As part of Transport Infrastructure Ireland’s (TII) Statement of Strategy 2021 to 2025 and for the drive for innovation in the asphalt industry Kilsaran and Irish Tar proposed to develop a binder and bituminous mixture which allows Reclaimed Asphalt Pavement (RAP) mixtures to be produced at lower temperatures in the cold addition process, write Dimitris Michailidis and David Hogan.

Alongside the environmental benefits of such an approach, it is also hoped that reduced ageing of the material through the mixing process will result in improved performance. The overall goal is to maintain material performance while driving down emissions and a reduction in the use of virgin materials (aggregates and bitumen).

The incorporation of RAP into Warm Mix Asphalt (WMA) is highly interesting, both environmentally and economically, and can contribute to promoting the circular economy.

The incorporation of RAP into WMA can provide the positive effects of both techniques – improved working conditions, energy and cost savings, reduction of emissions, and lower consumption of fuel, bitumen, and aggregates and their associated production impacts. 

Sustainable pavements 

The global situation regarding climate change makes it necessary to promote the circular economy and the use of more environmentally friendly technologies in the construction sector.

It is estimated that more than 90% of the 5.2 million kilometres of European paved roads and highways are surfaced with asphalt. Also, about 44% of goods are transported by road in the EU; maintaining their condition while in transit is crucial for the economy.

Although new technologies may have an initial cost increase, the long-term benefit assessment may justify it. Various aspects of asphalt mixture performance are the influential factors on life-cycle analysis (LCA) which determine the environmental-economical profitability. Better performance results in longer lifespan as well as higher serviceability, which, in turn, reduces reconstruction, maintenance, and rehabilitation costs.

Additionally, less natural resources, including fuel, aggregate and bitumen, are consumed in the long term. Therefore, the optimal balance between engineering (performances), environmental, and economic aspects brings sustainability to pavement assets (Figure 1). 

Figure 1: Sustainable Pavements

Reduction of atmospheric emissions 

WMA releases less heat and emits less pollution during its production and application and greatly diminishes the environmental degradation associated with Hot Mix Asphalt (HMA). The incorporation of a warm mixing agent also slows down the ageing process of the asphalt mixture.

The improved environmental impact of mixtures containing RAP and mixtures containing warm mix additives are widely covered in available research. Combining both in this project will result in maximising carbon savings and minimising raw material use while maintaining material properties.

During production of the trial, environmental measures were monitored including fuel usage at the plant and emissions. 

RAP historical data 

Recovered RAP materials need to be stored and properly stockpiled in dedicated area(s) before being subjected to further processing. Thus, stockpiles of traceable RAP are formed (and duly identified). The following information is recorded:

  1. Source (ie, layer type, pavement age).
  2. Date of receipt.
  3. Milling and extraction details.
  4. Processing history.
  5. Stockpile location.

The mix used was an SMA 14 mm with a PSV value of 60 and the pavement age was approximately 16 years old (construction of this section took place in 2005). In order to produce high quality material, we need to determine the aggregate properties of our reclaimed asphalt pavement.

In general, RAP should be also characterised for the mechanical properties of the aggregates and not only on the binder content and grading. For use in surface course mixtures, the source of RAP needs to be consistent and contain high PSV aggregates that are recoverable.

If RAP was obtained from the surface course with aggregate properties comparable to those required for the new recycled mixture, then it is assumed that there is no deterioration in the properties of aggregate within the RAP. The RAP will be considered suitable for use.

The mix design for high RAP contents should consider the properties of the RAP aggregate. At low RAP percentages, the effects may be minimal. When the aged binder from RAP is combined with a new binder, it will have some effect on the resultant binder grade. At low RAP percentages, the change in binder grade is small.

Stockpiles shall be identified through a documented tagging system and their status shall be clearly marked and made visible and understood by all responsible personnel. In order to be able to characterise our RAP properties and form homogeneous stockpiles, and, since PSV is a crucial characteristic, we decided, even though it is not described by the specs, to carry out a PSV test on the aggregates present in the RAP planings.

