Skip to main content
European Commission logo
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary
Contenido archivado el 2024-06-18

Re-road – End of life strategies of asphalt pavements

Final Report Summary - RE-ROAD (End of life strategies of asphalt pavements)

Executive summary:

Europe has more than 30 years of experience with reusing reclaimed asphalt in the production of new asphalt. Nonetheless, there are still big possibilities of optimising the process since at least one third of reclaimed asphalt end up in non-asphalt concrete applications. The focus of this project has been to look at and address all major aspects which hinder asphalt to be recycled to new asphalt concrete.

What is the environmental benefit of recycling asphalt in the first place' What additional benefit could be gained if we use a reclaimed surface to the same level in the pavement construction'

A life cycle assessment was the most appropriate way to establish the necessary framework to answer these questions on a transparent basis. The results of the LCA proved that, above all, recycling to a bound course was significantly more environmentally advantageous than recycling to an unbound course. Appreciable extra benefits can be realised if high specification aggregates are preserved in their original application by surface-to-surface course recycling. This, because the quarries that produce these specialised aggregates being widely spaced and hence requiring large transport distances for the aggregates. The moisture content that is sometimes present in reclaimed asphalt only mildly counteracts the recycling benefits.

The results also indicate that low level recycling, just 15 % to bound courses, is significantly more environmentally beneficial than warm mixing - a set of techniques for lowering the production and laying temperatures of asphalt concrete - particularly if the additives used to facilitate warm mixing are included in the analysis.

The project also addressed the performance of pavements made with reclaimed asphalt. A follow-up study of sites where significant proportion of reclaimed asphalt has been incorporated in new asphalt surface course layers was one approach used. For example, over 300 km of roads with so called "in situ" recycling in Sweden showed that there were no indications of a high recycling ratio having an adverse effect on the service performance compared with control sections. In situ recycling also proved to be effective in increasing the ride quality of the pavement and above all in situ recycling could be used multiple times without affecting the asphalt performance.

The impact of multiple recycling on the performance was also one of the topics of the laboratory studies. The performance will depend on the selection of the new binder which is added to the mix. A softer binder will be needed in order to obtain the right consistency after blending with the hard aged binder. The new and old binders also have to be compatible. The multiple recycling study showed no negative impact of multiple recycling on the performance characteristics of the mixtures or of extracted binders.

Project Context and Objectives:
The vast majority, 90 %, of European roads are paved with asphalt material. At the end of the service lifetime of a road, when the damaged pavement cannot further fulfill its purpose as a comfortable carrier of traffic, the road pavement must be renewed. Sustainable construction processes that conserve natural resources are well recognized within the asphalt industry although practices for asphalt recycling vary to a great extent across Europe. Today, a large amount of demolished asphalt pavement ends up as unbound granular layers where neither the bituminous binders nor special aggregates from old surface layers are reused at their full potential. Replacing new materials with recycled asphalt in the production of new asphalt reduces CO2 emissions significantly.

The Re-Road project aimed to develop knowledge and innovative technologies for enhanced end of life strategies for asphalt road infrastructures. Such a strategy has an important impact on the energy efficiency and the environmental footprint of the European transport system and fits within the lifecycle thinking which is being introduced in the waste policy at a European level.

The project covered the following topics that are important for the determination of an end of life strategy:
- Dismantling strategies: The impact and potentially adverse effect of different dismantling procedures on the quality of reclaimed asphalt (RA).
- Characterization strategies: Improved characterization of RA and technical evaluation of RA as a raw material requires considering both the heterogeneity of the material and the specific industrial processes for producing the asphalt mix.
- Handling strategies: Optimization of recycling to the highest possible level and for the original layer requires procedures for proper handling of the material and procedures for processing the data from the characterization done to have an environmentally sound reuse of the material, or disposal of marginal materials that cannot be recycled.
- Environmental criteria: Assessment of risks and benefits to the environment with the use of RA. Potentially harmful substances (like tar-containing asphalt) and life cycle analysis (LCA).
- Cost-effective recycling: Short and long term performance, life time prediction by modelling of asphalt mixes produced with different levels of RA and with different production techniques.
- Industrial processes: The potentially adverse effect on the final asphalt mix quality derived from the specific method for introducing the RA in the mixing plant. How to avoid problems in the re-cycling of polymer modified RA and how to take full advantage of their special qualities.

Project Results:
1 Sampling and Characterization
One of the objectives of the Re-Road project was to optimize the use of recycled asphalt pavement in wearing courses. The aim was to use higher amounts of Reclaimed Asphalts (RA) in mixes in comparison with current practices.

RA is complex materials and the use of significant proportions of RA involves a more critical management of heterogeneity and the characteristics of these materials. WP1 focused on issues that are specifically related to the characterization and technical evaluation of RA as a raw material, prior to the recycling processes. The main goal was to better assess the characteristics of RA that will allow an increase of the percentage of RA in new asphalt mixtures while ensuring final material performances.

The reports (deliverables) are denoted with a D followed by two numbers. The first representing the work package wherein it was produced. The reports could be downloaded from http://re-road.fehrl.org/

1.1 Results obtained in the task group for sampling procedure for RA
The first task of WP1 was dedicated to the development of a suitable sampling procedure, which considers the heterogeneity of materials that could be important sources of uncertainty in the assessment of physical, chemical and environmental properties of RA or materials including RA.

The main conclusions of the investigations are:
-Sampling by means of a spade in a roof shaped stockpile, according to the standard EN 932-1, seems to be an optimum way to make a representative sample. It can be noticed here that locations where RA is sampled by means of a shovel, are of high importance for obtaining a representative sample. This representativeness of a global sample is directly correlated to sample locations.

-Applying mixing laws for characterizing a mixture of RAs, even for RA containing PmB, seems to be reasonable, especially in terms of saving time. However, as is often the case, this study needs more tests in order to extrapolate and generalize.

