CHAPTER 3

Laboratory Work Plan

This chapter presents the laboratory work plan to evaluate the impact of adding PCR plastics via the dry process on the performance properties, surface characteristics, mix production, process control, and constructability of asphalt mixtures. The performance properties of interest include workability, rutting resistance, cracking resistance, moisture susceptibility, and aging resistance. The work plan consists of five experiments, detailed as follows.

3.1 Experiment 1: Characterization and Selection of PCR Plastics

Experiment 1 focused on characterizing the physical, thermal, and chemical properties of 12 PCR plastics from different sources and with different compositions, as shown in Table 2. When selecting the PCR plastics, priority was given to LLDPE, LDPE, HDPE, and PP for two reasons: (1) When combined, they account for the largest proportion of MSW plastics generated; and (2) they have relatively lower melting temperatures compared to other plastics, therefore they are considered most suitable for use in asphalt mixtures. As shown in Table 2, the 12 plastic samples selected in the experiment include 4 single-stream PCR plastics, 6 mixed-stream PCR plastics, and 2 commercial plastic products. One of the commercial plastic products was originally thought to be a blend of HDPE, PP, and calcium carbonate (CaCO₃), which is intended to be used as an aggregate replacement for asphalt mixture applications. However, analysis of the material determined that it is a blend of PET, PP, and CaCO₃. The other commercial plastic product is an HDPE-based asphalt additive processed from waste fishing nets with proprietary modification for performance enhancement.

Figure 10 presents the 12 PCR plastic samples. Most of the samples are in pellet form, except for Samples #6 and #8, which are in shredded form; Sample #11, which is an irregular granule form; and Sample #12, which is a stick-like form.

Table 3 presents the testing matrix of PCR plastics. Most of the tests listed have been well established in the field of plastics and were recommended by the joint NAPA/Asphalt Institute plastics task force for property characterization of recycled plastics (Willis et al., 2020).

3.2 Experiment 2: Laboratory Characterization of Plant-Produced RPM versus Control Asphalt Mixtures and Extracted Binders

Experiment 2 sought to evaluate the performance properties of control versus dry-process RPM mixtures and their corresponding extracted asphalt binders from two field projects in Ohio and Wisconsin. Each field project included a control mixture and an RPM mixture containing

Table 2. Experiment 1: Descriptions of 12 PCR plastic samples.

PCR plastics added via the dry process. A series of laboratory mixture and binder performance tests were conducted for performance evaluation.

Table 4 summarizes the mix design information of the two field projects, including the nominal maximum aggregate size (NMAS), virgin binder type, RAP content, and recycled plastic type and dosage. For the Wisconsin field project, a chemical WMA additive was used to reduce the mixing temperature to 149°C (300°F). Figure 11 shows the PCR plastics used in the two projects, and Table 5 summarizes their properties, including the initial melting temperature from DSC, specific gravity, particle size, and ash content from TGA.

For each field project, the control mixture was identical to the RPM mixture except no PCR plastics were added to the control. A fiber machine was used to feed the PCR plastic pellets into the mixing drums through the RAP collar, as shown in Figure 12. During production, loose mixtures were sampled in 5-gallon metal buckets and cardboard boxes for the Ohio and Wisconsin projects, respectively. After transferring to the NCAT laboratory, the mixtures were first reheated at compaction temperatures for two to four hours, depending on the container size, to split into individual sample sizes for performance testing. For each mixture, two sets of plant-mixed, laboratory-compacted specimens were prepared and tested: reheated (RH) and critically aged (CA) specimens. The split loose mixtures were first cooled to the ambient temperature and then reheated at the compaction temperature for two hours prior to compacting RH specimens. For CA specimens, the loose mixtures were subjected to additional long-term aging for six hours at 135°C (275°F) after reheating. Additionally, asphalt binders were also extracted and recovered from the RH loose mixtures for rheological and chemical characterizations.

Table 6 summarizes the performance testing plan for the control versus RPM mixtures for both field projects. To consider the impact of asphalt aging on cracking resistance, the Dongre Workability Test (DWT), Indirect Tensile Asphalt Rutting Test (IDEAL-RT), Hamburg Wheel Tracking Test (HWTT), and Tensile Strength Ratio (TSR) test were conducted on the RH specimens, while the Indirect Tensile Asphalt Cracking Test (IDEAL-CT), Disc-shaped Compact Tension (DCT) test, and Cyclic Fatigue (CF) test were conducted on CA specimens. The Dynamic Modulus (E*) test was conducted on both RH and CA specimens to assess the aging susceptibility of control versus RPM mixtures.

All mixture performance tests except the DWT were conducted following AASHTO or ASTM test standards. The DWT was conducted per Dongre et al. (2020) to assess the workability of control versus RPM mixtures at 107°C (225°F), 121°C (250°F), and 135°C (275°F). Three replicates were tested at each temperature. Prior to the DWT testing, approximately 4,800 g of the loose mixture at each target temperature was placed into a Superpave gyration

Table 3. Experiment 1: Testing matrix for property characterization of PCR plastic samples.

Table 4. Mix design information for Ohio and Wisconsin field projects.

