ITF Research Reports Long-life Surfacings for Roads: Field Test Results 9282108104, 9789282108109

Citation preview

This work is published under the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of OECD member countries. This document and any map included herein are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

Please cite this publication as: ITF (2017), Long-life Surfacings for Roads: Field Test Results, ITF Research Reports, OECD Publishing, Paris. http://dx.doi.org/10.1787/9789282108116-en

ISBN 978-92-82-10810-9 (print) ISBN 978-92-82-10811-6 (PDF)

Series: ITF Research Reports ISSN 2518-6744 (print) ISSN 2518-6752 (online)

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

Photo credits: Cover © Richard Elliot.

Corrigenda to OECD publications may be found on line at: www.oecd.org/about/publishing/corrigenda.htm.

© OECD 2017 You can copy, download or print OECD content for your own use, and you can include excerpts from OECD publications, databases and multimedia products in your own documents, presentations, blogs, websites and teaching materials, provided that suitable acknowledgement of OECD as source and copyright owner is given. All requests for public or commercial use and translation rights should be submitted to [email protected]. Requests for permission to photocopy portions of this material for public or commercial use shall be addressed directly to the Copyright Clearance Center (CCC) at [email protected] or the Centre français d’exploitation du droit de copie (CFC) at [email protected].

The International Transport Forum The International Transport Forum is an intergovernmental organisation with 57 member countries. It acts as a think tank for transport policy and organises the Annual Summit of transport ministers. ITF is the only global body that covers all transport modes. The ITF is politically autonomous and administratively integrated with the OECD. The ITF works for transport policies that improve peoples’ lives. Our mission is to foster a deeper understanding of the role of transport in economic growth, environmental sustainability and social inclusion and to raise the public profile of transport policy. The ITF organises global dialogue for better transport. We act as a platform for discussion and prenegotiation of policy issues across all transport modes. We analyse trends, share knowledge and promote exchange among transport decision-makers and civil society. The ITF’s Annual Summit is the world’s largest gathering of transport ministers and the leading global platform for dialogue on transport policy. The Members of the ITF are: Albania, Armenia, Argentina, Australia, Austria, Azerbaijan, Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Canada, Chile, China (People’s Republic of), Croatia, Czech Republic, Denmark, Estonia, Finland, France, Former Yugoslav Republic of Macedonia, Georgia, Germany, Greece, Hungary, Iceland, India, Ireland, Israel, Italy, Japan, Korea, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Mexico, Republic of Moldova, Montenegro, Morocco, the Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Russian Federation, Serbia, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine, the United Kingdom and the United States. International Transport Forum 2, rue André Pascal F-75775 Paris Cedex 16 [email protected] www.itf-oecd.org

ITF Research Reports ITF Research Reports are in-depth studies of transport policy issues of concern to ITF member countries. They present the findings of dedicated ITF working groups, which bring together international experts over a period of usually one to two years, and are vetted by the ITF/OECD Joint Transport Research Committee. Any findings, interpretations and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the International Transport Forum or the OECD. Neither the OECD, ITF nor the authors guarantee the accuracy of any data or other information contained in this publication and accept no responsibility whatsoever for any consequence of their use. This document and any map included herein are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

FOREWORD – 5

Foreword This report is the third phase of an important research project that began in 2002. Taken as a whole, the project was designed to address the issues of road maintenance and the potential of long-life wearing courses to increase the longevity of a road. The project is genuine research work at the forefront of road construction technology and is the fruit of the collaborative efforts of experts representing 12 countries. This report summarises the results of field trials of two innovative materials for road surfacing. This project has been possible thanks to the dedication of the experts of the working group and their colleagues at testing laboratories, to the audaciousness of road owners who made it possible the construction of road sections with these two materials, to the engineers who worked on the fine-tuning of the mixes, and to the construction site teams who worked very hard, sometimes night and day, on the laying out of the courses with tight timing constraints. The ITF would like to express, in particular, its gratitude to the following organisations and agencies, without which such an innovative research project could not have been undertaken: •

Agence Nationale de la Recherche (France)

•

Conseil Général de la Sarthe

•

Conseil Général de Loire Atlantique

•

IFSTTAR

•

Opus International Consultants Ltd

•

Fulton Hogan Ltd

•

New Zealand Transport Agency

•

UK Highways Agency

•

URS Infrastructure & Environment UK Limited.

The ITF would also like to express its gratitude to the main authors of this report: Mr Finn Thøgersen (Danish Road Directorate, Denmark) and Mr Richard Elliott (URS Infrastructure and Environment UK Limited), respectively co-ordinators of the HPCM and epoxy-asphalt trials, and Mr François de Larrard (IFSTTAR and Lafarge, France), Chairman of the Working Group.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

TABLE OF CONTENTS – 7

Table of contents Abbreviations ............................................................................................................................................. 9 Executive summary.................................................................................................................................. 11 Chapter 1. Increasing the longevity of wearing courses ...................................................................... 13 Summary of Phase 1: Economic evaluation of long-life pavements ...................................................... 14 Summary of Phase 2: Laboratory testing of epoxy asphalt and high-performance cementitious materials ................................................................................................................................................. 15 Objectives of the field trials (Phase 3) and working method ................................................................. 19 Monitoring of the trial sections .............................................................................................................. 19 Content of the report .............................................................................................................................. 21 References .............................................................................................................................................. 22 Chapter 2. Epoxy-asphalt road surfacing field trials .......................................................................... 23 Plant trials of epoxy asphalt in France ................................................................................................... 24 Field trials of epoxy asphalt open graded porous asphalt in New Zealand ............................................ 25 Field trials of stone mastic asphalt and epoxy asphalt in the United Kingdom...................................... 34 Assessment of performance.................................................................................................................... 45 Recommendations .................................................................................................................................. 46 Summary and conclusions ...................................................................................................................... 47 References .............................................................................................................................................. 50 Chapter 3. Field trials with high-performance cementitious materials ............................................. 53 Trials of HPCM in the United Kingdom ................................................................................................ 54 Trials of HPCM in France ...................................................................................................................... 56 Assessment of performance.................................................................................................................... 64 Recommendations .................................................................................................................................. 64 Summary and conclusions ...................................................................................................................... 65 References .............................................................................................................................................. 66 Working Group members ....................................................................................................................... 67

Figures Figure 1.1. Figure 2.1. Figure 2.2. Figure 2.3. Figure 2.4. Figure 2.5. Figure 2.6. Figure 2.7. Figure 2.8. Figure 2.9.

Principle of the HPCM pavement ....................................................................................... 18 Two components of epoxy-asphalt binder .......................................................................... 26 General view of the site ...................................................................................................... 27 Start of the 20% EMOGPA section .................................................................................... 28 Compaction of 20%EMOGPA ........................................................................................... 28 Traffic damage to 30% EMOGPA (plucked chip outside wheel tracks) ............................ 29 Curing of field trial specimens............................................................................................ 30 Start of 30% air voids content EMOGPA section, looking towards 20% air voids content OGPA section ........................................................................................................ 32 Start of 20% air voids content EMOGPA section, looking towards 30% air voids content EMOGPA section .................................................................................................. 32 Mean noise levels for cars, dual-axle and multi-axle trucks (SPBI =Statistical Pass By Index (according to ISO 11819-1)) ......................................... 33

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

8 – TABLE OF CONTENTS

Figure 2.10. Figure 2.11. Figure 2.12. Figure 2.13. Figure 2.14. Figure 2.15.

Composition of SMA mixture ............................................................................................ 36 General view of plant setup (bitumen delivery system on left) .......................................... 37 Failed patch reinstatement typical of pre-trial condition .................................................... 38 Paving of epoxy asphalt with conventional plant ............................................................... 39 Epoxy-asphalt trial section after 12 months trafficking (2013) .......................................... 42 Mean stiffness (ITSM) data at three temperatures, performed on in situ cured cores, 13 months after production .................................................................. 43 Figure 2.16. Mean stiffness (ITSM) data at 20°C, performed on laboratory cured cores, at various ages after production .......................................................................................... 44 Figure 3.1. Equipment used for the HPCM laying................................................................................ 55 Figure 3.2. Machines developed for laying HPCM .............................................................................. 56 Figure 3.3. Construction of the HPCM in Brettes-les-Pins ................................................................... 57 Figure 3.4. Aspects of the surfacing of the HPCM section in Brette-les-Pins ...................................... 58 Figure 3.5. Construction of the HPCM section in St. Philbert (France) ............................................... 62 Figure 3.6. Skid-resistance of GFRUHPC, as measured with the Wehner & Schulze machine ........... 63 Figure 3.7. Noise generation as measured by the CPX method on the St. Philbert test section after two weeks of traffic .................................................................................................... 63 Tables Table 1.1. Table 2.1. Table 2.2. Table 2.3. Table 2.4. Table 2.5. Table 2.6. Table 2.7. Table 2.8. Table 3.1.

Monitoring requirements before, during and after the construction of the trial sections.... 20 Mixture design for field trial............................................................................................... 26 Air temperatures near to trial site (December 2007–March 2010) ..................................... 30 Rutting ................................................................................................................................ 31 Skid resistance (British Pendulum Number) ...................................................................... 31 Water permeability ............................................................................................................. 33 Mixing temperature and maximum usable life of epoxy asphalt ........................................ 39 Summary of skid resistance, texture and rut depth data ..................................................... 41 Mean tensile strength (ITST) data at 20°C at various ages after production ...................... 44 HPCM mix-composition ..................................................................................................... 60

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

ABBREVIATIONS – 9

Abbreviations ALT

Accelerated loading tests

AADT

Annual average daily traffic

BBTM CL

Beton bitimineux très mince (Very thin asphalt concrete) Confidence Limit

CPX

Close proximity method to measure noise level

EA

Epoxy-asphalt

EMOGPA

Epoxy-modified open graded porous asphalt

FOR

Forever Open Road

FWD

Falling weight deflectometer

GFRUHPC

Grooved fibre-reinforced ultra-high performance concrete

HCV HPCM

Heavy class vehicles High-performance cementitious material

IRI

International rutting index

ISO

International Organization for Standardization

ITST LC

Indirect tensile splitting strength Lane centre

LLP

Long-life pavement

NSWT

Nearside wheel track

OGPA PVA fibres

Open graded porous asphalt Poly(vinyl alcohol) fibres

SCRIM

Sideway-force Coefficient Routine Investigation Machine

SMA

Stone mastic asphalt

UHPC

Ultra-high-performance concrete

Units of measure °C Celsius dBA A-weighted decibel kPa

Kilopascal

kN

Kilonewton

MPa

Megapascal

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

EXECUTIVE SUMMARY – 11

Executive summary Background Maintenance of high-traffic roads is expensive, and has indirect social costs that are less and less acceptable in the context of traffic gridlock in big cities. Therefore, a shift from low cost, low durability maintenance techniques to more durable and more expensive ones is desirable for the sake of the common interest. The long-life pavement (LLP) project launched by the ITF in 2002 addressed these issues in a series of three reports of which this is the final one. The first phase of the project focused on the economics of road surface maintenance. It concluded that for a sufficient level of traffic, long life wearing courses would be globally sustainable at a life expectancy of 30 to 40 years, with a total cost less than three times that of current techniques. Two families of material – epoxy asphalt and high performance cementitious material -- appeared to have the potential for this and a number of countries involved in the project commissioned their national road laboratories to investigate this question. During the second phase of the project, nine countries joined efforts to study two potential solutions. The first one was of the bituminous type: epoxy-asphalt (EA). This material has a good record in the field of long-span steel bridge decks. Extensive testing in the laboratory as well as in accelerated loading tests (ALT) gave a positive appraisal of this material’s potential use in longer road sections submitted to heavy traffic. The second, more innovative technique underwent laboratory tests for the first time as part of the the project. This high-performance cementitious material (HPCM) consists of a surface dressing with hard, polishing-resistant chippings embedded in a thin layer of ultrahigh-performance, fibrereinforced mortar. HPMC’s mechanical properties proved to be as encouraging as the alternative solution. The third phase consisted of full-scale field tests. These field trials were carried out between 2009 and 2012 in three of the participating countries, France, New Zealand and the United Kingdom.