1.2 Results obtained in the task group for improvement of the characterization of RA, especially RA containing polymer modified bitumen

RA are complex materials and the use in significant proportions in new asphalts involves a more accurate control of their characteristics are essential for asphalt mix design and key factors for good performance of new asphalt mixtures. Consequently, it is very important to focus on issues that are specifically related to the characterization and technical evaluation of RA and particularly RA containing modified binders. There is also clearly a lack of knowledge and adequate test methods to sample and analyse them (as shown in the state of the art presented in D1.1).

One of the topical problems is related to the determination of the binder content. The current European test methods for extraction and recovery of binders in RA are only normative for RA with pure binders and only give indicative informative for RA containing PmB. A key point for RA containing PmB is to know if the used recovery procedure permits to fully extract PmB while modifying the properties of PmB as little as possible. In this study, it has been tried to identify the impact of extraction procedure on the PmB properties. Binder extraction and recovery are essential steps to determine the basic characteristics of RA. The impact of the couple testing method/solvent on the binder content and binder characteristics has been assessed for RA with PmB in the frame of the European test methods EN 12697-1 and EN 12697-3.

Six laboratories participated in the round robin test which is presented in D1.2. According to the results obtained in this study, it seems that the couple testing method/solvent has no impact on the determination of the soluble binder content of a new asphalt mixture with modified bitumen and including a low content of RA. However, problems appear for RA containing aged bitumen with PmB, for which the ageing level, combined with the presence of polymers lead to a more difficult recovery. At the moment, the European standards EN 12697-1 and EN 12697-3 do not appear to be clear enough when asphalt contains PmB. The same trends can be observed in the characteristics (namely penetration, softening point, oxidation degree, polymer content, complex modulus, and ductility force) of recovered PmB, the values being more scattered for RA than for a new mix, whatever the measured properties.

The next study, described in D1.4 consisted of laboratory production of artificial RA with defined binder content and known binder properties. Then, the different recovery methods were applied according to the standards EN 12697-1 and EN 12697-3. As these methods describe a large range of methods and solvents that can be chosen to carry out the tests, the different results have been compared to the "true" ones in order to evaluate the adequacy of the methods and solvents proposed in standards to characterize RA with PmB.

The following conclusions have been drawn on all the analyses presented in D1.4:
- For the determination of soluble binder content, the testing method and the solvent have no impact on a new asphalt mixture with modified bitumen and including a low content of RA. However, problems appear when characterizing RA with Polymer modified Bitumen. More scattered results are obtained for RA with physical SBS modified bitumen, but the largest span is obtained for RA with chemically linked SBS modified bitumen. From a technical point of view, the recovery of the binder from an RA with PmB is more difficult than for a new asphalt with PmB, and the choice of the testing method and solvent seems linked with the type of binder (physically bond or cross-linked PmB): the impact is more pronounced for RA with a chemically linked SBS modified bitumen. That may be due to some solubility problems linked to the nature and temperature of used solvent.
- Two characteristics have been identified as strongly related to the assessment of the end of life of RA that are:
- the content of carbonyls (linked to the oxidative ageing)
- the complex modulus at 25°C and 52°C (either 1.6 Hz or 10 Hz) (linked to the hardening of binder).

Checking an RA grading curve is usually performed through sieving. However, this method is time consuming; hence it may not be appropriate to the production rate of mixing plants. To overcome this problem, the use of the VDG40 videograder was suggested since this device is able to rapidly determine the grading curve of aggregates. As it is an alternative optoelectronic device to the classical sieving method, comparative tests with traditional testing means have been performed on different kinds of RA in order to validate the use of this device for such materials.

This study, described more in detail in D1.5 shows that the videograder may be used for quick checking of the homogeneity of RA grading at the mixing plant. Despite the videograder not being appropriate to deliver the actual sieve grading curve of the RA aggregates, especially not the fines content, this apparatus presents a good repeatability in terms of measurement. It can also be used to detect big lumps that could be problematic for the reuse of the RA in a new hot bituminous mix.

1.3 Results obtained in the task group environmental characterization

To improve the environmental characterization of RA there is a need for laboratory test methods and test scenarios that can support practitioners in assessing emissions due to release of particles, vapours/fumes and leachates.

These methods include:
- Identification of potentially hazardous organic compounds susceptible to leaching from RA.
- Evaluation of methods suitable for assessing the potential for leaching of hazardous compounds from RA.
- Further characterisation of leaching, through appropriate ecotoxicity testing.
- Evaluation of a test method for assessment of the bioavailability of organic constituents associated with RA.
- Evaluation of an appropriate test method for airborne emissions arising in mixing at manufacturing temperatures.

The number of candidate contaminants that can be analysed through environmental testing is vast, as seen in the state of the art report D1.1. Effective characterization of RA requires that potentially hazardous compounds, susceptible to leaching from RA, are identified:
- For the inorganic compounds it was decided to analyse leachates for the following elements: barium (Ba), cadmium (Cd), cobalt (Co), chromium (Cr), cupper (Cu), manganese (Mn), magnesium (Mg), nickel (Ni), lead (Pb), antimony (Sb), vanadium (V) and zinc (Zn).
- For the organic substances, polycyclic aromatic hydrocarbons (PAHs) and benzothiazole (if the tested material contains rubber-asphalt) were identified as the most important organic compounds to collect data for. Alkanes, adipates and phthalates were rejected due to a combination of low concentrations and low toxicity levels.

The release of metals from the tested RA materials to water occurred in relatively small quantities, and levels of pollution among leachate were generally low. Differences between the material leachate (except leachate from the highly contaminated reference material) were small and the total PAH concentration in these leachate samples was nearly similar. Ecotoxicity tests can, however, bring additional important information. It has been demonstrated in this study that material leachate with the lowest concentrations of analysed metals and PAH may not represent the lowest risk for the water ecosystem. A toxic response obtained in tests on leachate with low concentrations of metals and PAH may be due to the prevalence of "unknown" substances (i.e. substances that were not analysed; other PAHs than the 16 common PAHs, degradation products of PAHs, other organic compounds with origin from the binder, rejuvenators, traffic etc.). It can also be a result of a synergy effect caused by two or more substances that alone exists in harmless concentrations but together cause a toxic response in the tested system. Therefore, this study has demonstrated the benefits of ecotoxicity tests for the evaluation of possible threats posed by RA to fresh water ecosystems. It is acknowledged that there is some uncertainty with the test results, but it should be noted that ecotoxicity testing gives information that is not available when using chemical tests alone.