Figure 2: PSV value of Reclaimed Aggregate

N80 Ballymacken project: introduction 

In April 2021, Kilsaran in collaboration with Irish Tar, TII and Laois County Council, and under the supervision of the latter, proceeded in the application of the innovative mixtures on the N80 Ballymacken pavement repair and renewal project.

This particular site was selected because there was full rehabilitation of all existing layers and therefore our base, binder and surface mixtures could be implemented at the same time. In total, nine mixtures were prepared and laid (Table 1).

Table 1: Asphalt Mixes

HMA AC 20 & AC32 with 25% RAP 

For the first section of the project, from chainage 0+036 to 0+420 in the outbound direction, our HMA base and binder layers with 25% RAP were implemented. Both these mixtures were produced at approximately 1650C.

In the inbound direction at the same chainages, we used the same mixtures but, in this occasion, they were produced with 0% RAP. The main reason for adopting this laying option was to be able to have comparative data and be able to monitor the behaviour of these mixes under the same traffic loads.

Overall, the material was laid and compacted following the same procedures as those normally applied for a HMA with no RAP. The only comment from the site was the extensive fumes and steam generated during the operation. However, these are expected when laying at the above-mentioned temperatures.

Photo 1: Laying AC32 with 25% RAP

Photo 2: Fumes During Compaction

WMA AC 20 & AC32 with 25% RAP 

For the following section from chainage 0+450 to 1+028 in the outbound direction, our WMA base and binder layers with 25% RAP were implemented. Both these mixes were produced at a temperature of 1200C.

In the inbound direction at the same chainages, we used the same mixtures but, in this occasion, they were produced with 0% RAP. Our on-site feedback indicated that in terms of laying and compaction there weren’t any significant differences from the procedures used for HMA.

The one major difference was in terms on fumes and steam, that were eliminated during the application of the WMA. This project was located in a residential area, and the lack of fumes made the site a lot easier for people commuting in the morning.

Photo 3: Laying & Compacting WMA

WMA SMA 10mm with 15% RAP & HMA SMA 10mm with 15% 

For the final two sections of the project our surface layer material was implemented. In the first section from chainage 1+136 to 1+300 in both directions we used our WMA SMA 10mm with 15% RAP and in the final section from chainage 1+375 to 1+650 we used our HMA SMA 10mm with 15% RAP. Both materials were laid and compacted following the same procedures as those adopted for a HMA with no RAP.

Photo 4: Laying of WMA SMA 10mm

Production temperatures 

Warm Mix Asphalt is produced and applied at a temperature of about 20-40 °C, lower than an equivalent 'Hot Mix Asphalt' (HMA). However, this figure depends on a number of factors such as moisture content of the RAP planings, the type of warm mix additive etc. During our production we were able to achieve a 400C drop in temperature with no resultant workability or compaction issues.

Figure 3: WMA vs HMA Production Temperatures (AC32 & AC20)

Figure 4: WMA vs HMA Production Temperatures (SMA 10 mm)

Plant emissions 

The majority of emissions at asphalt mixing facilities come from the combustion of fuel that is used to dry and heat the aggregate and to keep the temperature of the asphalt high.

The most commonly emitted gases are carbon dioxide (CO2), carbon monoxide (CO), nitrous oxides (NOX), and sulfur dioxide (SO2). Most of the other potential emissions, such as the dust generated during the drying of aggregate, are captured by baghouse filters or similar controls and never released to the environment.

HMA emits higher amounts of pollutant gases during both production and construction because of the high temperatures involved in these processes.

WMA releases less heat and emits less pollution during its production and application and greatly diminishes the environmental degradation associated with HMA. The incorporation of a warm mixing agent also slows down the ageing process of asphalt mixture. 

The main purpose for implementing Warm Mix Asphalt is to reduce our emissions and our overall carbon footprint. Therefore, in collaboration with Odour Ireland we measured our emissions during normal production and then repeated the same measurements during the production of our warm mix SMA.

The analysis of the reading showed a significant reduction of our CO2 emissions by 9.5%. This figure was lower than anticipated as studies showed that the reduction can be up to 50% and an acceptable value is about 20%.