Findings The epoxy-asphalt (EA) solution has so far met initial expectations. Feedback from the UK and New Zealand test sites is encouraging, suggesting the transfer of this technology from bridges to pavements at an industrial scale seems possible. The HPCM solution is not yet at the same level of maturity. The technique proved difficult to apply in an industrial scale. It also tends to create noisy pavements, which are becoming less and less accepted. Grooved fibre-reinforced ultrahigh-performance concrete (GFRUHPC) seems more promising, but needs further development to achieve full control of production and placement. Specifically, issues like shrinkage-induced cracking need to be mastered. Once optimised, the quality of the material should provide road owners with an alternative, long-life and sustainable road surfacing option. The short, experimental test sections built during Phase 3 did not allow a sound economic assessment of what could become a mature technique. Even if the economic justification can be only demonstrated through a tender process looking at whole-life costs, it is likely that the recommended

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

12 – EXECUTIVE SUMMARY

techniques could meet the goal of keeping a total cost at less than three times the cost of current solutions.

Policy insights Long-life surfacing is essential for advanced and affordable transport infrastructure A long-life surfacing is an essential requirement for the advanced and affordable transport infrastructure envisaged by the Forever Open Road (FOR) concept. Long-lasting overlays as part of durable and integrated pavements are one of the key research and innovation themes of FOR in order to produce an affordable road for a society that cannot afford road closures. The higher cost of long-life road surfacing materials is justified particularly for road network hot spots While the additional cost and marginally increased construction complications will limit the use of advanced long-life road surfacing materials for many conventional applications, the justification for advanced surfacing materials can only increase as traffic levels continue to rise. In certain road network hot spots where any loss of serviceability is unacceptable, they will be the material of choice. It will be important to continue monitoring existing test sections in the future to corroborate findings over the road pavement life cycle In 2025, the test sites in the UK and New Zealand will have reached about 15 years of use; an age where normal wearing courses are reaching the end of their life cycle. It is recommended to continue with site visits to obtain further corroboration regarding the usefulness of this technique. Research on potential health issues related to the use by epoxy binders should continue in the light of concerns in some countries regarding the application of epoxy binders on site.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

1. INCREASING THE LONGEVITY OF WEARING COURSES – 13

Chapter 1. Increasing the longevity of wearing courses

This chapter summarises the results of the two first phases of the project: the study on the economic viability of long-life wearing courses and the laboratory testing of candidate materials for long-life surfacing. It presents the objectives of the third and final phase of the project: the full scale field tests of two candidate materials (epoxy asphalt and high performance cementitious material).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

14 – 1. INCREASING THE LONGEVITY OF WEARING COURSES

Maintenance and rehabilitation of existing roads constitute a permanent challenge for road administrations and an increasing part of their budget. In 2011, the share of public expenditure on road maintenance represented between 25% and 35% of total road expenditure (ITF, 2013). These costs do not take into account user costs due to disruption and congestion, which also increase and are of particular concern for heavily trafficked roads. In periods of budget pressure that affect many ITF countries, savings on public expenditures and user costs to maintain the road infrastructure would be welcome. It is in this context that the notion of “long-life pavements” started to interest policy makers. Long-life pavements would be expected to show high quality performance without the need for significant repair for more than 30 years and, under certain conditions, the benefits of avoiding major repairs and re-pavements may become large enough to justify the higher initial costs of such pavements. It has been demonstrated that long-life as just described is achievable for the subsurface pavement layers, but the surface layer or wearing course, which is critical for safe and comfortable driving, remains the Achilles’ heel of the concept. This thin uppermost pavement layer is, more than any other part of the structure, exposed to air, sun and weather, and to the wear, tear and deformation from the traffic it carries. Today, pavements with bitumen or cement binders dominate the market. They function well in a wide range of traffic and climate conditions and have few environmental disadvantages. However, although quality products are available, most pavements exhibit shortcomings in terms of durability, road-user qualities, strength and repair needs. This translates into poor maintenance economy when these pavements face the challenge of the increasing vehicle-mass limits and higher density of traffic on the arterial roads of today and the near future. It is in this context that the OECD decided to launch in 2001 a major research project to develop a new long-life pavement surfacing. This project was conducted in three distinct phases: Phase 1 (2001-2003): Economic viability of long-life wearing courses and identification of candidate materials for long-life surfacing Phase 2 (2005-2007): Laboratory elaboration and accelerated load testing of suitable candidate materials Phase 3 (2009-2013): Field tests of the selected materials. This 12-year project ended in 2013 after two to six years monitoring of the trial sections. This report presents the detailed results of the field tests, including the preparatory work, the construction phase and the monitoring phase, and presents the final conclusions of the project.

Summary of Phase 1: Economic evaluation of long-life pavements European Conference of Ministers of Transport (ECMT, 2005) assessed the likely envelope of costs for economic viability of new long-life wearing courses, taking into account all the costs involved including initial construction costs as well as savings in maintenance costs and user cost savings expected in the longer term. Phase 1 provided guidelines for a research programme to assess the real capacity of the candidate materials and their suitability as long-life wearing courses and recommended to undertake laboratory testing on both materials (Phase 2).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

1. INCREASING THE LONGEVITY OF WEARING COURSES – 15

From a cost viewpoint, Phase 1 concluded that long-life pavement surfacing costing around three times that of traditional wearing courses per square metre would find a market for a range of high-traffic roads, depending on an expected life of 30 years, discount rates of 6% or less and an annual average daily traffic (AADT) of 80 000 or more with at least 15% of heavy trucks. A review of advanced surfacing materials confirmed that that there were indeed materials that could be feasible for long-life surfacing and from the review of materials it was concluded that two types of materials in particular had the potential to fulfil the requirements: epoxy asphalt and high-performance cementitious materials (HPCM). Epoxy asphalt Epoxy asphalt is a premium material, which has been used for many years as a road surface on stiff bridge decking. The first such application, on the San Mateo Bridge (California, United States), is still meeting performance requirements, after 40 years of service. Over time, epoxy asphalt has been more widely used for stiff bridge decking applications in a number of other countries. Administrations have not used epoxy asphalt for regular road pavement surfaces as cheaper materials have been available which, although they may not last as long, could be replaced relatively easily and each time at moderate cost. High-performance cementitious materials In the last years of Phase 1, there were already references to cementitious materials which exhibited superior strength and durability properties. For instance, the family of ultra-high-performance fibre-reinforced concrete was gaining recognition, and recommendations were released in 2004 to practical applications, mainly in the precast concrete industry or for some special, niche markets. Therefore it was stated that such solutions could be studied for the pavement domain, even if the material would need substantial adaptations to match the specifications of a road wearing course. A new solution, proposed for the next project phase, was entitled high-performance cementitious material (HPCM). It is an innovative product which was developed and tested for road surfacing applications for the first time during this project. This pavement surfacing consists of a layer of ultra-high performance, fibre-reinforced fine mortar, in which hard, polish resistant aggregate particles are embedded.

Summary of Phase 2: Laboratory testing of epoxy asphalt and high-performance cementitious materials The objective of the second phase of the project (ITF, 2008) was to research the behaviour and properties of the materials identified as candidates and test them sufficiently to assess their suitability for use in long-life wearing courses. Co-ordinated testing, including accelerated load tests, was conducted to assess the durability and other essential properties identified for long-life wearing courses. These tests were done in the following national testing laboratories: •

Australia: New South Wales Roads and Traffic Authority

•

Denmark: Danish Road Institute and DBT Engineering

•

France: Laboratoire Central des Ponts et Chaussées (LCPC, now known as IFSTTAR, for Institut Français des Sciences et Technologies de l’Aménagement et des Réseaux)

•

Germany: Federal Highway Research Institute (BASt)

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

16 – 1. INCREASING THE LONGEVITY OF WEARING COURSES

•

New Zealand: Transit New Zealand (now known as The New Zealand Transport Agency) and Opus International Consultants Ltd.

•

United Kingdom: Transport Research Laboratory (TRL) Ltd, Scott Wilson (now URS Infrastructure and Environment UK Limited) and the Highways Agency

•

United States: Turner Fairbank Highway Research Center.

Main findings regarding epoxy asphalt On the basis of the comprehensive testing undertaken, acid-based epoxy-asphalt mixtures were found to have greatly improved performance compared to conventional mixtures. In particular, compared to conventional asphalts, cured epoxy asphalts are significantly: •

stiffer (higher modulus) at service temperatures, with greater load-spreading ability

•

more resistant to rutting

•

more resistant to low temperature crack initiation

•

more resistant to surface abrasion from tyre action, even after oxidation

•

more resistant to fatigue cracking (although the benefits are less marked at higher strain levels)

•

less susceptible to water induced damage

•

more resistant to oxidative degradation at ambient temperatures.

A limited accelerated pavement testing (APT) trial of epoxy-modified open graded porous asphalt (OGPA) resulted in early signs of surface abrasion in the control section but not in the epoxy one. Tests on the APT sections demonstrated that the skid resistance of epoxy asphalt was not significantly different from that of conventional asphalt. The tests undertaken confirmed that epoxy asphalt is a premium material that outperforms conventional binders. Test performance of the epoxy-asphalt materials studied in Phase 2 was considered greatly superior when compared with conventional materials, on the important indicators central to assessment of the potential for long service life. Performance expectations for the longevity and durability of epoxy-asphalt surfaces were built up during the project taking into account the results of the tests undertaken and experience with their relationship to longevity in the field. Nearly all the testing has indicated that epoxy asphalt should provide a durable long lasting surfacing, even in the most heavily trafficked road situations. The tests undertaken showed the type of epoxy materials and the choice of aggregates had an important impact of the performance of the surfacing. In addition, epoxy asphalt needs close supervision at time of production and laying to ensure full mixing is achieved; time and temperature need to be carefully monitored to achieve the best performance outcomes. Phase 2 recommended that further research work be undertaken, in particular on the following aspects: •

Curing and construction time. Further laboratory studies are needed prior to any demonstration projects to optimise the curing profile with the desired rate of reaction for the local conditions (for example, time for curing, distance of transport and laying). LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

1. INCREASING THE LONGEVITY OF WEARING COURSES – 17

•

Curing period. It is important to establish when after the initial blending of the epoxy asphalt the reaction is complete.

•

Curing temperature. Some epoxy systems have shown the ability to cure rather rapidly at a lower temperature than might be expected. The prospects for lower temperature curing – and the related potential for energy and cost savings during production – need further research.

•

Production process: Epoxy asphalt is a material with high stiffness that can be applied in thin surface layers. Production experience to date for the relatively small quantities used has almost exclusively been with a batch plant that gives good control of mixing time – an important part of its subsequent curing and post-curing properties. However, for the trials in New Zealand a continuous mix drum plant was used without problems.

•

Construction process: Due to the thermosetting nature of the material, extra care is required in the timing of manufacturing and construction phases to ensure the product is not overcured before compaction. The risk of construction failures and damage to plant is greater than with conventional bitumen. For both these areas, the perceived risk is likely to diminish in importance as experience with the material grows.

•

Health impact: When uncured, certain epoxy materials contain compounds harmful to people and the environment. These were not used for the epoxy asphalts in this project. However, if such materials are used, special equipment and safety precautions would be required for all involved in handling them while uncured.

In summary, Phase 2 concluded that if all aspects of the process are correctly handled, epoxy asphalt should be able to provide a surfacing material that can be expected to meet the aim for a much extended, practically maintenance-free life, i.e. 30 years or more. It was recommended to undertake field tests to further analyse and research the areas above mentioned before making general recommendations on the future use of this material for long-life surfacing. Main findings regarding high-performance cementitious material High-performance cementitious material (HPCM) is an innovative product developed by the Working Group and tested for road surfacing applications for the first time during this project. The initial mix design was improved during the project. It evolved through a number of stages which included: selection of constituents, mix design and laboratory application processes and assessment of behaviour. It was assessed against critical properties such as: strength and abrasion resistance, E-modulus, coefficient of thermal expansion, bond with the bituminous substrate, cracking behaviour, skid resistance and durability in harsh environment. Overall, the thickness of the fibre-reinforced mortar layer needed to be minimised for cost reasons. At the same time, it needed to be thick enough to allow for good penetration of the chippings in the fresh mortar. The test programme focussed on the following performance issues: •

general physical properties of HPCM particularly in regard to bond to substrate and capability to establish a lasting bonding of chippings to the matrix

•

ductility and fatigue properties

•

durability under environmental impact

•

surface properties, noise and skid resistance.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

18 – 1. INCREASING THE LONGEVITY OF WEARING COURSES

By comparison with epoxy asphalt, the HPCM solution needed more development, including operational laying techniques, before being ready for commercial introduction as a long-life surfacing. However, the tests undertaken in Phase 2 indicated there is a high probability that the current uncertainties about HPCM applications will be overcome. Figure 1.1. Principle of the HPCM pavement

Source: Laboratoire Central des Ponts et Chaussées. A number of issues were identified for future research and testing, including: •

Effect of water dosage on HPCM properties. The water dosage has a significant impact on mortar engineering properties, such as: ease of mixing (at industrial scale) and workability; chippings loss, and; bond with the asphalt.