Finally, in laboratory the airborne emissions arising when mixing at manufacturing temperature have been assessed thanks to a model experiment using a new fume generation system.

View of IFSTTAR fume-generation system (asphalt mixer equipped with a heated stack).

2 Laboratory Performance, Mix Design, Field Validation
2.1 Introduction
The objectives of Work Package 2 "WP 2: Impact of RA quality and characteristics on mix design and performance of asphalt containing RA" was to analyse the potential use of RA in new asphalt surface layer mixes considering the use of modified binders. Therefore, the chemical compatibility of new binders with old (polymer modified) bitumen in RA and the physical and mechanical performance of the resulting binder and asphalt mixes were examined in laboratory. Whereas case studies were evaluated further for practice validation in order to develop mix-design guidelines to ensure a long service life of asphalt mixes with reclaimed asphalt.

2.2 Results obtained in the task group for RA compatibility
In addition to traffic loading, asphalt surface courses are subjected to long-term aging due to the direct exposure to air (oxygen) and sunlight (UV-radiation).

The short-term aging properties of binders are commonly investigated e. g. by means of Rolling Thin Film Oven Test, according to EN 12607-1, and more seldom by long-term aging by means of Pressure Aging Vessel, according to EN 14769. The evaluation of aging properties of asphalt mixture is still uncommon. Despite the fact that the aging of the asphalt mixture is mainly influenced by the aging properties of the asphalt binder itself and the void content of the compacted asphalt layer, the aggregates may influence the aging effects. Therefore, a procedure for analysing the effect of aging on asphalt properties is needed, taking the whole asphalt mixture into account.

When RA is reused in new hot-mix asphalt, the binder of the RA (usually highly aged) is melted during RA heating and/or mixing with virgin aggregates and new binders. The properties of the resulting mixture are highly dependent on the mixed binder properties. Therefore, the compatibility between the aged RA binder and the virgin binder is of great interest for the evaluation of suitability of RA addition to new mixtures. In order to evaluate the binder compatibility, physical (mechanical) as well as chemical test procedures were applied on various binders extracted from mixtures containing high percentages of RA.

As the bituminous binder is the constituent material dominating the particular physical properties of the resulting asphalt mixture, the work being done concentrates on the binder-related topics:
- Optimisation of a laboratory aging method to analyse the recyclability of asphalt mixtures. This method can also be used to prepare RA with controlled constituent materials for other project tasks.
- Development of procedures to evaluate chemical compatibility of the aged (polymer modified) binder with new binders and other additives (e.g. rejuvenators) added during the recycling process as well as the application on different modified RA samples.
- Determination of the end of life of an asphalt mixture after several recycling cycles.

The following observations could be made when optimising the aging procedures and comparing the aging effects on binder and asphalt mix properties:
- Aging of loose asphalt mixture in an oven is a suitable method in order to simulate long-term aging of asphalt mixture.
- Additional UV radiation tends to show that it only leads to a slight increase in aging (shown for aging temperature of 60 °C) of an asphalt mixture containing unmodified binder. No effect was observed for the asphalt mixture containing SBS modified binder.

The results of a study (D2.3) set to analyse the precision of the applied aging procedures indicate that the laboratory aging of asphalt mixtures results in feasible repeatability and reproducibility levels. The results presented also indicate that the relative effect of aging in the applied methods is different from one mixture to another.

2.3 Results obtained in the task group for impact of RA on mix design and laboratory performance

A study was made to investigate the effect of recycling a mixture containing PmB more than once. Important issues like compatibility of the new PmB with the old PmB and the validity of mixing laws for predicting the properties of a mixture with up to 50 % RA were also addressed in this study. The main conclusions of the multiple recycling studies were:
- The multiple recycling of 50 % RA in new asphalt mixes, in 3 recycling cycles with optimal laboratory conditions, results in asphalt performance comparable to the properties of SMA mixes containing 100 % virgin material. This conclusion is valid for the laboratory conditions applied, where the influence of different grading in RA and new mix was not considered, and "pure" simulated RA without any other materials (e.g. other layers, rehabilitation patches) was added to the new mixes. In terms of binder reactivation, the laboratory-ageing, simulating long-term site ageing won't affect the recyclability of surface course SMA if virgin binder of reduced viscosity is added to the mixes.
- By using a polymer modified binder of lower viscosity as virgin binder, the increase in viscosity due to long-term aging effects can be balanced. According to the binder properties analysed, the mixtures composed of material which has already gone through several recycling cycles still shows suitable performance. No lifetime reduction of the mixture containing 50 % of simulated RA should be expected.

Based on the results obtained, a five-stage mix design procedure is proposed. Details of the proposed mix design procedure could be found in D2.7.

2.4 Results obtained in the task group for field validation
Several existing test sites were monitored in Denmark, Sweden and the UK in order to evaluate the medium-term performance of the asphalt surface course variations applied. Besides the test sites, on which plant mixed asphalt surface mixtures with considerable percentages of reclaimed asphalt, were applied, also hot in-place remixed surface courses were monitored. Monitoring schedule and results are discussed in detail in D2.6.

View of the Renishaw site in June 2012 after 10 years' service

From the monitoring undertaken in Denmark, Sweden and the UK on various in situ and plant mixed surface course recycling sites, the following conclusions, based on comparative performance between control and recycled sections, were drawn.

Plant mixed recycling
The performance of plant mixed recycling with a proportion of RA, added to a new surface course has been shown to provide comparable service for the medium to long term. With additions of RA of between 23 – 30 %, this has been demonstrated in pilot scale trials for over 10 years, road trials for over 8 years and in re-surfacing schemes after nearly 6 years' service.

In situ surface course recycling
There have been improvements in the operational process for in situ recycling since the late 1990's and this has led to a more consistent end product.

Studies of over 300 km of roads in Sweden showed that there were no clear indications of a high recycling ratio having an adverse effect on in service performance compared with control sections.