As a rule of thumb, the release of fume is reduced by about 50% for each 12 °C reduction in temperature. So, a temperature reduction of 25 °C will lead to fume emission reduction of about 75% (EAPA).

Figure 5: Temperature Reduction & Fume Emission reduction

However, it is positive that even though Warm Mix Asphalt was only produced for a few hours, there was a reduction of almost 10% (Table 3). We strongly believe that through continuous production of WMA a significantly higher reduction of our CO2 emissions is possible.

Table 3: Reduction of Plant Emissions

Laboratory testing 

Upon completion of construction a series of cores were extracted and tested in order to determine the properties of our mixtures.

Stiffness results (ITSM & 4PB) 

The modulus (stiffness) of the various HMA or WMA types used for the construction of the bituminous layers of the test sites are estimated using the indirect tensile stiffness modulus (ITSM) test.

Table 4: Stiffness (ITSM) – average values [MPa]

From the ITSM laboratory results, the following conclusions can be drawn:

  • The AC32 stiffness moduli values are generally higher than the related of the AC20;
  • Warm-RAP mixtures exhibit lower values than the Control/Hot-RAP ones;
  • Control and Hot-RAP mixtures exhibit approximately equal moduli values.

The modulus (stiffness) values of the various HMA or WMA types used for the construction of the bituminous layers of the test sites are considered also during the four-point bending test (4PB), referred to as 'initial stiffness'.

The specimens used for 4PB testing are produced in the laboratory, using similar material and mix design data, with the ones used for the construction of the various test sites.

Table 5: Initial stiffness (4PB) – average values [MPa]

Figure 6: Initial stiffness (4PB)

From the above results, the following conclusions can be drawn.

  • The AC32 stiffness moduli values are generally higher than the related of the AC20;
  • Warm-RAP mixtures stiffnesses exhibit lower values than Hot-RAP ones;
  • 4PB results confirm the ITSM results.

Fatigue testing 

While not yet a standard test in Irish pavement design standards, it was decided to carry out fatigue testing on the proposed materials to compare their performance with the control materials.

Conventionally, bending tests are supposed to represent the repeated bending forces caused in the pavement by the passage of vehicles, while tensile tests represent the tensile force induced at the base of the pavement by this bending. Simple flexure and diametral tests are the most common testing procedures used by testing laboratories for routine investigations.

Fatigue characteristics (ITFT) 

Fatigue testing was carried out in accordance with IS EN 12697-24 Annex D and Annex E. Table 1 shows results from laboratory trials tested in accordance with IS EN 12697 Annex E.

The fatigue characteristics of the various HMA or WMA types are estimated with the indirect tensile fatigue test (ITFT). The test is conducted on cores from the constructed test pavements. The results (fatigue curves at temperature of about 20oC) are presented graphically in Figures 10 and 11 for AC32 and AC20 respectively.

Figure 7: Fatigue curves (ITFT) AC32

Figure 8: Fatigue curves (ITFT) AC20

From the above test results, the following conclusions can be drawn:

  • AC32 fatigue curves show better fatigue performance for both Warm-RAP and Hot-RAP mixtures, against the Control mix. However, further investigation is needed due to the relatively small number of tested specimens;
  • AC20 mixtures have shown different fatigue behaviour, with better performance for the Control mix.

Fatigue characteristics (4PB) 

The fatigue characteristics of the various bituminous mixture types are also estimated with the four-point bending test (4PB). The specimens used for 4PB testing are produced in the laboratory, using similar material and mix design data, with the ones used for the construction of the various test sites.

The results, ie, number of cycles at 200 μstrain (average values) at a temperature of about 20oC, are presented graphically in Figure 12. It must be noted, that for the SMA, the number of cycles is estimated at 400 μstrain, due to extremely high number of cycles at 200 μstrain amplitude.

Figure 9: Number of cycles to failure (4PB: H-RAP & W-RAP at 200 μstrain, SMA at 400 μstrain)

The number of cycles at 200 μstrain indicates better performance of the Hot-RAP mix than the Warm-RAP.