Table 5. Properties of PCR plastics used in Ohio and Wisconsin field projects.

Table 6. Experiment 2: Mixture performance testing plan.

compactor mold. The DWT was then conducted by applying force on the loose mixtures with a constant rate of 0.05 mm/s without gyrating. During the test, the applied force and specimen height were recorded every 0.1 seconds to calculate the stress and strain. The DWT value of the mixture is defined as the slope of the non-linear compressive stress versus the volumetric percent strain curve at a stress level of 600 kPa. This value was calculated by measuring the ratio of stress change between 650 and 550 kPa to the corresponding change in volumetric strain. A higher DWT value is desired for better workability.

In addition to the mixture performance tests listed in Table 6, slab specimens for each mixture were also prepared and subjected to 150,000 cycles of surface polishing in the NCAT Three Wheel Polishing Device (TWPD) and 1,000 hours of surface weathering in the NCAT Accelerated Weathering System (NAWS), as shown in Figure 13 and Figure 14, respectively. The NAWS operates in accordance with ASTM D4799, Standard Practice for Accelerated Weathering

Conditions and Procedures for Bituminous Materials (Fluorescent UV, Water Spray, and Condensation Method). The slabs were tested with a Circular Track Meter (CTM) and Dynamic Friction Tester (DFT) to evaluate their surface texture and friction properties, respectively, at the following weathering and polishing intervals: (1) before polishing and weathering, (2) after 50,000 polishing cycles and 333 weathering hours, (3) after 100,000 polishing cycles and 666 weathering hours, and (4) after 150,000 polishing cycles and 1,000 weathering hours. The CTM measured the macrotexture profile of the slab surface and provided the macrotexture’s mean profile depth (MPD) in millimeters. The DFT measured surface friction using the friction coefficient at 40 km/h (DFT40). Higher MPD and DFT40 values are desired for better surface texture and friction properties.

Table 7 and Table 8 summarize the laboratory testing plans for characterizing the rheological and chemical properties, respectively, of asphalt binders extracted and recovered from the control and RPM mixtures from the two field projects. All tests were conducted on as-extracted binders without further aging, whereas only the tests for cracking resistance or intermediate PG temperature evaluation were conducted on extracted binders after 20 hours of Pressure Aging Vessel (PAV) aging.

Table 7. Experiment 2: Extracted binder rheological testing plan.

3.3 Experiment 3: Survey on Recycled Plastics in Asphalt Mixtures and Exploratory QC Testing of PCR Plastics

The focus of this experiment was twofold: (1) to survey asphalt contractors experienced in producing RPM asphalt mixtures, and (2) to conduct exploratory QC testing of PCR plastics using a handheld near-infrared spectrometer. The survey was intended to gather information on the supply of PCR plastics in plant-produced RPM asphalt mixtures as well as the constructability, production control, and process control of these mixtures in recent field projects in the

Table 8. Experiment 2: Extracted binder chemical testing plan.

^*Test results subjected to the solubility of waste plastics in the solvents used for binder extraction and analytical evaluation.

^†Test results subjected to the solubility of waste plastics in the solvent used for binder extraction only.

United States. Note that the survey did not request information from contractors who had used recycled plastics added by the wet process. A copy of the survey questionnaire is provided in Appendix B, which can be found on the National Academies Press webpage for NCHRP Research Report 1143 (https://doi.org/10.17226/28867) under “Resources.” For exploratory QC testing of PCR plastics, a handheld near-infrared spectrometer (Figure 15) was used to detect the primary plastic composition of the 12 PCR plastics selected in Experiment 1. The results were compared against the composition analysis results from FTIR. Furthermore, the near-infrared spectrometer

was used to evaluate the batch-to-batch production consistency of PCR plastics from six suppliers on the market. Each selected PCR plastic was tested with 15 production batches.

3.4 Experiment 4: Selection of Laboratory Method for Adding Recycled Plastics via the Dry Process

Experiment 4 examined four laboratory methods of adding PCR plastics in asphalt mixtures via the dry process to determine the best method of simulating the production of dry-process RPM mixtures at asphalt plants. This experiment used the control and RPM mixtures from the Ohio and Wisconsin field projects evaluated in Experiment 2. All the mixtures were prepared by matching the plant mixing temperature, asphalt content, aggregate gradation, and RAP content from the respective contractor’s QC data. Table 9 summarizes the materials and mix design information of the two field projects, including NMAS, virgin binder type, RAP content, and PCR plastic type and dosage.

The four laboratory methods of adding PCR plastics evaluated in this experiment are described as follows:

• Method 1: mixing ambient-temperature PCR plastics with preheated aggregates at 170°C, followed by adding preheated RAP at 135°C and then adding the virgin asphalt binder.
• Method 2: mixing ambient-temperature PCR plastics with superheated aggregates (at a temperature selected per plant manufacturer recommendations based on the RAP content, moisture content of RAP, and plant production temperature of the mixture), followed by adding ambient-temperature RAP and then adding the virgin asphalt binder.
• Method 3: preheating PCR plastics with RAP for two hours at 135°C, then mixing with preheated aggregates at 170°C and the virgin asphalt binder.
• Method 4: preheating PCR plastics with aggregate overnight at 170°C, then mixing with preheated RAP at 135°C and the virgin asphalt binder.