Recycling in situ has proved to be effective in reducing the longitudinal unevenness (expressed by International Roughness Index IRI) and hence increasing the ride quality of the pavement.

The vast majority of remixed sections gave comparable performance despite high recycling ratios. Furthermore, in situ recycling could be applied multiple times without adversely affecting the asphalt performance.

3 Environmental Performance

3.1 Introduction
The main objective of Work Package 3 research program was to develop and improve tools that can be used to assess and characterize the environmental performance of the use of RA and to identify and assess environmental benefits and burdens rising from the different stages of an asphalt pavement's life cycle.

To allow the Re-road partners to conduct a risk assessment and life cycle analysis of pavements incorporating RA (relevant to all European partners) it was first necessary to obtain an accurate overview of the road construction practices employed across Europe.

In D3.3 and D3.4 the modelling work is presented in full details together with information on the data collected from literature, lab- and field experiments. This report summarises the main findings of these two assessments and provides an integrated evaluation of the results. In D3.5 the findings in the previous deliverables are summarised and an integrated evaluation is provided.

The asphalt pavement's lifecycle can be divided into the following four stages:
i) asphalt production from new raw materials;
ii) paving and in service use (including maintenance);
iii) dismantling of the asphalt and
iv) production of RA-material (crushing, sieving etc.).

3.2 Results obtained in the task group for risk assessment of RA

There is always an element of risk to the environment associated with road construction work. This is an inevitable result of the construction process, but it is important to ensure that the level of risk remains within an acceptable range. The use of RA within asphalt surface layers presents many advantages with respect to material use and sustainability. However, there exists a perception that by using RA in surface courses, there is an increased risk of environmental damage.

This raises a number of questions, including:
- Is there an increased risk of contaminants leaching by having the recycled materials closer to the surface (and infiltrating water)'
- Does the recycling process, including stockpiling of the materials, present any environmental hazards'
- What are the most likely pathways for contaminants that can lead to health issues (airborne, waterborne, other)'

Effect on groundwater quality
There are considered to be three potential sources of contamination associated with reclaimed asphalt:
- Contaminants in the bitumen, additives and aggregates;
- Contaminants which build up on the surface of the road during its previous pavement life/lives;
- Contaminants which build up during the life of the newly laid pavement containing RA.

These contaminants may then be available for leaching into road runoff or water infiltrating the road surface or they can be blown from the surface by the wind.

In assessing the potential effect on groundwater, a leaching model was developed that incorporates data from the leaching tests reported in D1.6.

The leaching simulations were conducted for a range of pavement types. These included:
- Pavement constructed using stone mastic asphalt (SMA) with 0% RA (Ref Mix 1);
- Pavement constructed using SMA with 15% RA (Ref Mix 2);
- Pavement constructed using SMA with 30% RA (Ref Mix 3);
- Pavement constructed using a mixture with 50% RA and a rejuvenator (Storbit);
- Pavement constructed using a tar containing RA material (TCA);
- Two pavements constructed using material from mixed source stockpiles (MSS and Stockpile).
Further details on these pavement constructions and materials can be found in D3.3.

Sample leaching behaviour – fluoranthene leaching from various pavement surfaces after 1 year's rainfall. Left – frequency distribution of quantity leached per square meter of road area. Right – cumulative distribution of quantity leached per square meter of road area.

Based on these distributions a number of observations can be made:
- For the reference mixes with 0%, 15% and 30% RA, the quantities of fluoranthene leached are far below what is observed for the TCA and MSS materials;
- There is no clear difference between any of the reference mixtures, suggesting that there is no increased risk associated with using the material in surface courses;

Similar trends were observed for all PAHs. The highest concentration of individual PAHs was always associated with either the MSS or TCA material or reaching levels associated with the reference mixtures, always significantly below these values.

Based on the values a number of comparisons can be made both with groundwater limits and contamination levels associated with regular traffic effects.

Some conclusions include:
- The levels of leaching associated with the reference mixtures are significantly below the guideline values. These materials will not exceed the guideline values under any circumstances.
- The highest leaching levels were associated with the tar containing RA, supporting that restrictions are required on its use in road construction.

Effect on air quality
The health hazard related to airborne bitumen fume generation is primarily relevant for paving crews while there is little opportunity for exposure related to asphalt plant workers. Exposures to asphalt fume have declined since the 1970's and is normally below occupational threshold limits. Recycling of RA and introduction of additives can potentially create new exposures and health concerns.

The complex chemical composition of asphalt fumes and the high variability in exposure makes the occupational assessment of bitumen fume susceptible to large variability, including:

- Climate, (wind speed and direction, temperature etc.)
- Work tasks (paving, screeding, raking, rolling etc.)
- Ambient environment
- Asphalt application (type, source, temperature, engineering controls etc.)
- Sampling device (type, rate, duration etc.)

This high variability in occupational exposure can mask significant changes in emissions and makes it very difficult to relate quality and quantity of bitumen fumes to the quality of RA, binder and binder additives. In order to correctly assess the influence of these parameters, laboratory test methods performed under controlled conditions can be of great value. Therefore, such experiments were conducted, using a laboratory bituminous mixtures fume generator presented in D1.6 making it possible to isolate and identify the impact of RA and asphalt binders on both quality and quantity of emitted bitumen fumes. By making use of a sequential mixing protocol it was further possible to relate these emissions to specific parts of the RA recycling cycle e.g. paving or asphalt plant.

Adding uncontaminated RA in the hot mix asphalt process did not increase the total emissions or the quality of bitumen fumes. Data on emissions from mixes containing tar contaminated RA showed that these types of mixes were more emissive, especially for PAH, in agreement with earlier findings, and were also more emissive for benzene soluble matter. The effects were pronounced even for mixes with a rather limited contamination of PAH at approximately 300 mg/kg (PAH-16).

3.3 Results obtained in the task group for life cycle assessment of RA
The life cycle assessment (LCA) study in D3.4 focused on the environmental basis for recycling asphalt and how it measured up against other environmentally-focused initiatives in the asphalt industry. Also within the scope of the study, it was possible to investigate specific parts of the life cycle that could be optimised in order to enhance benefit realisation, or negate any particular negative effects that may be associated with RA use.