•

Industrial application technology. The adaptation of existing equipment or the practical development of new pavement laying equipment was a high priority for the Phase 3 field testing.

•

Two-dimensional cracking tendency. The test pad chosen for testing two-dimensional cracking tendency needs to be fully representative of a real pavement and laid on a sufficiently stiff asphalt material.

•

Production of HPCM. Production is seen as a manageable process using existing know-how and equipment. However, some modification of existing equipment or development of new equipment will be required for laying the HPCM mortar and inserting the chippings. Construction factors that are important include the availability of constituent materials, the mixing process and the workability of the freshly mixed material. The application of the chippings should ideally take place immediately after placing the thin mortar layer, i.e. with the same machine or with a chip spreader. A light rolling or tamping action is required to ensure the desired embedment of the chippings and a flat, even running surface.

Based on the test results, Phase 2 concluded that if the HPCM layer performs well for the first one-two years, then it is unlikely to fail in the following years. It is the expectation that this surface, based on further trials, can be developed into a final product characterised by high safety, comfort, durability and limited noise emission. Because HPCM is a full innovative concept, further work and research is needed to test in real conditions, in particular the production process.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

1. INCREASING THE LONGEVITY OF WEARING COURSES – 19

Objectives of the field trials (Phase 3) and working method The research in Phase 2 provided comprehensive results from laboratory testing and trials in various accelerated pavement testing machines. The expectations for the durability and long-life capabilities of the materials were based on extrapolations of observations made during this testing. The innovation process naturally requires that the materials be tested in larger scale under real traffic and environmental conditions. In particular, the aims of the field tests were to: •

confirm the performance of the two materials under real traffic and environmental conditions

•

develop construction methods

•

improve cost estimates

•

optimise material mixes

•

increase contractor experience levels.

A Working Group composed of experts from 11 countries was in charge of co-ordinating field trials undertaken by voluntary countries. Although interest in the construction and monitoring of trial sections in a joint research effort was initially expressed by several countries, a number of factors, including the impact of the post-2008 global financial crisis, made locating funding and stimulating an appetite to trial new materials extremely difficult. Nevertheless, five road sections were built using the innovative materials. Sites using epoxy asphalt: •

New Zealand: two 60 m long sections in Christchurch, built in 2007 (yet during Phase 2 of the project).

•

New Zealand: three 210 m long sections in Christchurch, built in 2012.

•

The United Kingdom: a 110 m long section near Truro (in Cornwall), built in 2012.

•

Sites using HPCM:

•

France: a roundabout near Le Mans, built in 2010.

•

France: a 150 m section, near Nantes, built in 2011.

In addition, some further laboratory testing on HPCM was also conducted in the United Kingdom and in Belgium to further develop the HPCM mix and the pumping equipment; and tests were also conducted in Germany and France on epoxy asphalt.

Monitoring of the trial sections The five trial sections were regularly monitored. On each site, an idealised set of data and information to be collected before, during and after the construction phase was agreed (Table 1.1), although individual countries were free to interpret these requirements as appropriate for their local situations.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

20 – 1. INCREASING THE LONGEVITY OF WEARING COURSES

Table 1.1. Monitoring requirements before, during and after the construction of the trial sections Preparation of the trial

Construction phase

• • • • • • • •

Immediately after the construction and before opening to traffic

• • • • •

Yearly monitoring

• • • • •

Specific monitoring regarding epoxy asphalt

• • •

Specific monitoring regarding HPCM

• • • •

Visual condition survey Structural assessment Evenness of the base course Traffic assessment Temperature and weather conditions Material QC data (mix-design, material properties) Site schedule (production and laying rates) Unexpected events (problem identification and management) Visual condition survey (including ravelling/aggregate loss and cracking) Profile (longitudinal/transverse), IRI, rutting Texture depth (EN 13036-1) Skid resistance (e.g. portable skid resistance tester according to EN 13036-4 and/or measurement vehicle) Noise measurement (e.g. SPB according to ISO Standard 11819-1 or CPX according to ISO/CD 11819-2) Visual condition survey (including ravelling/aggregate loss and cracking) Profile (longitudinal/transverse), IRI, rutting Texture depth (EN 13036-1) Skid resistance (e.g. portable skid resistance tester according to EN 13036-4 and/or measurement vehicle) Noise measurement (e.g. SPB according to ISO Standard 11819-1 or CPX according to ISO/CD 11819-2) Traffic assessment Monitoring of curing (such as by stability or stiffness) Removal of 4-10 cores and determination of mechanical properties (stiffness) Assessment of HPCM thickness at the fresh state (during trial) Photo documentation of aggregate spreading procedure and result Removal of a few cores for documentation of layer thickness Special attention during visual condition surveys to possible delamination and end joint problems

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

1. INCREASING THE LONGEVITY OF WEARING COURSES – 21

Content of the report This report describes each field test and summarises the findings. It is composed of the following chapters: •

Chapter 2 describes the experiences with the epoxy asphalt sites in New Zealand and the United Kingdom.

•

Chapter 3 describes the experiences with the pad tests in the United Kingdom and the two HPCM sites in France.

•

Chapter 4 presents the general conclusions from the trials and the recommendations from the Working Group on the potential of long-life pavement surfacing.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

22 – 1. INCREASING THE LONGEVITY OF WEARING COURSES

References ECMT (2005), Economic Evaluation of Long-Life Pavements: Phase 1, OECD Publishing, Paris. DOI: http://dx.doi.org/10.1787/9789264008588-en ITF (2013), Spending on Transport Infrastructure 1995 2011: Trends, Policies, Data, http://www.itfoecd.org/sites/default/files/docs/13spendingtrends.pdf. ITF (2008), Long-Life Surfaces for Busy Roads, OECD Publishing, Paris. DOI: http://dx.doi.org/10.1787/9789282101209-en

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 23

Chapter 2. Epoxy-asphalt road surfacing field trials

This chapter describes the findings of the third phase of this project (i.e. the field trials) in respect to the epoxy-asphalt materials tested. Three test sections using epoxy asphalt were built and monitored: two in epoxy-modified open graded porous asphalt, in New-Zealand, and the third one in stone-mastic asphalt, in the United Kingdom. No significant difficulties were encountered during the construction phase, and the two pavements have exhibited to date satisfactory behaviour under service.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

24 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

The overall objective of Phase 3 was to demonstrate that the performance envisaged from the laboratory and accelerated testing would hold true within the period of the trial under real traffic and environmental conditions. Additional aims were to develop construction methods, optimise mixtures, and increase contractor experience. As in Phase 2, the research approach taken in the epoxy-asphalt group was to have participating organisations identify and utilise local sources of epoxy asphalt and compare the characteristics and performance results with those of a standard high quality reference material. In conducting the comparison, the epoxy-asphalt binder would replace the bitumen in the reference material and other aspects of the selected system would remain the same. Adopting this approach allowed the participating countries to focus on their likely application of epoxy asphalt using local materials and standard test procedures, while providing a qualitative assessment of actual field performance. Epoxy-asphalt surfacing systems are not new. Pioneering work was done by Shell in the 1950s, initially to develop surfaces to resist jet fuel damage and as a heavy duty surfacing in industrial areas. Epoxy asphalt was also used at the time for a limited number of highway applications, but it was particularly adopted as a thin, stiff surfacing on steel bridge decks, where the material has been used on a number of major bridges around the world. Excellent performance has been recorded, most notably on the San Mateo-Hayward Bridge, where epoxy-asphalt surfacing has been in service for more than 45 years without failure (Lu et al., 2012). Due to its superior resistance to aircraft fuel and jet-blast, the material has also been used on a number of military airfields in the United States (Simpson et al., 1960). Prior to the current project, the main use of epoxy asphalt in the United Kingdom in recent times had been on a limited number of steel bridge decks (Erskine and Humber), where the design was based on hot rolled asphalt. To the authors’ knowledge, except as mentioned above, epoxy-asphalt surface course has not been used to any significant degree for highway surfacing. However, as part of the current project, successful trials of epoxy-modified materials have been carried out on live road sites in New Zealand and the United Kingdom and initial indications from these trials are very encouraging. Interest in participating in a joint research effort was initially expressed by national institutions from nine countries: France, Germany, Israel, New Zealand, Norway, South Africa, Spain, United Kingdom and the United States. Unfortunately, a number of factors, including the impact of the post-2008 global financial crisis, made locating funding and stimulating an appetite to trial new materials extremely difficult. Therefore, only France, New Zealand and the United Kingdom were able to play an active role in the field trials of the epoxy-asphalt material. In the event, for a number of practical reasons, France was able to complete plant trials only. However, as part of the current project, successful trials of epoxy modified open graded porous asphalt were carried out in New Zealand (Herrington, 2010), and of epoxy-modified stone mastic asphalt in the United Kingdom.

Plant trials of epoxy asphalt in France In 2009, France selected a site near Le Mans and carried out laboratory trials of typical French mixtures manufactured with an amine curing type of epoxy modified bitumen, a proprietary product supplied by the contractor. The binder was described as 20% epoxy (component A) and 80% bitumen (component B) of 80-100 penetration grade. Seven binders were investigated in order to try to identify a very slow (amine) curing epoxy. This work revealed that the curing time of the premixed components (A+B) was critically dependent on mix temperature, and the “workability window” reduced from four hours at a mix temperature of 120°C to 1.5 hours at a mix temperature of 140°C. The contractor planned to produce the epoxy modified material through a turbulent-mass drum mix plant (continuous type). However, it proved difficult to ensure that precise quantities of each component were added using this type of mixing process and that the temperature within the mixing drum were maintained at 120°C. Cleaning the epoxy binder distribution system was also an issue. Although the use

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 25

of a batch mixing plant was envisaged, the trials were finally aborted due to these technical risk concerns.

Field trials of epoxy asphalt open graded porous asphalt in New Zealand Research outline The New Zealand Transport Agency’s (NZTA) contribution to the research focused on the potential benefits of epoxy-modified open-graded porous asphalt (EMOGPA), specifically to undertake the construction of a full-scale road trial and monitor its performance. A summary of this work is provided below; full details have been provided by Herrington (2010). Although the safety and noise-reduction properties of open-graded porous asphalt (OGPA) are well documented, binder oxidation is a major problem and is the principal factor governing the ultimate life of porous asphalt. Because of the very open nature of the material, oxidation and consequent binder embrittlement are more rapid than in conventional mixtures. Oxidation ultimately leads to failure of the mixture through loss of material from the surface (ravelling or fretting) under traffic-shearing stresses. The result is a reduced life compared with denser mixtures, adversely affecting their cost-benefit ratios and thus inhibiting the more widespread use of this safe and environmentally friendly surfacing. EMOGPA appears to offer the potential for a significant improvement in life for open-graded surfacing. EMOGPA uses the same mixture designs as conventional OGPA, but the bitumen component is replaced with a bituminous binder incorporating a reactive epoxy resin and curing agent.

