP from diverse sources varies. The exact make-up of milled RAP depends on age, type, properties of bitumen, configuration and performance of the milling process. The consistency of aggregate is determined by evaluating gradation, specific gravity, coarse aggregate angularity, and fine aggregate angularity (Pavement Interactive, 2011). The following, affects the consistency of binder: fatigue factor (G*sin (δ)), complex modulus (G*), and phase angle (δ°). Lee and others (1983) in an effort to quantify the plant mixing efficiency of asphalt mixtures with incorporation of RAP observed that the performance of the recycled mix depends on interaction between the virgin aggregate and asphalt binder, which indicates the need for consistency of samples. Mixing efficiency is usually measured by bituminous mixtures appearance in respect to distribution and coating. Lee and others (1983) resolved the difficulty associated with rejuvenating agents by a “Dye Print Technique”. A variability study on RAP stockpile (Figure 2.3), different test results, such as gradation of aggregate, asphalt content, air void, penetration and viscosity, and stability, are shown in (Solaimanian and Tahmoressi, 1996). Figure 2.4 shows the variations in gradation obtained in their research. Incorporating higher percentage of RAP shows a higher variation in gradation as observed in Figure 2.5. Changes in the viscosity of the binder from the RAP and plant mix are shown in Figure 2.6.
Figure 2.3 RAP Stockpile
Figure 2.4 Daily gradations from extraction
Figure 2.5 Mean deviations from Job Mix Formula (JMF) for Sieve #10
Figure 2.6 Sample daily viscosities
Table 2.3 shows that RAP obtained from road cores were highly variable and the aggregate gradation became finer after milling and processing (Han and others, 2011).
Table 2.3 RAP compositions of cores and stockpiles
Location of roads % passing
2.36 mm % passing 0.075
mm Asphalt Cement
n ave. σn-1 ave. σn-1 ave. σn-1
California Road Cores 12 54 8.30 9.9 2.01 5.4 0.71
California Stockpiled after
Milling 5 69 6.50 11.8 0.34 5.2 0.04
North Carolina Road Cores 12 69 3.20 6.1 0.66 5.7 0.11
North Carolina Stockpiled
after Milling 5 72 0.90 8.0 0.11 5.7 0.11
Utah Road Cores 12 52 3.80 8.7 2.60 6.5 0.28
Utah Stockpiled after Milling 10 58 2.80 9.9 1.15 6.2 0.44
Virgina Road Cores 12 41 2.10 9.7 0.79 5.3 0.20
Virginia Stockpiled after
Milling 6 52 1.10 13.0 0.30 5.2 0.12
Average σ of HMA Surface
Course 2.81 0.94 0.28
A laboratory test was carryout by Huang and others (2004) using four category of mixtures consisting of various kind of bitumen (PG 64-22, PG 70-22, and PG 76-22) each made-up of varying RAP percentages (Han and others, 2011). 0% RAP mixture serves as the control mixes. Various characterization tests were carryout. The result shows the effect of variability of RAP on virgin mixtures. Figure 2.7 exemplify no major dissimilarity in the aggregate gradation from the worksite stockpiles, different worksites shows a little variation. Changes was recorded FAA and specific gravity of RAP from diverse worksites are shown in Figure 2.8 however, samples obtained from altering locations show no significant difference (Han and others, 2011). Figure 2.13 shows the differences in phase angle of the binder from the worksite CNC 302 and other worksite while Figure 2.9 shows the values of binder from worksite CNC 302.
Figure 2.7 RAP Gradation for worksite a
Figure 2.8 RAP Gradation for worksite b
Figure 2.9 RAP Gradation for worksite c
Figure 2.10 RAP Gradation for worksite d
...(download the rest of the essay above)