The research answered questions, including the ones listed below:
- What are the benefits of recycling asphalt'
- What is the additional benefit of recycling surface course back into new surface course'
- How does the toxicity of reclaimed asphalt compare to that of virgin aggregates and bitumen'
- How do the benefits of recycling compare to those of warm mix asphalt'
- By how much does moisture in RA diminish the benefits of recycling'
- How significant is durability in relation to recycled mixtures'

Undertaking a LCA was the most appropriate way to establish the necessary framework to answer these questions on an equitable, transparent basis. A range of ten impacts were assessed in the LCA, these are briefly summarised below. These were selected to meet the EN 15804:2012 standard on the Sustainability of Construction Works and were supplemented with four categories to assess toxicity.

Impact categories assessed

Impact category Unit Brief description

Abiotic depletion kg Sb eq The depletion of "non-living" finite natural resources.
Acidification kg SO2 eq Results in lowering the pH of water courses.
Eutrophication kg PO4--- eq Nutrient enrichment of water courses.
Global warming (GWP100) kg CO2 eq Leads to climate change effects.
Ozone layer depletion (ODP) kg CFC-11 eq A reduction of stratospheric ozone leads to a proliferation of UV rays at the Earth's surface.
Human toxicity kg 1,4-DB eq These four impact categories assess the potential toxicity levels associated with chemical compounds on humans and organisms in the fresh water, marine and terrestrial environments respectively.

Fresh water aquatic ecotoxicity kg 1,4-DB eq
Marine aquatic ecotoxicity kg 1,4-DB eq
Terrestrial ecotoxicity kg 1,4-DB eq
Photochemical oxidation kg C2H4 eq Photochemical oxidation results in summer smog production in the troposphere.

Key results from LCA study
The results of the LCA suggest:

- Recycling of asphalt to bound courses should be maximised.
- Surface-to-surface course recycling can realise even greater gains.
- Recycling should be prioritised over other environmental initiatives since it yields greater benefits.
- Moisture content in RA is not critical, though it should be minimised to maximise benefits.
- Durability is potentially very significant and preservation of the road surface should be considered alongside any environmental initiative.

Varying recycling routes and rates

The durability of asphalt mixtures remains one of the "biggest unknown quantities" in pavement design since variable durability can result from any of a number of factors of which recycled content may be one. The results of this and other past studies are inconclusive as to whether RA content is really a factor in enhancing or diminishing the durability of pavement courses. The resultant effect of variable durability is indicated in terms of environmental performance. Aside from recycling, durability is the factor that has the most potential to influence the life cycle environmental impacts of a road pavement.

Evaluating the effects of variable durability

The results of the LCA demonstrate that, above all, recycling to a bound course was significantly more environmentally advantageous than recycling to an unbound course. Appreciable extra benefit can be realised if high specification aggregates are preserved in their original application by surface-to-surface course recycling, due to the quarries that produce these specialised aggregates being widely spaced (hence requiring large transport distances for the aggregates). The moisture content that is sometimes present in reclaimed asphalt only mildly counteracts the recycling benefits. The sensitivity analysis shows that durability is the factor that has the most potential to influence the life cycle environmental impacts of a road pavement.

4 Production, Processing and Management at the Mixing Plant
4.1 Introduction
The objectives of the Work Package 4 research program was to facilitate the optimal use at the highest potential level of the natural resources and non-renewable materials already invested in the existing pavement structure. Recycling of wearing course material for new wearing courses is a huge benefit to the environment as it preserves natural resources of high quality aggregates as much as possible, and reuses the non-renewable resources "stored" in the bituminous binder.

Work Package 4 focus has been the production and management of reclaimed asphalts at the mixing plant, specifically on issues that are related to the processing and handling of reclaimed asphalt and its introduction into the mixing plant. The aim was to highlight factors that are considered essential to increasing the level of recycling of RA and to ensure optimum use of this RA.

The research program was divided into three task groups and topics:
- Production of reclaimed asphalt
The purpose has been to optimize the milling operation for optimal use of reclaimed asphalt as well as the possible influence of milling on gradation and aggregate properties.
- Handling of reclaimed asphalt
The objective has been to gain insight in the steps at the pre-processing unit or the asphalt plant with regard to initial inspection at delivery, crushing, sieving, storage, interim transport prior to reuse.
- Introduction of reclaimed asphalt into the mixing process
This part of the work has been intended to obtain information and gain experiences from recycling at the asphalt plant and the pros and cons of different asphalt plant designs in that respect.

4.2 Results obtained in the task group for production of reclaimed asphalt
The literature review for production of reclaimed asphalt by milling revealed that in practise only cold milling is performed. Hot scarifying of an old pavement surface as part of the remixing process is not considered as production of reclaimed asphalt because the material resource is not obtained to be used in a general manner for purposes elsewhere.

Cold milling works can be performed in order to reach various objectives:
- Improve the adherence conditions: fine milling machines can just remove the very first millimetres of the smoothed existing surface in order to regain roughness and adherence.
- Reshape the road surface: milling allows restoring the initial road profile by removing some material instead of overlaying where there is a lack of road material (collapsing, stripping, potholes, etc.). In this case, milling also restores drainage flow.
- Lower the road surface: milling prevents the road from being elevated too much because of regular overlays that are not consistent with bridge clearances, thresholds for entrance ramps and kerbs, bridge overloads, etc. The defective materials are removed without deconstructing the whole structure.
- Totally deconstruct the existing road: in this case, all materials are removed and a new structure is laid.
- Special purpose (e.g. traffic safety): Milling can be used to change the texture of the surface to introduce acoustical or vibrational impact on the driver as part of traffic safety features. This kind of milling is not considered as production of reclaimed asphalt as the tonnage produced is minimal.

Rough milled surface which will be overlaid by a new asphalt layer.