Design The road trial was laid on the outer north-bound lane of Main North Road (part of State Highway 1) at Belfast in Christchurch on 5 December 2007. The trial site consisted of three sections. A standard PA 14 OGPA, meeting NZTA Specification P/11 (NTZA, 2007) was used as a control and a 20% air voids content EMOGPA and a 30% high air voids content EMOGPA were laid. The 30% air voids content epoxy asphalt represented an attempt to match the highest air voids content achieved with polymer-modified conventional OGPAs – such 30% air voids mixtures have been noted as having particularly good acoustic performance. The site had an unbound granular pavement foundation, with 15 850 vehicles per day heading north, of which 6% were heavy commercial vehicles. The three sections were contiguous, each 60 metres long, 5 metres wide and 30–35 mm thick. Looking north, the order of the sections was 20% EMOGPA, 30% EMOGPA and 20% control OGPA. The trial site was in the outer of the two lanes heading from the city, where an existing OGPA laid in 1992 was being replaced because of fretting. The inner lane was laid using standard 20% OGPA. Falling weight deflectometer (FWD) testing suggested that the site was structurally sound. The existing OGPA surface of the site was milled out and a grade 5 (10 mm) chipseal was constructed over the remaining thin asphaltic concrete surfacing.

Materials The epoxy bitumen used was supplied by ChemCo Systems Ltd of California. This is a two-part product that is blended just before use (Figure 2.1). Part A (used at 14.6% by weight) consists of an epoxy resin formed from epichlorhydrin and bisphenol-A. Part B type V (85.4%) consists of a fatty acid curing agent in an approximately 70 penetration grade bitumen. The bitumen used for all control mixtures was an 80–100 penetration grade bitumen, manufactured from Middle Eastern crudes,

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

26 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

comprising both air-blown and butane-precipitated material, and conforming to the NZTA M/1:2011 specification (NTZA, 2011). This bitumen is commonly used in OGPA surfacing in New Zealand. Figure 2.1. Two components of epoxy-asphalt binder

Source: Turner-Fairbank Highway Research Center, USA.

The mixture designs (grading, aggregate source and binder content) were nominally the same as those used in the laboratory work reported in Phase 2. Compaction of 100 mm diameter specimens for testing was by Marshall Hammer (75 blows per side) and carried out at the Fulton Hogan Ltd laboratory in Christchurch, according to ASTM D6926 (ASTM, 2004). Production testing of the mixtures gave the results shown in Table 2.1. Table 2.1. Mixture design for field trial Test section

Passing (%) sieve size (mm) 0.075

Bitumen content (%)

Air voids content (%)

13.2

9.5

4.75

2.36

20% air voids OGPA

100

94

31

19

2

5.0

17.3

20% air voids EMOGPA

100

94

24

15

3

5.3

20.6

30% air voids EMOGPA

100

94

8

3

1

4.0

31.3

The target bitumen content in each case was 5%; however, production testing for bitumen contents gave 5.0%, 5.3% and 4.0% for the 20% OGPA, 20% EMOGPA and 30% EMOGPA, respectively. The low binder content of the 30% air voids content site was due to significant binder drain-down that occurred in the asphalt plant which had not been apparent in laboratory-scale work.

Manufacture and construction A turbulent-mass continuous-mix drum plant was used to manufacture the EMOGPA. An in-line blending system was used to introduce the epoxy binder. Epoxy component A, heated to 85°C, entered the line carrying component B (120°C) about 4 metres from the point of discharge into the drum to LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 27

provide premixing of the two components. Ordinary positive displacement gear pumps fitted with electronic mass flow meters were used. The flow metres reported to the plant control system controlling the pumps. The epoxy-asphalt mixtures were held at 125°C for 45 minutes before compaction. The epoxy mixtures were manufactured first; the first four-five tonnes of the control mixture that followed was run through and then discarded, in order to clean the plant. When the manufacture of the asphalt was complete, the pumps and lines used to introduce the epoxy bitumen components were disconnected from the plant and drained and flushed with bitumen and kerosene. Manufacture of the epoxy mixture was found to be straightforward and completed without difficulty, except for the unanticipated drain-down of the 30% EMOGPA. Figure 2.2. General view of the site

Source: NZTA (New Zealand).

Construction and compaction by a standard tandem steel-wheel vibratory roller required about 20−30 minutes for each section. Temperatures during compaction were 55−70°C for the epoxy mixtures. The total time from the commencement of the manufacture of the first epoxy mixture to the commencement of construction was 45−60 minutes. The mixture was manufactured at a temperature of 117°C and 122°C for the 30% and 20% sites respectively. There were some problems in compaction of the EMOGPA, particularly the 30% material, because of concern that the epoxy might cure before compaction occurred. Although the initial viscosity of the epoxy binder is somewhat lower than that of 80−100 bitumen at the same temperature, when excessive curing occurs, the epoxy bitumen becomes “dry” and is not adhesive. To increase the working-time, the period for which the mixture was held at high temperature was kept to a minimum. Although this time was similar to that used in the laboratory work (45 minutes), a longer mixing time at high temperature in the plant, or a higher plant temperature, would have been desirable to increase the viscosity of the binder. As a result, there was some initial “pick-up” on the roller, and the road surface had to be cooled with water before traffic was allowed on it that afternoon. The mixture at the 30% site was still “lively” some

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

28 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

three hours after compaction was complete. Other than this, the behaviour and appearance of the EMOGPAs was indistinguishable from that of the control material. No unusual fuming or smell was noted, as was also the case in the earlier CAPTIF trial (Herrington et al., 2007). Figures 2.2−2.5 show aspects of the trial construction. The trial sections were opened to traffic that afternoon. Figure 2.3. Start of the 20% EMOGPA section

Source: NZTA (New Zealand).

Figure 2.4. Compaction of 20%EMOGPA

Source: NZTA (New Zealand).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 29

Figure 2.5. Traffic damage to 30% EMOGPA (plucked chip outside wheel tracks)

Source: NZTA (New Zealand).

Curing of the EMOGPA The rate of curing of the EMOGPA was monitored by measuring the indirect tensile modulus of 100 mm diameter cylindrical blocks (in triplicate) prepared at the time of construction and placed outside at the Fulton Hogan Ltd laboratory (Pound Road, Christchurch), close to the trial site. These blocks were wrapped in silicone release paper to help prevent deformation of the blocks, and were tested periodically to determine the increase in modulus. The modulus measurements were conducted on a 5 KN test frame at 25°C, according to AS2891.13.1 (Standards Australia, 1995). Data for the 20% air voids content EMOGPA (shown in Figure 2.6) suggested that it had cured rapidly over the first 30 days and the modulus was still increasing slowly beyond that. The control modulus increased over the first summer but changed little thereafter. The 30% air voids content EMOGPA modulus appeared to be unchanged. The very open structure and low bitumen content meant these samples were very fragile, and it is not certain that they were representative of the in situ material. The results for the 30% EMOGPA were also in contrast to those of the 25°C curing experiments carried out on the laboratory-prepared mixture, where a small increase was observed over 140 days at 25°C. This could have been because the average road temperature was below 25°C, or more likely, the lower binder content (4%) compared with the 5% of the laboratory mixtures. The curing rate of the isolated blocks was likely to have been lower than that of the material on the road, where the greater mass could have acted to produce higher overall average temperatures. Road temperatures were not measured directly; however, air temperature data from a weather station near the site (Christchurch Airport) are given in Table 2.2.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

30 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

Figure 2.6. Curing of field trial specimens

Table 2.2. Air temperatures near to trial site (December 2007–March 2010) Parameter

Temperature (°C)

Average daily maximum

17.5

Average daily minimum

6.6

Assessment, testing and monitoring The sites were monitored visually and vehicle noise was measured over three years. Measurements were also made for rutting, skid resistance and water permeability every 10 metres, both inside and outside the wheel tracks (except for rutting). The general appearance of the site after three years is shown in Figures 2.7 and 2.8. The surfaces were found to be in good condition, with the exception of small patches at the end of each of the EMOGPA sections (e.g. lower edge of Figure 2.7), where some ravelling had occurred. These patches corresponded to locations where the paving machine sat for long periods waiting for new material to lay, which appears to have caused damage. Discolouration in these patches was apparent immediately after construction.

Rutting The 2-metre straight-edge rutting results in Table 2.3 show minimal rutting. The results suggest that the epoxy sections were at least as strong as the control in early life, even without significant curing.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 31

Table 2.3. Rutting Mean rut depth (mm) ± 95% CL

Section

January 2008 4 ±1

Control

March 2010

January 2009 3 ±1

4 ±2

20 % EMOGPA

2 ±1

2 ±1

1 ±1

30 % EMOGPA

4 ±1

2 ±1

1 ±1

Skid resistance Skid resistance, measured using the British Pendulum Tester (according to NZTA draft method T/2:2000), was comparable for all sites (Table 2.4), demonstrating that the epoxy surface was not inherently more slippery than conventional OGPA. The results showed that skid resistance in the wheel tracks had dropped slightly compared with un-trafficked areas, but the effect was comparable for all three sections. The skid resistance of all sites appeared to have increased after one year, which may have been due to wear of bitumen from the aggregate surface. However, comparison of skid resistance measurements made over such long periods may be unreliable and can be strongly influenced by, for example, weather over the days or weeks preceding the measurement. It is planned to obtain high-speed SCRIM data on the sites to confirm the skid resistance results. Table 2.4. Skid resistance (British Pendulum Number) British Pendulum Number ± 95% CL Section

January 2008 Wheel tracks

Outside wheel tracks

January 2009

March 2010

Wheel tracks

Outside wheel tracks

Wheel tracks

Outside wheel tracks

Control

53 ±2

51 ±2

59 ±2

63 ±3

50 ±2

55 ±5

20 % EMOGPA

50 ±2

45 ±5

52 ±2

56 ±3

48 ±2

56 ±6

30 % EMOGPA

53 ±2

49 ±3

57 ±2

61 ±3

49 ±2

57 ±3

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

32 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

Figure 2.7. Start of 30% air voids content EMOGPA section, looking towards 20% air voids content OGPA section

Source: NZTA (New Zealand).

Figure 2.8. Start of 20% air voids content EMOGPA section, looking towards 30% air voids content EMOGPA section

Source: NZTA (New Zealand).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 33

Water permeability The results from field water-drainage tests (using an in-house method based on NZTA P23:2005 specification,) detailed in Table 2.5, show that drainage times had increased, with the exception of the 30% air voids content material, which showed relatively little change and was the most free-draining of the three sites. Data for 2010 for the control and 20% EMOGPA sites was an underestimate, as several readings were, in each case, greater than 100 seconds. Measuring longer drainage times was not practicable with the method that was used. Table 2.5. Water permeability Mean water permeability (seconds) ± 95% CL January 2008 Section

Wheel Tracks

Outside Wheel Tracks

January 2009

March 2010

Wheel Tracks

Outside Wheel Tracks

Wheel Tracks

Outside Wheel Tracks

Control

22 ±13

20 ±14

33 ±13

34 ±28

>65*

>41*

20 % EMOGPA

17 ±5

12 ±7

36 ±19

23 ±6

>64*

>75*

30 % EMOGPA

14 ±7

5 ±2

15 ±6

6 ±3

23 ±7

21 ±12

* Lower bound of real result – see text.

Figure 2.9. Mean noise levels for cars, dual-axle and multi-axle trucks (SPBI =Statistical Pass By Index (according to ISO 11819-1))

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

34 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

Noise The results of noise measurements are shown in Figure 2.9. These show that all three sites generated similar levels of tyre noise. Surprisingly, the 30% air voids content EMOGPA did not produce a noticeably quieter surface. This may have been because of the problems associated with “pick-up” of the material on the roller during compaction and early trafficking, which may have led to more surface texture than would normally be expected, and hence higher noise levels, cancelling out the benefits of the higher percentage voids. The noise level had increased for all sites (between 2008 and 2009), consistent with the reduced water permeability measurements. It had rained heavily the night before the January 2009 measurements were taken, and although the surface was dry, water may have still been present in the voids and affected the result. It is interesting to note that there was no comparable increase in noise level between 2009 and 2010; more measurements are needed to properly assess trends.