The primary differences between the four methods are the temperatures of the various mixture components (i.e., virgin aggregate, RAP, and PCR plastic) for mixing, as shown in Table 10.

Table 9. Experiment 4: Materials and mix design information for the field projects.

Table 10. Experiment 4: Component material temperature for different methods of adding PCR plastics.

Table 11. Experiment 4: Performance tests for laboratory-prepared mixtures.

The performance properties of laboratory-prepared RPM mixtures created with the four PCR plastic–mixing methods were evaluated, and the results were compared to those of the corresponding plant-produced mixtures from Experiment 2. Laboratory-prepared control mixtures (without PCR plastics) were also tested for comparison. In this experiment, the IDEAL-RT, IDEAL-CT, and DCT test were conducted to evaluate the mixture’s rutting resistance, intermediate-temperature cracking resistance, and low-temperature cracking resistance, respectively. Table 11 summarizes the mixture aging condition, test standard, and test parameter of each performance test. The short-term aging condition consisted of two hours of loose mix aging at 135°C per AASHTO R 30, while the long-term aging condition consisted of six hours of loose mix aging at 135°C following short-term aging.

Fume emission analysis was conducted to evaluate fumes released during the laboratory preparation of the control versus dry-process RPM mixtures (with different methods of adding PCR plastics) from the Ohio and Wisconsin field projects. As illustrated in Figure 16, the analysis followed the National Institute for Occupational Safety and Health (NIOSH) Methods 5042 (gravimetric evaluation of benzene-soluble fraction and total particulate) and 2549 (screening for HAPs and VOCs), in addition to the U.S. EPA Toxic Organic Compounds in Ambient Air Compendium Method TO17 (EPA TO17) and International Organization for Standardization (ISO) 16000-6 for PAHs.

3.5 Experiment 5: Laboratory Characterization of Laboratory-Prepared RPM versus Control Asphalt Mixtures and Extracted Binders

Experiment 5 aimed to characterize the performance properties of laboratory-prepared RPM mixtures containing different types of PCR plastics versus the control mixtures without plastics. Two mix designs, one each from Alabama and Minnesota, were selected to represent mix designs from southern and northern states, referred to as the southern mix design and the northern mix design, respectively. Table 12 summarizes the job mix formulas (JMFs) of the two mix designs.

Table 12. JMFs of southern and northern mix designs.

^*P_ba = absorbed binder.

^†P_be = effective binder content.

Six mixtures were prepared with each mix design: a control mixture without plastics and five dry-process RPM mixtures. The RPM mixtures were prepared with five PCR plastics selected in Experiment 1 (Table 13; see further discussion in Chapter 4). The PCR plastics were added at a dosage of 0.5% by weight of the total aggregate following Method 1 (i.e., mixing ambient-temperature PCR plastics with preheated aggregates at 170°C, followed by adding preheated RAP at 135°C and then adding the virgin asphalt binder), the laboratory mixing procedure selected in Experiment 4 (see further discussion in Chapter 4).

A series of mixture performance tests were conducted to characterize the performance properties of the control versus five dry-process RPM mixtures for each mix design. These tests included the DWT for workability evaluation; the HWTT and IDEAL-RT for rutting evaluation; the TSR test for moisture susceptibility evaluation; and the IDEAL-CT, DCT test, and CF test (including the E* test) for cracking evaluation. The DWT, HWTT, and IDEAL-RT were performed on laboratory-mixed, laboratory-compacted specimens with short-term aging for 2 hours at 135°C per AASHTO R 30. Three tests, IDEAL-CT, DCT, and CF, were conducted on specimens with additional loose mix aging for either 8 hours at 135°C (for the southern mix design) or 6 hours at 135°C (for the northern mix design) after short-term aging (i.e., critical aging). TSR testing was performed on specimens conditioned according to AASHTO T 283. The E* test was performed on both short-term and long-term aged specimens to assess the aging susceptibility of the mixture. Furthermore, slab specimens were prepared for each mixture and subjected to 150,000 cycles of surface polishing with the TWPD and 1,000 hours of surface weathering in the NAWS to simulate the field polishing and weathering of asphalt pavement surfaces. Surface texture and friction of the slabs were measured using the CTM and DFT, respectively, at four weathering and polishing conditions: (1) no polishing nor any NAWS weathering, (2) after 50,000 cycles of polishing and 333 hours of NAWS weathering, (3) after 100,000 cycles

Table 13. Five PCR plastics selected from Experiment 1.

of polishing and 666 hours of NAWS weathering, and (4) after 150,000 cycles of polishing and 1,000 hours of NAWS weathering.

In addition, asphalt binders were extracted and recovered from the control and dry-process RPM mixtures, which were fingerprinted using the FTIR to detect the presence of plastics. The extracted RPM binders showed no presence of PCR plastics; therefore, they were excluded from further rheological and chemical characterization. This finding is discussed in more detail in the next chapter.