The expected changes of the reclaimed asphalt as opposed to the materials in place are mainly changes of the particle sizes (for any fraction) which are supposed to decrease because of possible fragmentation and attrition. In parallel it is possible to expect:
- The flakiness to be modified, but it is quite difficult to anticipate its evolution (increase or decrease of the flakiness index).
- The angularity and roughness to change (creation of new broken faces delimited by sharp angles, attrition).
- The fines content to increase, because of the aggregate "crushing" under the cutting tools action. However, the fine content may also decrease because of the loss of fines during milled surface sweeping. Usually it increases 2 or 3 % (order of magnitude).

In order to assess the impact of milling on RA aggregates, several road works were monitored. The mechanical and physical characteristics of milled products depend on the existing road pavement (binder, aggregates, material density and stiffness, etc.) and milling parameters like milling depth, drum rotation speed, level of wear of cutting tools etc.

Milling drum and two different types of cutting tools with various level of wear

The evolution of flakiness index was also studied. This property showed a slight decrease due to milling of about 2-3 points of the index, but this decrease remained in the order of repeatability of the test method.

Quantifying the correlation between RA water content and quantity of water added during the milling operation was difficult, despite some trends observed. The two effects of the speed and the magnitude of the water flow may compensate each other: going slower led to more water added for given water flow. The question of RA moisture, which is very important when the RA must be reused straight into a new warm or hot bituminous mixture, was set aside in all studied cases. It seemed that the water flow is mainly adjusted and optimized to prevent cutting tools to wear and to save time, while trying to maintain a reasonable quality of RA.

4.3 Results obtained in the task group for handling of reclaimed asphalt
Laboratory mixing questionnaire
Early on in the project WP2 needed to have some input from the road sector on how reclaimed asphalt was treated in laboratory mix design. A European standard described the principle for mix design but it was not clear what level of implementation this standard had reached. WP4 put together a short voluntary questionnaire which was issued to the road sector in the WP4 partner countries and more than 23 responses were collected. The results of the questionnaire are presented in D4.1 and reveals that a great variety exists in the utilisation of reclaimed asphalt, percentage used and how laboratory mixing is performed (if used at all).

Some results of the questionnaire are highlighted below:
1. Generally a cross section on the responses on percentage of RA indicated that 15-20 % is often the maximum unless parallel drums or similar devices for preheating the RA are available.
Several points are mentioned as reason to stay at this level:
a. Concern whether or not investment in production equipment for reaching higher percentages with the present and future level of availability of RA will have a reasonable payback period.
b. European products specification provide more lenient quality control for the RA if the asphalt producer stays below 10 % RA in surface layers and 20 % RA in bituminous base courses.
c. Technical limitations of the RA (like aggregate gradation and binder properties) enforced restriction in what is possible in normal practise.

2. Marshall mix design is in general still the most used guide for development of new mix recipes.

3. Even though a European standard exists for laboratory mixing (EN 12697-35) it is only used by half of the companies responding.

4. Some asphalt producers do not use laboratory mixing at all, but use full scale asphalt production facilities for their development of mix recipes. One responder even claims it is cheaper.

The following comments and focus points have also been mentioned in the questionnaire on the percentage of RA in the produced asphalt material. The percentage is often limited by a complex set of constraints which prohibit a high percentage of reuse in a particular mix. Among these factors are:
- binder ageing in the RA
- energy consideration (heat transfer limitation versus plant modification investments)
- gradation of RA in relation to the gradation of the new asphalt (e.g. gradation for noise reducing surface layers)
- availability of RA (large enough amount of homogenous RA material)
- area and environmental approval of stock piles
- cost/benefit considerations for
- virgin materials versus RA
- upgrading RA for reuse in surface layers
- demand for laboratory control of RA

Questionnaire for handling of RA
The situation at the pre-processing site or at the asphalt plant has been assessed with a questionnaire, presented in D4.2. Contractors in Denmark, Finland, Sweden, Norway, United Kingdom, Ireland, Slovenia, Austria and Italy were asked to answer the questionnaire. The focuses were on practical aspects of control issues at the gate when reclaimed asphalt is delivered (including sampling and analysis), the use of pre-processing of the RA and interim storage before reuse. To a certain extent the responses to the questionnaire reflects national legislation, contractual relations and tradition that all influence the objective of the project with respect to reaching high percentages of RA in surface layers.

Results of the questionnaire were:
1. Some differences in validation of RA between countries were expected. Due to the climate in the northern countries, and use of studded tires, surface layers with high quality aggregate are separated for reuse in surface layers. Also the maximum aggregate size is higher (16 mm) due to better wear resistance.
2. Denmark uses nearly all RA for base layers, since road authorities are reluctant to allow it in surface layers to avoid the risk of frost sensitive aggregate particle from old base courses. Therefore all RA go into the same pile.
3. All countries document their RA according to EN 13108-8.
4. The value of the RA varies, it can be positive, zero or negative depending on the local situation. In some cases RA is the property of the road owner.
5. Beside contamination of earth or gravel, coal tar is the only real contaminant to consider. The only speed screen method used is for coal tar.
6. Size and type of interim storage depends on local legislations. Some companies have paved foundations and sewer systems with oil separators to take care of run-off water from RA piles in one plant, while the other plant has its stock piles on gravel. In order to minimize water content and save fuel, tents or some kind of cover is used, at least for finer fractions.
7. Interim transport of RA is handled as virgin material. RA is normally crushed in batches and the amount varies from "not very big" to 10,000 - 15,000 t. Roof shaped piles are common and actions to prevent segregation are taken.
8. Crushing and sieving is performed in one operation, with a feedback loop for oversized material.

Many more details can be found in the deliverable D4.2 Annex 7.1.

In D4.4 selected case studies are presented for handling of RA and in D4.5 Case studies for introduction of RA in the mixing process, are presented.

4.4 Results obtained in the task group for introduction of RA in the mixing process
The questionnaire on introduction of RA in the mixing process was used in order to find the benefits and problems of different designs of asphalt plants, with the goal to find all possible scenarios enabling us to achieve high levels of RA into the mix. The results have been presented in D4.2.

In situ recycling
Several countries responded that they sometimes use in situ recycling for the replacement of wearing courses. Also foamed bitumen (and emulsion) technology is frequently used. Base asphalt layer is commonly produced with foamed bitumen (and emulsion) technology.