Laboratory work During the same period that the field trial was being constructed and monitored, a programme of complimentary laboratory work was undertaken to investigate the curing behaviour and durability properties of EMOGPA, and to examine the effects of diluting the epoxy binder composition as a way of reducing costs. A brief summary of this work is provided below. Open-graded porous asphalt (OGPA) specimens were treated in an oven at 85°C for up to 171 days, resulting in oxidation equivalent to approximately 20 years in the field. Although the moduli (25°C) of both materials increased with oxidation time, that of the EMOGPA was much more pronounced, reaching 12 000 MPa after 171 days at 85°C, compared with 7 800 MPa for the control. This time period results in oxidation considered to be equivalent to about 20 years in the field (80 days is considered to be equivalent to 12 years in the field). Data from storage in oxygen and nitrogen atmospheres suggested that the greater hardening of the EMOGPA is largely attributable to gradual curing, rather than to oxidation. Despite the very high modulus, the Cantabro Test results showed that oxidation had no significant effect on the abrasion resistance of the EMOGPA. After 171days, the EMOGPA mass loss was within error of the initial value, whereas that of the control had increased by 1.8 times. Similarly the fatigue life of oxidised EMOGPA was markedly greater (more than 25 times) that of the control. Dilution of the epoxy binder to 25% or 50% of the mixture binder composition, using the standard 80-100 penetration grade bitumen used in the control mixture, gave an OGPA mix with properties inferior to that of the undiluted material, but still markedly superior to conventional OGPA in terms of abrasion resistance after oxidation. Cantabro Test losses for the 25% and 50% mixtures increased only 1.4 times after 171 days oxidation. The fatigue life of the oxidised 25% and 50% EMOGPA mixtures were equivalent to that of the control. For clarity, and as an example, the 50% EMOGPA mixture comprised 14.6 parts of epoxy component A, 85.4 parts of epoxy component B and 100 parts of 80-100 penetration grade bitumen. It was noted that the usefulness of this approach may depend on the selection of the appropriate bitumen to ensure compatibility with the epoxy components, and may not be successful with all bitumens.

Field trials of stone mastic asphalt and epoxy asphalt in the United Kingdom In the United Kingdom, the Highways Agency (HA), an executive agency of the Department for Transport with responsibility for operating, maintaining and improving the Strategic Road Network in England, provided support to Scott Wilson (now URS) to assist in the organisation of the full scale

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 35

epoxy-asphalt trial. The scope of the support also encompassed monitoring performance over time, and relevant in situ and laboratory tests, the latter on cores extracted from the trial section. The reference material selected for the trial, as in the earlier phases of the project, was stone mastic asphalt (SMA). SMA has had a track record of successful use in the United Kingdom for the last 20 years, particularly as a proprietary Thin Surface Course System. It was intended that the epoxy asphalt would be laid as a 30 mm or 40 mm thick SMA surface course as a like-for-like and cost-neutral replacement for the originally specified material; the benefit to the road owner permitting the trial was that its life would be expected to be at least as long as that of the traditional material. Although EAPA quotes lives ranging from 16 to 25 years where SMA is used on secondary roads (EAPA, 2007), the Phase 1 study identified an expected life for conventional SMA surface course ranging from 5 to 15 years, depending particularly on incidence of studded tyre use and level of traffic. Ten years was selected as the average expected life to be used for the economic evaluation for heavily trafficked roads (ECMT, 2005), and this ties in reasonably well with UK experience.

Logistics and design considerations The UK Trial Site is located in Threemilestone, on Lane 1 (L1) of the A390 inbound carriageway towards Truro, in the Southwest of England. The trial encompasses an area approximately 110 metres long by 3.65 metres wide where approximately 40 tonnes of 10 mm nominal size epoxy-asphalt SMA trial material was laid 30 mm thick; a comparable length and volume of control material was laid immediately afterwards. The traffic data for this section of the A390, factored up to January 2012 from a 2006 traffic count, indicate an AADT of 29 100 vehicles, with 830 (3%) heavy class vehicles (HCV) of gross weight > 7 500kg and 1 100 (4%) medium class vehicles (MCV) of gross weight 3 500–7 500 kg. The pavement structure is flexible with a full depth asphalt construction ranging in thickness from 312 to 390 mm. The surfacing was due for replacement in 2012 due to surface deterioration (predominantly crazing and fretting, with poor profile due to patch reinstatement) but the substrate was considered to be sound. The logistical, practical and procurement issues were resolved in two meetings involving URS, the supplier (Colas) and the local authority (Cornwall Council, CC) held on 11 January and 30 March 2012, and the material sourcing and production, by Colas, and laying of the road trial, by CORMAC, was successfully completed on 28 April 2012.

Materials and mixture design As noted above, the mixture design developed for this project was based on a generic surface course system that is widely used for the surfacing of major road networks around the world, namely stone mastic asphalt (SMA). Essentially, the design of the epoxy-asphalt SMA material was similar to that of conventional SMA, with the substitution of a slightly increased quantity of the epoxy binder components, in place of the standard binder. Full details have been reported elsewhere (ITF, 2008; Elliot et al., 2008). As the specific requirements of the site necessitated a 30 mm thick surfacing layer, it was decided that 10 mm nominal maximum aggregate size was more appropriate, rather than 14 mm as used in the Phase 2 work. Two mixture designs were therefore required, a 10 mm nominal size epoxy-asphalt SMA and a 10 mm nominal size control SMA. The specific mixture designs for the Truro trial were carried out by Colas Product Development Department. Both mixtures used PSV65 aggregate and had nominally identical compositions (Figure 2.10) apart from the binder type and content; in particular, the binder content of the epoxy asphalt and control material was 6.9% and 6.6%, respectively. Although the epoxy

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

36 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

binder was sourced by Colas, it is understood to be practically identical to the ChemCo product used in New Zealand. Figure 2.10. Composition of SMA mixture

Manufacture and construction Plant trials were carried out using a discontinuous mixer at Colas’ Carnsew Quarry, in particular to resolve any difficulties with handling the material in the plant. For the purposes of the trial, a bitumen distributor with the facility to heat and pump the two materials was mobilised, and suitable piping was sourced to enable the materials to be pumped from ground level up to the mixing pugmill of the Carnsew batch plant. Large scale production of epoxy asphalt would necessitate automated blending of the epoxy-asphalt binder, and, for example, ChemCo in the United States use an automatic “meter/mix” machine to premix the two components and inject the proper amount for each batch into the pug-mill. The liquid components are stored in heated tanks from which the components are drawn by the meter/mix machine. The pre-mixer is cleaned after each batch by blowing compressed air and cleaning agent (Gaul, 2011). However, in this case, the small scale of the trial precluded the manufacture of a bespoke and expensive bitumen delivery system and some ingenuity was therefore required in order to achieve the desired end result. Component B was piped from the bitumen distributor through the pump to the bitumen “kettle” for weighing directly into the mixer pugmill; component A was piped from the bitumen distributor through the pump to a three-way valve to allow weighing into containers on a balance and subsequent manual introduction into the mixing pugmill. At the same time as these plant modifications were being made, quality control testing of the mixed epoxy-asphalt product was underway. Following the successful completion of the Plant and Quality Trials on 26 April 2012, the decision was made to proceed to full trial on 28 April 2012. LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 37

Colas supplied both the epoxy-asphalt trial and control materials out of Carnsew Quarry, and transported the material to site (a haul distance of around 30-45 minutes). Figure 2.11. General view of plant setup (bitumen delivery system on left)

Source: URS Infrastructure & Environment UK Limited.

The material was then handed over to Cornwall Council Direct Labour Organisation (CORMAC), who laid the material using its own plant and labour. The weather during the trial was dry, mainly overcast but with sunny intervals. After planning out the original surfacing (Figure 2.12 for view of typical pre-trial condition) and before laying the trial and control material, Colbond 50 polymer modified bond coat was applied to the planed surface at a nominal spread rate of 0.7 litres/m2 (equivalent to 0.35 kg residual binder/m2) and left until clearly “broken” before laying the new surfacing.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

38 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

Figure 2.12. Failed patch reinstatement typical of pre-trial condition

Source: Cornwall Council.

Although epoxy modified asphalt material may be handled in a similar way to other asphalt, time and temperature are critical to the successful laying and compaction of the material. Colas therefore provided a representative on site to supervise the laying operation, in particular to provide the laying gang with a timed schedule for laying and compacting the material. This included providing an indication of the maximum usable life of each batch of material, dependant on mixture temperature, as shown in Table 2.6, and the resulting ‘must lay after” and “must be substantially compacted before” times, and controlling the speed of the paver to ensure the optimum mixture curing time. Before the day of the trial, a “Toolbox Talk” was given by Colas to the laying gang so that everybody was clear about the special requirements. Colas also provided a specification for materials, transport, laying and compaction of the epoxy asphalt, which sets out the requirements for mixing and delivery of the material. In particular, the temperature of binder component A is maintained between 80°C and 100°C prior to and during addition to the mixer, and that of binder component B is maintained between 120°C and 135°C prior to and during addition to the mixer, and during storage. The asphalt is mixed for a minimum of two minutes from the time part A is introduced into the mixer, and the minimum temperature of the completed mixture is 110°C (100°C during transportation). The adopted procedure for laying and compaction of epoxy asphalt was generally in accordance with standard procedures for asphalt as set out in Volume 1 of the UK Manual of Contract Documents for Highway Works (MCHW1) (Department of Transport, n.d.), although the temperature of the asphalt leaving the paver must not be less than 90°C, and not less than 70°C at the commencement of rolling. The transport, paving and compaction plant were the same as used for conventional material and, visually, the epoxy-asphalt material looked and behaved like a traditional SMA (Figure 2.13).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 39

Table 2.6. Mixing temperature and maximum usable life of epoxy asphalt Mixing temperature

Maximum usable life

Maximum usable life

(°C)

(minutes)

(hours)

*Must be laid by these times

*Must be laid by these times

110

180

3:00

115

150

2:30

120

120

2:00

125

100

1:40

130

85

1:25

135

75

1:15

Figure 2.13. Paving of epoxy asphalt with conventional plant

Source: URS Infrastructure & Environment UK Limited.

The only difficulty observed related to the paving of the fourth load, which was delayed in traffic between the quarry and the site. While waiting for this load to be delivered, the paver pulled back around 0.5 m from the fresh edge of the asphalt and stopped moving forward; the fourth load duly arrived, the paver re-connected with the fresh edge of the asphalt and the laying was completed without further incident. However, after completion of compaction, the visual appearance of the asphalt where the paver LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

40 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

had stopped showed some surface blemishes and the profile appeared to be less consistent than the remainder of the laid material. It is not known whether a similar issue would have occurred with the control material, as the paving here proceeded uninterrupted. Laying of the epoxy asphalt was completed within 60 minutes, and the control material followed on directly.

Curing and opening to traffic The early life traffic ability of epoxy modified SMA is the subject of on-going research, but, for the trial, L1 was closed to traffic until Monday night (30 April 2012), when the Traffic Management was lifted. In order to monitor the quality control and “strength” development of the manufactured product, and so that they could advise when it was appropriate to open to traffic, Colas produced specimens for Marshall testing, to BS EN 12697-34 (EN, n.d.. The requirements were a minimum stability and maximum flow of 5 KN and 4 mm, respectively, after 24 hours curing at ambient temperature, and 30 KN and 7 mm, respectively, after 20 hours curing at 120°C. Provisions were in place such that, in extreme circumstances, it would have been possible to extend the TM window, although this extra time was not needed.

Assessment, testing and monitoring Scope In order to monitor the visual condition and properties of the field trial, two visits were made to the site: (i) an initial monitoring visit on the night of 09 May 2012, 11 days after the material was laid and nine days after the trial section was re-opened to traffic, and; (ii) a final, 12-month visit on 30 April 2013. On both occasions, the monitoring work was done under a partial lane closure from late evening onwards, with L1 closed to traffic over the trial and control sections, and all inbound Truro traffic using L2. The work carried out comprised the following: •

photographic record of visual condition

•

longitudinal profile, in accordance with Volume 1 of the Manual of Contract Documents for Highways Works (MCHW1) Clause 702, Table 7/2 (Department of Transport, n.d.)

•

surface macro-texture by volumetric patch method as described in BS EN 13036-1 and BS 594987, Clause 8.2, (with additional tests across full width of lane)

•

falling Weight Deflectometer (FWD) testing, broadly to the requirements of HD 29/08 (Volume 7 of the UK Design Manual for Roads and Bridges (DMRB7); Department for Transport, 2008)

•

coring to establish depth of bound material and to recover samples for materials testing.