The SMA mixture exhibits much better performance than the AC32 and AC20 mixtures. Although this is an indication of the robustness of the SMA using a polymer modified bitumen.

Falling weight deflectometer analysis – in-situ collected data (deflections) 

The FWD device used was a Dynatest 8002-452. For deflection measurements, seven sensors were utilized with distances from the centre of the loading plate: 0, 300, 600, 900, 1,200, 1,500 and 1,800 mm. The measured pavement surface temperature was in the magnitude of 19oC, 21.7oC and 20.6oC, for the test sites with codes: 207, 208 and 209 respectively.

The deflection indicators used for the characterization of the structural condition of the pavement layers are the following:

  • The Central (maximum) deflection D0 for the characterization of the condition of the whole pavement;
  • The Surface Curvature Index (SCI): (SCI = D0 – D300), for the characterization of the condition of the pavement bituminous layers;
  • The Base Damage Index (BDI): (BDI = D300 – D600), for the characterization of the condition of the unbound base layers;
  • Considering the results of the deflection analysis (deflection indicators), the following conclusions can be drawn;
  • ✓ The structural condition of the bituminous layers can be characterised as 'very strong' without any significant variations along the constructed test sites;
  • ✓ The overall condition of the constructed test sites can be characterised as 'strong'. The variability can be attributed mainly to the differences of the structural condition of the unbound layers and/or the subgrade.

Conclusions 

The aim of this project was to try and compare the performance of WMA and HMA (with 0% and a percentage of RAP) in terms of in-situ densities (compaction), stiffness, fatigue, ageing effects, and overall durability.

Also, our aim was also to investigate how these mixes are laid and compacted in comparison to traditional HMA. Furthermore, observations were made regarding failure mechanisms of our HMA and WMA which did not necessarily reflect in our results.

Regarding production and site operations, implementing a WMA mix showed no issues with mixing, laying or compaction, with the physical properties of the WMA being equivalent to HMA in all respects. It is possible to meet the required volumetric properties when laying and compacting WMA using standard paving machinery and rolling patterns.

The paving-related benefits of using WMA mixtures include: 

  • The ability to pave in cooler temperatures and still obtain density. In addition, the ability to pave and compact asphalt mixes at cooler temperatures could allow an extended paving season;
  • The ability to have the workability to lay and compact the mixture after longer haul distances;
  • Reduced effort needed to compact the mixture which will lead to increased paving speeds and greater in-place densities;
  • The ability to incorporate higher proportions of RAP at reduced temperatures.

With lower mixing temperatures, the temperature of the asphalt at the end of compaction should be lower with WMA hence, closer to the service temperature at which it can take traffic without damage.

We noted that the failure mode of our HMA with 25% RAP was significantly more catastrophic in comparison to our WMA with the same amount of RAP. The failure pattern of those samples appears to be more brittle, and we believe, the reason for this is the absence of rejuvenator in those mixtures (AC20 with 25% RAP).

Photo 5: Failure comparison HMA vs WMA

Considering the laboratory test results, the following conclusions can be drawn.

As far as the stiffness of the various bituminous mixtures is concerned, the Control and the Hot-RAP mix exhibit approximately equal moduli values, while the moduli values of the Warm-RAP are lower.

In some cases, the results of the various fatigue tests are ambiguous. For this reason, no clear conclusions can be drawn, considering the fatigue characteristics/ranking of the various bituminous mixtures. This may be, due to the relatively small number of tested specimens, which creates the need for further investigation.

4PBT was performed at one strain level in terms of four specimens. Perhaps the response could have been different if there were additional strain levels, including an increase of the number of test specimens. 

Authors: Dimitris Michailidis is technical manager (road surfacing) with Kilsaran Concrete and has more than 17 years’ experience in large-scale infrastructure projects holding a technical QA/QC and construction management role. David Hogan is a Chartered Engineer and is technical manager with Irish Tar & Bitumen Suppliers.He joined Irish Tar on completion of his degree in structural engineering at DIT Bolton Street and has been with the company for more than 15 years.