Plant recycling
Roughly 95% of asphalt plants are batch plants with charge production.
The maximum capacity of asphalt plants varies significantly from 50 to 600 t/h. The optimum content of reclaimed asphalt in asphalt mixtures varies in different countries, from 7 to 50 % (mass). In the Netherlands the wearing course optimum content of reclaimed asphalt is 30 % (mass) and in the base course 50 % (mass).

The production capacity of an asphalt plant can increase by 30 % when parallel drums are used due to the parallel heating of RA in a second drum. However, in cases when cold reclaimed asphalt is introduced in the mixer, the production capacity can decrease with 20 %.

Prolongation of the mixing time varies from 0 s up to 15 s when reclaimed asphalt is introduced.

Softer bitumen is the most common virgin material added to the mixture. Some producers use additives like rejuvenator oils as a lubricant in a parallel drum. Other additives include hydrated lime, pulverized fly ash and Sasobit synthetic wax.

Heating reclaimed asphalt indirectly with hot mix of stone fractions is, in the majority of countries, the most common way to introduce lower quantity of reclaimed asphalt. Only in Belgium and the Netherlands separate heating of reclaimed asphalt and hot mix of stone fractions is the predominantly used practise.

5 Performance Modelling
5.1 Introduction
The objectives of the Work Package 5 research program, Performance modelling of RA, was to gain deeper insight into the mechanics of asphalts composed of recycled and new components. A broad material test program was conducted to investigate and understand the elastic, visco-plastic and visco-elastic deformation as well as the fatigue performance of asphalt mixes containing RA and to determine the input parameters required for numerical simulations. The results obtained from the material test program should form the basis for models describing the constitutive characteristics of asphalt materials. A prerequisite was to design and select representative RA mixes with nearly identical grading and clearly different amounts of aged and unaged binders. In the research program the performance of two asphalt mixtures with RA were evaluated in comparison to a virgin asphalt mixture.

The aims were to develop visco-elastic and plastic models that can be used to design pavements made of recycled materials, and to predict pavement life as well as their sensitivity to damage.

5.2 Materials tested
Asphalt mixes with nearly identical grading and clearly different amounts of aged and unaged binders were taken into account in this investigation. As reference material, Material I SMA 11 S was chosen. For Material II and III reclaimed asphalt replacement levels of 15 M.-% and 30 M.-% were used. The chosen reclaimed asphalt originated from porous asphalt. Virgin binder for all asphalt materials was PmB 25/55-55A. All virgin greywacke aggregate came from the same quarry and the filler was made from dolomite.

Grading curves of the asphalt mixes. Material I (0% RA); Material II (15% RA); Material III (30% RA)

Binder content in the mixes and properties of recovered binders
Property – test method Material I
(0% RA) Material II
(15% RA) Material III
(30% RA)
Binder content (% mass) 6.55 5.63 7.20
Penetration at 25°C (mm/10) – EN 1426 34 32 20.5
Softening Point (°C) – EN 1427 63.3 66.2 69.6
Viscosity at 135°C, (Pa s) – EN 13302 1.74 1.80 2.71
Viscosity at 150°C, (Pa s) – EN 13302 0.63 0.64 1.06

The extracted binders for Material I and II were quite similar but the binder in Material III was stiffer and had a more elastic character than the other binders. The binders as well as mastic made from the binders were further characterized for fatigue and healing performance and the results are presented in D5.5.

5.3 Performance tests
Elastic modulus
The compressive dynamic test was used to determine the elastic modulus and Poisson's ratio of the specimen under the action of cyclic stresses. The test provides the material properties needed for the parameter model identification of the elastic-visco-elastic model. For the characterization of the absolute modulus, |E1|, of each material, repeated load triaxial tests with seven load cases each at several temperature-frequency combinations were conducted.

Absolute modulus |E1| master curves (TR=20°C) (left), and as function of temperature at f=1Hz (right).

From the comparison of the three master curves, it can be concluded, that at high temperatures and low frequencies (i.e. left part of the curve) the stiffness of the two asphalt mixes with RA is higher compared with the stiffness of the virgin mixture (Material I).

A diagram that represents the asphalt mixture rheology is the so called Cole-Cole plot. This diagram is the results of plotting the storage modulus E' vs. the loss modulus E''. The Cole-Cole plots are shown for the three materials.

Schematic Cole-Cole-Plot with storage and loss modulus (left) and Cole-Cole Plots for Material I, II and III (right).

Material III behaves more elastic than the other two materials. At low temperatures and high frequencies the visco-elastic behaviour of Material III differs from the trend observed on Materials I and II. The behaviour of Material II could be due to the significantly lower binder content of this mix. The master curves and the Cole-Cole plots are required for the determination of the parameters of the visco-elastic model.

Permanent deformation
A multistage triaxial test program was developed for testing permanent deformation under repeated loading. This progressive increase of accumulation of permanent deformations at high stress ratio is further evidenced in the asphalt mixes containing RA.

Variation of permanent strain with number of load cycles

Fatigue and healing
Repeated load tests are required to obtain the fatigue properties of asphalt mixes. It became quite clear that the fatigue relation is changing significantly due to the RA content of the asphalt. However, the number of load cycles to fatigue failure in-situ (in a real pavement) cannot be estimated directly from the results of laboratory fatigue tests. For this reason fatigue models that utilise the results of fatigue tests are needed to estimate the number of load cycles to fatigue failure.

5.4 Model development
Two different models, using different input parameters, have been developed.

The first approach was two decoupled models: a visco-elastic and a plastic model. While the visco-elastic model was meant to determine the stresses and visco-elastic strains at all relevant points in the pavement considering influencing factors such as temperature and frequency etc., the empirical plastic model would use these stresses and strains to estimate the accumulation of permanent strain. Both models, the visco-elastic and the plastic one, may not be very scientifically sophisticated, but very practical. With these two models at hand, even pavement engineers that are not too familiar with constitutive modelling techniques of pavement materials can estimate the performance of pavements with recycled asphalt material. In addition, the computational time needed with this first approach is very low.