In the event, the weather during the initial visit was very poor (heavy rain), and the initial surface macro-texture could not be determined. A noise assessment was also planned but local topography and proximity of dwellings made this impractical.

Visual and surface characteristics Visually, the surface appearance of the epoxy-asphalt test section was indistinguishable from that of the control section (although see below) and from conventional SMA; in addition there was no discernible change in appearance between the initial and 12 month visits. The longitudinal profile complied with the requirements of MCHW1 Clause 702, Table 7/2(Department of Transport, n.d.), and the results from the two visits were essentially similar, indicating no change with time; two 7 mm LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 41

irregularities found in the epoxy-asphalt test section were on the joints at the beginning (0 m) and end (110 m) of the section respectively, and are a reflection of the site situation rather than a failure to achieve a consistent profile. These surface irregularities at the beginning and end of the test section can be detected when travelling over the test section by car, as can the location (around Chainage 80) where the surface profile was influenced by the paver coming to a halt between the third and fourth loads of epoxy-asphalt material (as detailed above). However, in the latter case, any impact on ride quality was not reflected in the surface regularity assessment. The surface macro-texture data at 12 months show that the mean texture depth for the trial and control sections (set of 10 measurements) were both 1.1 mm. These values comply with the current UK requirements (Department of Transport, n.d.) for thin surface course systems for lower speed roads (≥ 1.0 mm (average per 1000 m section); ≥ 0.9 mm (average per set of 10 measurements)). In addition to this information, routine Surface Condition Assessment of the National Network of Roads (SCANNER) measurements, traffic-speed surveys of the network0 to determine rut depth, texture, profile and cracking, carried out according to the User Guide and Specifications published by the UK Roads Board (2011) (Department of Transport, 2013), and; Sideway-force Coefficient Routine Investigation Machine (SCRIM) measurements in-service skid resistance , carried out according to HD 28/04 of DMRB7 (Department of Transport, 2004) reveal that the surface has so far not shown any indication of change since it was opened to traffic. A summary of the skid resistance, texture and rut depth data from these surveys is reproduced in Table 2.7 below. Table 2.7. Summary of skid resistance, texture and rut depth data MSSC Material

2012

Rut depth (mm)

2013

2012

Macro-texture (mm)

2013

2012

2013

LHS

RHS

LHS

RHS

LHS

RHS

LHS

RHS

Epoxy

0.49

0.57

3

3

3

3

1.1

1.0

0.9

0.9

Control

0.52

0.55

3

3

5

3

1.0

0.9

0.9

0.8

Notes: MSSC = Mean Summer SCRIM Coefficient, as described in HD 28/04 (DMRB7); LHS, RHS = Left hand side, right hand side.

Personal feedback from Cornwall Council staff who drive this section of road frequently is that nothing registers apart from the section where the paver stood for a period (see above), and even here there has been no sign of change since the road was opened to traffic. Figure 2.14 shows the condition of the epoxy-asphalt trial section (Lane 1 in the photograph below) after 12 months trafficking.

Falling weight deflectometer testing The falling weight deflectometer (FWD) testing was done at 2 m centres in two lines running longitudinally from Chainage 0 to 220 m, in the nearside wheel track (NSWT) and lane centre (LC), at contact pressures of 700, 850 and 1 000 kPa. The NSWT tests were offset by 1 m from the LC tests, and similarly offset from the cores (see below) where applicable. The primary purpose of the FWD testing was to establish that the pavement was structurally competent in both the trial and control areas, and that the trial and control pavements were essentially similar. A detailed summary of the testing is beyond the scope of this chapter. However, in overview, the analysed FWD data show that the asphalt stiffness were generally low throughout (it is possible that this LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

42 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

may in part be a reflection of variable bond between asphalt layers influencing the back analysis) and the foundation stiffness were indicative of good performance throughout. Although there was some individual variation, overall the structural assessment of the site carried out using the FWD confirmed broadly similar performance in the trial and control areas, and no significant change between the initial and final visits. Figure 2.14. Epoxy-asphalt trial section after 12 months trafficking (2013)

Source: URS Infrastructure & Environment UK Limited.

Coring and laboratory testing For both visits, twenty number 100 mm diameter cores were taken, evenly spaced along the centre of L1, ten in the epoxy-asphalt test material and ten in the control material. Selected cores were drilled to full depth, revealing an asphalt construction comprising six to eight discernible layers, typically: •

new SMA surface course, overlying

•

two-three layers of asphalt concrete, overlying

•

one-two layers of hot rolled asphalt, overlying

•

one-two layers of asphalt concrete.

The full depth of asphalt ranged from 312 mm to 390 mm, and was generally found to be slightly thicker in the test section (370-390 mm) than in the control section (312-350 mm). The full depth cores showed some lack of bond in their lower layers; the depth to the first layer showing lack of bond ranged from 111 to 240 mm, measured from the top of the pavement. Generally, the cores revealed a relatively thick asphalt construction encompassing several different material types, typical of a structure that has developed over time as new material has been laid over existing.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 43

In order to provide baseline data on the mechanical and volumetric properties of the trial and control material, and to assess any change in mechanical properties occurring over time, a selection of the extracted cores from each section and from each visit was subjected to a bespoke programme of storage and testing. Thus the cores from the final visit, after 12 months curing in situ, were tested as soon as possible after extraction, while the cores extracted from the initial visit were stored at laboratory ambient temperature (20°C), after extraction from the road, until required for testing. The testing carried out comprised bulk and maximum density, and calculated air voids, to BS EN 12697 Parts 5, 6 and 8 (BS EN 12697-5; -6; -7), Indirect Tensile Stiffness Modulus (ITSM) to BS EN 12697 Part 26 and Indirect Tensile Splitting Strength (ITST) to BS EN 12697 Part 23 (BS EN 12697-26). The indirect tensile stiffness and strength data are important parameters to assess likely pavement material performance over time, and are particularly useful here to illustrate the special properties of epoxy asphalt. A summary of the key mechanical property data from these two tests is presented below. Figure 2.15 shows the mean stiffness data at 0, 20 and 30ºC, for in situ cured cores tested after 13 months (similar results were obtained with laboratory cured cores, tested one month after production). Figure 2.15 shows that the stiffness of epoxy asphalt is higher than that of the control mixture, regardless of the test temperature. At 20°C and 30°C, the stiffness of the epoxy asphalt is significantly higher than that of the control, which is likely to be beneficial with respect to the rutting potential of asphalt mixtures. Indeed, the dramatically improved deformation resistance of epoxy modified mixtures, compared with that of standard material, has been demonstrated in earlier work (Standards Australia, 1995). In addition, the results indicate that there is a strong exponential relationship between stiffness and temperature for both mixtures. Figure 2.15. Mean stiffness (ITSM) data at three temperatures, performed on in situ cured cores, 13 months after production Epoxy

Control

Expon. (Epoxy )

12690

9910

3140

y = 40190e-1.214x R² = 0.9925

1120

1120 420

0°C

Source: URS Infrastructure & Environment UK Limited.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

20°C

30°C

44 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

Figure 2.16. Mean stiffness (ITSM) data at 20°C, performed on laboratory cured cores, at various ages after production 9000 8000 7000

ITSM at 20°C (MPa)

6000 5000 4000 3000 2000 1000 0 Epoxy Control

1 Month 2850 1050

3 Month 5660 1550

6 Month 6220 1770

9 Month 6900 1910

13 Month 8240 1790

Source: URS Infrastructure & Environment UK Limited.

The changes in stiffness (ITSM) over time for the epoxy modified and control mixtures are shown in Figure 2.16, for laboratory cured cores tested at 20ºC. It can be seen that the stiffness of the epoxy-asphalt mixture is significantly greater than the control mixture at all ages. Furthermore, the stiffness of the latter remains at a constant (and relatively low) level between six and 13 months, whereas the stiffness of the epoxy modified mixture continues to rise throughout the period of monitoring. Table 2.8. Mean tensile strength (ITST) data at 20°C at various ages after production Mean ITST (MPa) at 20°C Material

Laboratory cured cores

In situ cured cores

1 month

13 months

13 months

Epoxy

1.6

2.4

2.0

Control

0.6

0.6

0.8

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 45

Finally, Table 2.8 summarises the results of tensile strength (ITST) tests performed on laboratory and in situ cured cores. A direct comparison of the results of laboratory cured cores shows that the strength of the epoxy-asphalt mixture increased by 50% between one and 13 months, while the strength of the control mixture did not experience any change. The in situ results also indicate that the strength of the epoxy modified mixture is substantially higher than that of the control. In comparison with earlier results generated during Phase 2 of this project (ITF, 2008; Elliot et al., 2008), it is noteworthy that the laboratory cured stiffness at 20ºC of the field cores is substantially lower (8 240 MPa after 13 months compared with 13 400 MPa after 60 days). The reason for this difference is unexplained, but the slightly reduced stiffness of the field cores may be beneficial in the long term, as discussed further below.

Assessment of performance During this project, field trials of epoxy-asphalt surfacing have been successfully carried out in two locations: Christchurch, New Zealand, and Truro, United Kingdom. Although different generic surfacing materials were trialled on each site (open graded porous asphalt and stone mastic asphalt, respectively), the epoxy binder (sourced from ChemCo and Colas, respectively) was understood to be very similar, essentially comprising specially formulated epoxy resin and a fatty acid curing agent in a penetration grade binder. Thus, even though the two sites are on opposite sides of the world and separated by some 19 000 kilometres, an assessment of the performance of the two field trials reveals some common themes. On both sites, it was noted that paving proceeded smoothly using conventional plant to produce a surface with comparable surface characteristics to those of standard materials, and difficulties were only encountered when there was an interruption in the paving. Thus it was reported from New Zealand that discoloration and ravelling occurred in small patches at the end of each EMOGPA section, corresponding to locations where the paver stood waiting for new material. Similarly, in the United Kingdom, the paver stood for a period between the third and fourth, and the resulting compacted surface now shows some surface irregularity in this location (detectable when travelling over by car, but nonetheless compliant with the UK surface regularity criteria). The importance of a good understanding of mixing temperature and time, and their relationship to curing and “strength” development, was recorded from both sites. In New Zealand, the period for which the material was held at high temperature was kept to a minimum, because of concerns that the material might cure before compaction had been completed. However, as a result, the mixture remained relatively mobile during laydown and there was some initial “pick-up” on the roller. Conversely, in the United Kingdom, following their experience laying epoxy-modified asphalt on bridge decks, Colas provided the laying gang with a timed schedule indicating when the material must be laid and substantially compacted by, related to the mixing temperature. Colas also monitored the quality control and strength development of the manufactured product, so that they could advise when it was appropriate to open to traffic. In the event, the UK trial site was closed to traffic for around 55 hours. Conversely, in New Zealand, the road surface was cooled with water to allow it to be trafficked on the same day that the material was laid. Suitability of newly laid epoxy-asphalt surfacing for trafficking is a subject where a better understanding would be beneficial, of which more below. The primary objective of the field trials was to demonstrate that the performance envisaged from the laboratory and accelerated testing would hold true within the period of the trial under real traffic and environmental conditions. To date, this objective has been comfortably met on both sites, although, as noted in Chapter 2, the stiffness of the epoxy modified SMA mixtures in the United Kingdom,

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

46 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

determined on cores removed from the laid material, are generally lower than previously determined in the laboratory. A similar reduction in stiffness compared with that of laboratory prepared material was observed from the accelerated pavement testing work carried out in Phase 2. However, the stiffness is still comfortably in excess of that of standard surfacing mixtures, with consequent benefits for predicted life, and a slightly lower stiffness may not be a bad thing to help mitigate any potential for cracking. Finally, both mixtures were manufactured, laid and compacted at lower temperatures than conventional materials incorporating paving grade bitumen, with obvious benefits for energy usage and reduction of carbon footprint. Thus the 20% and 30% EMOGPA mixture in New Zealand were manufactured at 122°C and 117°C, respectively, whereas the epoxy modified SMA in the United Kingdom was manufactured between 110°C and 135°C. This temperature range is similar to that which applies for so-called “warm” mixtures (i.e. those manufactured using additives such as wax or zeolite or foamed bitumen with a foaming bar), and is substantially lower than that for typical “hot” mixtures (140-190°C, depending on binder grade). Similarly, the temperature of the epoxy modified SMA leaving the paver must not be less than 90°C, and not less than 70°C at the commencement of rolling, compared with typical temperatures of 130°C and 100°C, respectively, for conventional SMA manufactured with 40/60 pen paving grade binder (BS 594987 ).