The second approach is based on a fully coupled visco-elastic-plastic model which was developed in collaboration with the group of structural mechanics at TU Delft. From a scientific point of view, this model is more sophisticated than the first approach and a higher accuracy of the results produced with this model can be expected. This model provides a platform more suitable for scientists/researchers and its application requires a background of computational mechanics, programming and constitutive modelling. The model presents the following advantages:
- It is an available formulation that could be implemented in any FE code with an open library of materials.
- It only requires a few experimental tests: Single Uniaxial Creep and Recovery tests within the in-service range of strains and stresses

Therefore, the two different models proposed have been developed to satisfy practical needs as well as scientists'/researchers' requirements. The details of the models are presented in D5.5.

Model validation was carried out by comparing the results achieved by the models with the Hamburg and Dublin wheel tracking test results, both tests were performed at 40 °C.

For the Dublin wheel track test the rut depth was measured after 150 000 wheel passes both using a manual method and using a laser based technique, LiDAR.

Average permanent deformation calculated using LiDAR and using manual measurements in the Dublin Wheel track test
Mix Type Average Measured Rut Depth (mm) Average LiDAR Calculated Track Path Depth (mm)
Material I (0% RA) 1.11 0.89
Material II (15 %RA) 0.52 0.16
Material III (30% RA) 0.31 -

As can be seen from these figures, the level of permanent deformation observed was very low. The use of increased levels of RA in the asphalt mixtures resulted in increased resistance to permanent deformation. Based on the values, it can be seen that the use of 15 % RA resulted in a 53 % reduction in permanent deformation and the use of 30 % RA resulted in a 72 % decrease.

Pavement Design Life Calculations
The experimental data on the three materials were used in pavement design life calculations. Calculations were conducted using an estimation of the tensile strains at the bottom of the asphalt layer and fatigue life is defined by the occurrence of the macro-crack at the bottom of the asphalt layer. The calculations were made using the pavement design software TISAD. This design software is based on multi-layered elastic analysis and proceeds according to the German analytical design procedure for asphalt pavements RDO Asphalt. Two pavement structures were investigated, one with a 10 cm and one with a 20 cm thick asphalt surface layer consisting of either Material I, II or III. Beneath the asphalt layer 30 cm of unbound granular layer was modelled with a stiffness of 150 MPa and 20 cm of a subbase with a stiffness of 120 MPa. The subgrade was modelled with a stiffness of 45 MPa.

Potential Impact:

Re-road aimed at contributing to improve recyclability and increase the energy efficiency with regards to asphalt pavements. The targets of the project were very ambitious as we were aiming at 99 % of re-use of asphalt materials. This target is far from the general level of recycling of asphalt material today. Much reclaimed asphalt ends up as unbound material. A rough estimate from polls done by the European Asphalt Pavement Association suggests that approximately 30 % of reclaimed asphalt is re-used as unbound material.

In the project we have provided tools to assess the potential of recycling of particular asphalt materials and tools to assess any environmental risk associated with reclaimed asphalt.

The life cycle assessment within the project proved that recycling to a bound course was significantly more environmentally advantageous than recycling to an unbound course.

If recycling to bound layers is achieved instead of recycling to unbound layers significant life cycle benefits are achievable with average reductions across all impacts of 15.4 %, with the biggest potential reductions in abiotic depletion, terrestrial ecotoxicity and the least in marine aquatic ecotoxicity.

Recycling to bound layers lowers the "carbon footprint", or "global warming potential (GWP100)", by 13.0 % at the highest recycling level (30 %) investigated.

Since at least 30 % of the 55 million tonnes of yearly available reclaimed asphalt in Europe end up in unbound layers, there is a great potential to reduce the "carbon footprint" and other environmental footprints of the road building sector if the results of this project is implemented at a pan-European level.

To reach these ambitious goals on a European scale it will be absolutely necessary to continue to work at a pan-European level. Both the experience from the different countries involved in the project and the dissemination on a regional scale are crucial for the achievement of the expected impact.

Dissemination activities
The key dissemination and implementation activities of the project included:

- Internet site. Creation and maintenance of the public website
The web page for this project was placed within a clustered website area for FEHRL Strategic European Road Research Programme; this provided users with a streamlined portal for access to all related projects in the programme.

- Preparation of papers. Preparation of papers for international conferences, articles for national journals, newsletters for end users.
The papers and articles for scientific journals and conferences were crucial for giving the scientific community a possibility to review our results. Articles and newsletters etc. were efficient means to communicate our results to a wider audience, e.g. road authorities and industry.

- Contacts with end-users. Preparation and implementation of meetings with reference end-users, feeding guidance from these meetings into the project.
Interaction with the key stakeholders, such as road owners, was crucial for (future) implementation of the results. The clustering action managed by FEHRL provided a critical mass of related highway research projects, which contributed to dissemination of the Re-Road project. The FeRMM conference is but one example on this clustering effect.

- Launching regional seminars for key stakeholders
The regional seminars played a major role in catching the awareness of the stakeholders as it was also recognized that many stakeholders do not have time to attend focused scientific international conference or read scientific papers.

- Final symposium. Organisation of the final symposium for presentation of the results of the project.
The final seminar attracted 75 people from across Europe representing Re-Road target groups and demonstrating the efficiency of the FEHRL partner as providing easy access to a broad network of stakeholders across Europe.

The project was presented at a number of events and through a range of distribution channels. The following tables summarize the dissemination activities and publications:

Exploitation
The project was focused on the production of new knowledge and characterization tools. Therefore, it has not been desired by anyone within the project to secure the Intellectual Property Rights for himself or herself and no money has been allocated in the budget for the protection of knowledge.

The output of the project has been oriented towards the public domain in order to have the highest possible impact on the level of re-cycling of reclaimed asphalt, preserving natural resources and reductions of energy usage in maintenance of the infrastructure.

The most important factor for success for the exploitation of the foreground in this collaborative research project is to continue the dissemination of the results

- via the website,
- via databases,
- via seminars and conferences
- through tandardization bodies
- to use the results, e.g. laboratory tools for optimizing re-cycling of asphalt

List of Websites:

http://re-road.fehrl.org
141994871-8_en.zip