Recommendations Two types of epoxy-asphalt binder were investigated in Phase 3 of this project, specifically an amine curing and a fatty acid curing type. Field trials with the amine curing epoxy were not possible, due to difficulties encountered with the type of mixing plant (controlling the precise quantities required proved impossible with a continuous type of plant), with cleaning of the epoxy binder distribution system and, most significantly, with the temperature sensitivity of the “workability window”. Clearly, more work would be required with this material if it was to be more widely used. Conversely, mixtures manufactured with the fatty acid curing type epoxy have been successfully applied and tested in the field using both continuous and discontinuous mixing plant. They appear ready for more widespread use, although they would benefit from further development in some areas, as detailed below. It is clear that the epoxy-modified asphalt mixtures can be laid at significantly lower temperatures (30-40°C lower) than their conventional counterparts manufactured with paving grade bitumen. This was not foreseen at the outset of the project but is nonetheless a substantial benefit in today’s carbon conscious environment, and one which has become much more important since the inception of this study in 2001. More work is warranted in this area, to optimise the benefit. Linked to this is the need for a better understanding of how temperature and time interact in relation to curing, strength and/or stiffness development and suitability for trafficking in the field. Thus, even though ChemCo Systems and Colas have developed specifications for the installation of epoxy-asphalt concrete surfacing (Gaul, 2011) and epoxy-modified SMA (assessment, testing and monitoring above), respectively, it is clear that the development of mechanical properties in the field occurs at different rates to that predicted from laboratory work. Thus, while it might be expected that the greater mass of material in the field result in higher overall average temperatures and consequent lead to faster strength development than in the laboratory, the reverse was noted in accelerated pavement and field testing in the United Kingdom. It is likely that factors such as substrate temperature and ambient conditions are significant here, but more work is required to quantify their respective influence. One issue which caused some concern when discussions were held with prospective contractors (for example, in England and in France) was procedures for dealing with plant breakdown and cleaning. For the UK trial, a location close to the site was identified in case there was a need to tip material because of a plant breakdown (this did not prove necessary here). The epoxy binder distribution system for both the

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 47

New Zealand and UK trials were temporary and bespoke set-ups, and cleaning the system after production was similarly bespoke. Thus, in New Zealand, the tank, pump and lines carrying Part B (the major component of the epoxy binder) into the plant drum were cleaned by pumping through conventional bitumen and making several tonnes of mixture which were discarded. A separate pump was used for Part A. The line carrying Part A joined the line carrying Part B a few metres before the latter entered the plant drum (giving some time for mixing). Afterwards the pump and line carrying Part A were disconnected from the system, and when it had cooled, this section of line and the (explosion proof) pump only were cleaned with kerosene. Clearly, for larger scale production of epoxy asphalt, an automated mixing process would be essential, and this has been solved by ChemCo Systems in the United States with their automatic machine for blending of the epoxy binder components, which is cleaned after each batch by blowing compressed air and cleaning agent (section 2.5 above). However, it remains the case that contractors will be reluctant to take on board this new technology unless there are full proof procedures in place to ensure any plant used can be cleaned and put back into use for conventional materials with a minimum of effort, and there is work to be done to reassure them that this is the case. One final point relates to the surface blemishes that were observed in both New Zealand and the United Kingdom when the paving process was interrupted. It is in all situations preferable for paving of asphalt to be one continuous operation but it is clear that this is more significant for successful laying of epoxy modified mixtures. The use of asphalt feeders or shuttle buggies can help to solve this problem; the advantages of the latter, for example, are that they can store a delivered load of asphalt to maintain a uniform temperature of the material, a mixer auger reduces segregation, the system helps throughput by minimising the need for the paver to stop and the paver hopper is continuously fed, reducing cold spots. The high initial cost is counteracted by increased productivity and reduced compaction problems with improved durability. However, because of the cost implications, it is likely that the use of such devices would need to be written into relevant contract documents.

Summary and conclusions In earlier phases of this project, an economic appraisal had demonstrated the likely benefits from development of road surfacing materials with a service life in excess of 30 years, which were subsequently characterised in laboratory and accelerated load testing of two candidate materials (epoxy asphalt and HPCM). This chapter summarises the work done in Phase 3 of the project; specifically, to plan, execute and monitor full scale trials of the optimum epoxy-asphalt mixture design formulation that had been developed in the earlier stages of this project. Although interest in participating was initially expressed by national institutions from a number of countries, it was only possible to carry out successful field trials in two countries, New Zealand and the United Kingdom. The sites in question were located on relatively heavily trafficked sections in Christchurch (New Zealand) and near Truro (United Kingdom). In each case, the surfacing required replacement due to surface deterioration but the substrate was considered to be sound. The epoxy-asphalt mixture designs were both 10 mm nominal size, comprising EMOGPA in New Zealand and SMA in the UK, together with appropriate controls. Both the epoxy asphalt and control materials were manufactured, transported and laid using conventional plant, generally without incident. Full details of the field trial installations have been published elsewhere (Herrington, 2010; Elliot et al., 2013). Idealised requirements for assessment, testing and monitoring of the epoxy-asphalt field trials were established by the OECD/ITF steering group in July 2009, involving monitoring before, during and after the trial, the latter including monitoring at early age and after one year in service. However, it was left up to the individual countries to decide upon the most appropriate parameters to measure. Data collected for

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

48 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

the EMOGPA in New Zealand over a period of three years include visual condition, rut depth, skid resistance, permeability, noise and stiffness (indirect tensile modulus at 25°C): data collected for the SMA in the UK over a period of one year include visual condition, rut depth, texture, skid resistance, profile, cracking, FWD, coring, indirect tensile strength and stiffness (indirect tensile stiffness modulus at 20°C). Feedback from New Zealand indicated broadly similar rut depth, skid resistance and noise characteristics for the epoxy modified and control materials, with minimal absolute levels of rut depth. The permeability data also indicated similar drainage times for the control and 20% EMOGPA mixtures, with permeability reducing over time; conversely, the 30% EMOGPA showed relatively little change over time and was the most free draining of the three materials (albeit this was not reflected in beneficial noise characteristics). The indirect tensile modulus data showed that the 20% EMOGPA mixture cured rapidly over the first 30 days and slowly thereafter, reaching an ultimate value of 3 300 MPa at around 800 days, whereas the control OGPA modulus increased more slowly to an ultimate value of 2 300 MPa and the 30% EMOGPA modulus has remained relatively unchanged throughout, at around 650 MPa. Data from two monitoring visits in the UK over the first year of service, and feedback from the local authority, indicate good performance to date, with comparable surface characteristics (particularly regularity, skid resistance and texture, with negligible cracking and rutting) on the trial and control sections and no indication of change since the trial site was opened to traffic. A structural assessment of the site carried out using the falling weight deflectometer (FWD) has confirmed broadly similar performance in the trial and control areas. Testing of cores extracted from the site and subsequently stored in the laboratory, indicate that the epoxy-asphalt material has substantially greater stiffness and tensile strength than the control material. The mean stiffness at 20°C of the epoxy asphalt has increased from 2 850 MPa at 1 month to 8 240 MPa at 13 months, compared with a comparable change for the control material from 1 050 MPa to 1 790 MPa. In addition, the mean tensile strength of epoxy asphalt has increased from 1.6 MPa at 1 month to 2.4 MPa at 13 months, whereas the control mixture has shown no change in strength over time and a much lower absolute value (0.6 MPa). Surface blemishes were observed in both New Zealand and the UK when the paving process was interrupted, at the end of each of the EMOGPA sections (discoloration and some ravelling) and in the final quarter of the SMA section (visual indication of irregularity). In order to overcome this problem, the use of asphalt feeders or shuttle buggies is recommended to ensure the asphalt paving is completed as a continuous process. Notwithstanding, although the surface blemishes in the SMA are visible to the naked eye and detectable when driving over the site in a vehicle, the surface regularity complies with the current UK requirements. The primary objective of the field trials was to demonstrate that the performance envisaged from the laboratory and accelerated testing would hold true within the period of the trial under real traffic and environmental conditions. To date, this objective has been comfortably met from the work carried out in New Zealand and the UK. Secondary objectives were to develop construction methods, optimise mixtures and increase contractor experience levels. It is certainly the case that the field trials have successfully demonstrated that the full-scale manufacture and construction of epoxy modified OGPA and SMA surfacing can be accomplished with standard plant and equipment, and with only very minor changes to practice. In terms of increasing contractor experience levels and optimising mixtures, it is likely that this will only come with increased uptake of the concept. From the present research, obvious targets for mixture optimisation would include extending the workability window, improved knowledge of how curing temperature influences early traffic ability and ultimate mechanical properties, and making best use of the potential to manufacture and lay epoxy modified mixtures at lower temperature than conventional material.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 49

The economic case for long-life surfacing of the type described in this chapter depends on achieving a service life at least twice as long as that typically expected from asphalt surfacing currently used in road pavements. As an example, the expected life of “conventional” SMA surface courses popularly used in Europe and North America were reported to range between 5 and 15 years (EAPA, 2007). Clearly, the current trials have a long way to go before a life of 30 years can be demonstrated, but early signs are encouraging. Experience in New Zealand is that in high demand urban areas and motorways, OGPA lives in the order of eight years are common. Evidence from laboratory studies (Herrington, 2010) and practical application of dense epoxy-asphalt mixtures overseas (Lu et al., 2012) would suggest that lives substantially in excess 16 years should be obtainable for epoxy-modified OGPA. Against this background, additional EMOGPA trials were constructed on the Christchurch Southern Motorway in November 2012 involving concentrations of EMOGPA ranging from 25% to 100% (Herrington, 2013). These trials suggest that the life of 25% EMOGPA (with superior performance to that of unmodified OGPA but inferior to that of 100% EMOGPA, but at a reduced cost) would only need to achieve a life of 11 years before whole life cost benefits begin to accrue.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

50 – 2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS

References ASTM (American Society for Testing and Materials) (2004), Standard practice for preparation of bituminous specimens using Marshall apparatus. ASTM D6926–04. BS EN 12697-5, Bituminous mixtures. Test methods for hot mix asphalt. Determination of the maximum density. BS EN 12697 -6, Bituminous mixtures. Test methods for hot mix asphalt. Determination of bulk density of bituminous specimens. BS EN 12697-8, Bituminous mixtures. Test methods for hot mix asphalt. Determination of void characteristics of bituminous specimens. BS EN 12697-23, Bituminous mixtures. Test methods for hot mix asphalt. Determination of the indirect tensile strength of bituminous specimens. BS EN 12697-26, Bituminous mixtures. Test methods for hot mix asphalt. Stiffness. BS EN 12697-34, Bituminous mixtures — Test methods for hot mix asphalt — Part 34: Marshall test. BS EN 13036-1, Road and airfield surface characteristics. Test methods. Measurement of pavement surface macrotexture depth using a volumetric patch technique. BS 594987, Asphalt for roads and other paved areas. Specification for transport, laying, compaction and type testing protocols. Department of Transport (n.d.), Specification for Highway Works, Manual of Contract Documents for Highway Works, The Stationery Office, London, Vol. 1. Department for Transport (2004), Skid resistance, HD28/04, Design Manual for Roads and Bridges Volume 7, Section 3. Department for Transport (2008), Pavement design, HD29/08, Design Manual for Roads and Bridges Volume 7, Section 3. Department of Transport (2011), SCANNER Surveys for Local Roads, User Guide and Specification, Volume 3, Advice to Local Authorities: Using SCANNER survey results, Version 1.0Edition: www.pcis.org.uk/iimni/UserFiles/Applications/Documents/Downloads/SCANNER%20and%20T TS/SCANNER%20Specification/SCANNER_Spec_2011_Volume_3.pdf (accessed 28 August 2013). EAPA (2007), Long-Life Asphalt Pavements, June. Elliott, R.C., C. Fergusson, J. Richardson, A. Stevenson and D. James (2013), “Field Trials of a Long Life Surfacing Material”, Asphalt Professional, Issue 57, in preparation, September. Elliott, R.C., I. Widyatmoko, J. Chandler, A. Badr and W.G. Lloyd (2008), Laboratory and pilot scale assessment of long life surfacing for high-traffic roads. Paper 300-005 in Proceedings of the 4th Eurasphalt & Eurobitume Congress, 21–23 May 2008, Copenhagen. LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

2. EPOXY-ASPHALT ROAD SURFACING FIELD TRIALS – 51

Gaul, R. (ChemCo) (2011), personal communication. Herrington, P. (2010), Epoxy-modified porous asphalt. NZ Transport Agency research report 410. Herrington, P. (2013), Epoxy Modified OGPA – Christchurch Southern Motorway 2012, Contract No. 12-838, Opus International Consultants Ltd. Herrington, P., D. Alabaster, G. Arnold, S. Cook, A. Fussell and S. Reilly (2007), Epoxy modified opengraded porous asphalt. Land Transport New Zealand research report 321. pp. 27. ITF (2008), Long-Life Surfaces for Busy Roads, OECD Publishing, Paris. DOI: http://dx.doi.org/10.1787/9789282101209-en Lu, Q., R.W. Gaul and J. Bors (2012), Alternate Uses of Epoxy Asphalt on Bridge Decks and Roadways, Proceedings of the 5th Eurasphalt & Eurobitume Congress, Istanbul. NZTA (2000), Standard test procedure for measurement of skid resistance using the British PendulumTester. Draft NZTA T/2:2000. NZTA (2005), Notes to the performance based specification for hot mix asphalt wearing course surfacing, Appendix A, NZTA P23:2005. NZTA (2007), Specification for open-graded porous asphalt (OGPA). NZTA P/11:2007. NZTA (2011), Specification for roading bitumen. NZTA M/1:2011. ECMT (2005), Economic Evaluation of Long-Life Pavements: Phase 1, OECD Publishing, Paris. DOI: http://dx.doi.org/10.1787/9789264008588-en Simpson, W.C., H.J. Sommer, R.L. Griffin and T. K. Miles (1960), Epoxy asphalt concrete for airfield pavements, ASCE Journal of the Airport Division, Vol 86/ 1, pp. 55–71. Standards Australia (1995) Methods of sampling and testing asphalt – determination of the resilient modulus of asphalt – indirect tensile method. AS 2891.13.1:1995.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 53

Chapter 3. Field trials with high-performance cementitious materials

High-performance cementitious materials (HPCM) is an innovative technique that was elaborated during the previous phase of the project. This chapter summarises the results of the field tests conducted in France and further testing conducted in the United Kingdom regarding large-scale mixing and pumping equipment and laying. The material seems promising, but requires further refinement to reach the project goal in terms of durability in real conditions.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

54 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

This chapter is concerned with the high-performance cementitious materials (HPCM) and describes the field trials conducted during Phase 3 of this project. The overall objective of Phase 3 was to demonstrate that the performance envisaged from the laboratory and accelerated testing would hold true within the period of the trial under real traffic and environmental conditions. Additional aims were to develop construction methods, optimise mixtures, and increase contractor experience levels. Initially the following countries expressed interest in participating in the field trials: Belgium, Denmark, France, Hungary, Spain and the United Kingdom. Unfortunately, due to lack of national funding and a general reluctance to test high risk solutions, together with the impact of 2008 economic crisis, most of the countries did not actually succeed in carrying out the full scale trials. The UK did some preparatory testing of mixing equipment and small scale paving equipment, but a full scale paving trial was eventually not realised. In France a national collaborative project was set up to carry out two full-scale trials along with further development of the HPCM.

Trials of HPCM in the United Kingdom In the UK the project was funded by the Highways Agency and the work was carried out with the Transport Research Laboratory (TRL) as the main contractor and other involved parties were Tecroc Products and Balfour Beatty. A number of laboratory mix studies and small scale paving applications took place first. Initial testing in 2008 focused on the mixing equipment and developing procedures for manual paving of small areas. For the first trials the mix proportions suggested from Phase 2 were used, with some variations in the water content with the purpose of trying to reduce the required mixing time. In the first trials in 2008 it was realised that the mortar had a tendency to develop a surface skin which made it very difficult to achieve satisfactory embedment of the chippings. It also proved very difficult to place an 8 mm thick layer on a surface with some irregularities. Increasing the mortar thickness to 20 mm made placing much easier. From these initial trials the following conclusions were made: •

Mortar thickness should be increased to 15-20 mm.

•

A high shear mixer should be used. The mix should be able to be pumped.

•

The skinning problem should be addressed, possibly by changing the mix.

•

Chippings should be vibrated into the surface using a wide, flat screed. With the increased mortar thickness it was also decided that a larger 6-10 mm size chipping should be used.

•

The spreading rate should be increased from 4 kg/m2 to 6 kg/m2.

In 2009, the first trial focused on the “pumpability” of the fibre-reinforced mortar, and the results were encouraging. Along with this, laboratory studies were performed in order to try to solve the skinning problem, and it was found that by reducing the microsilica content considerably, the time before skinning could be increased from 10 minutes to 40 minutes. The new mix was tested at a trial on 1 May 2009 at Tecroc’s depot. The mix was successful and good embedment of the chippings was achieved. However, as a result of the mixing equipment being of an inferior quality to that previously used, the second part of the trial was not successful. The pump was blocked by the mix, and this finding led to the conclusion that a new rotor-stator design should be developed. LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 55

A new rotor-stator with larger amplitude and smaller diameter was tested on 9 September 2009 and was found to perform well. This trial also led to the conclusion that a Baron pan mixer would be the best choice for the final trial in 2009, which took place at Tecroc’s depot. For this trial two Baron pan mixers and a separate pump were used. It was found that to get good mixing results, the amount of mixture in the mixer should not be too small. Pumping and placing worked well and good chip embedment was achieved. Figure 3.1. Equipment used for the HPCM laying

The Baron pan mixer

Laying the HPCM with the pump

The levelling screed

Spreading of chippings

After these initial trials, most problems had been solved satisfactorily: •

The HPCM did not start skinning for 30 – 40 minutes.

•

The HPCM could be pumped to speed up the laying.

•

There was good embedment of the chippings.

In 2010, the UK focused on further development of mixing and pumping equipment. A new standard size rotor-stator design gave significantly better results than the previous purpose-made version. A larger size mixer was tested, as a result of previous problems with the mixer electronic tripping out when large amounts of material were put in the mixer. A 300-litre Baron pan mixer had sufficient capacity and performed well.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

56 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

A full scale trial was planned for the spring of 2011, but unfortunately this trial was for various reasons not realised, and no further activities have taken place in the United Kingdom.

Trials of HPCM in France France launched a national project called CLEAN (Chaussée à Longévité Environnementale Adhérente et Nettoyante, or Long-Life, Environmentally Friendly, Skid Resistant and Air-Cleaning Pavement) with the aim of developing a semi-industrialised technology for HPCM application and validating the concept through experimental construction sites. The eight project partners included research laboratories, road owners, material providers, a construction facility maker and a contractor. Full documentation of the project can be found on the project web-site (CLEAN, n.d.). The first activities in 2009 focused on preparations for two test sites. Initially, the mix design from Phase 2 was adapted to fit with the planned laying specifications. The constituents were re-considered based on further laboratory testing. Mixing was planned to take place in an ordinary annular pan mixer. The dry materials were to be delivered as premix, and the mixing time was planned for eight minutes, with fibres added during the last two minutes. Pumping tests indicated that the material should be pumpable, however it was noted that care was needed to avoid fibre plug formation during cleaning. The equipment producer SAE developed three separate machines for HPCM laying. It was the intention that they should eventually be assembled into one laying unit. Figure 3.2. Machines developed for laying HPCM

Mortar spreader

Chippings spreader

Roller

During the autumn of 2009, three testing campaigns were carried out at IFSTTAR (formerly LCPC) in Nantes. Some of the conclusions from these tests were: •

Mortar slump value should be between 130 mm and 170 mm.

•

Laying speed should be around 2 m/min and kept at a constant rate to achieve good evenness.

•

Base course should be moistened before applying the mortar.

•

Chippings should be spread quickly after mortar laying.

•

Roller should be driven by a motor on the axle (not pulled).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 57

After the trials in Nantes, it was observed that some steel fibres were visible on the surface. After passes with a bicycle it was realised that these fibres could puncture a bicycle tyre, but not motorised vehicle tires. Other types of fibres (PVA, aramid and glass) were then being considered for the field trials. Most of the work performed within the CLEAN project in 2010 dealt with the preparation, execution and monitoring of the first experimental site, in Brette-les-Pins (10 km south of Le Mans, Sarthe). The site preparation began in August 2009, where a high-modulus asphalt layer was placed on the roundabout. The HPCM was laid between 26-28 April 2010. Figure 3.3. Construction of the HPCM in Brette-les-Pins

Mixer

Paving train

Chipping spreading

Manual compaction

With the background of the potential bicycle tyre puncture problem when steel fibre reinforcement is applied, a mix with 1% PVA fibres was chosen for this first trial. The mixing equipment was placed 30 m from the roundabout. The mortar mix was delivered in premix big-bags, leaving only fibres, retarder and water to be added. The asphalt substrate in the roundabout was shot blasted prior to paving in order to clean the bitumen surface layer from the aggregate and thus improve the bond. The mix quality was monitored with various tests, including compressive strength, which ranged from 117 to 134 MPa (28-day values), as measured on 4x4x16 cm prisms. LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

58 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

The HPCM layer was placed with specially developed machinery including a mortar spreader with a vibrating screed followed by a chipping spreader. The chipping compaction was planned to take place using a vibrating roller. However, this was not successful and manual compaction by workers walking on wooden boards was chosen as the solution for this procedure. After completion of the paving, various tests took place including skid resistance and noise measurement. Skid resistance values were satisfactory whereas the noise generated was slightly higher than values found on a surface dressing and significantly higher than values for a reference asphalt pavement. Visual cracking surveys conducted some months after paving showed both transverse cracking with mean distances between the cracks of 1-2 m, and also a few longitudinal cracks and examples of crazing, probably caused by improper application of curing compound or too early trafficking. The transverse cracking was as expected given the low volume and the soft nature of the PVA fibres, and is well in line with the results from Phase 2 of the project. Generally, the roundabout looked rather heterogeneous in terms of colour, amount of aggregate particles and surface height. Figure 3.4. Aspects of the surfacing of the HPCM section in Brette-les-Pins

Overview

Transverse and longitudinal cracking

Transverse cracking

Typical transverse crack (opening > 1 mm)

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 59

The Brette-les-Pins roundabout trial led to the following conclusions: •

Laying of the three rings, a total of about 750 m2, was carried out within a three-day period. As a result of the rapid hardening of the mortar, it was possible to load the material with site machines after about 24 hours.

•

The fine tuning on the mortar and chipping spreaders led to globally satisfactory operation (when the speed could be kept constant).

•

Regarding mortar production, the obtained results were both satisfactory and consistent.

•

Interruptions in the mortar delivery to the paver led to difficulties in terms of longitudinal evenness and compaction quality (owing to the high thixotropy of the fresh mortar).

•

Poor chipping compaction was the main problem: given the high mortar thixotropy, the penetration of chippings through the surface became difficult a short while after laying the mortar. Both mortar rheology and compaction technique need to be revised.

•

Finally, the site must be closed in order to avoid any untimely circulation on the fresh material (pedestrians, two-wheel vehicles, etc.).

Based on these experiences, the following improvements were envisaged: •

To modify the mortar mixture design with the aims of decreasing the proneness to cracking and the thixotropy.

•

To change the compaction process (by developing a new machine or another more efficient manual technique).

•

To perform automatic curing with a sprayer mounted on a trailer.

•

To connect the two spreaders in a single machine.

•

To further investigate the method of joint construction.

Other activities in the French CLEAN project included numerous tests performed to optimise the mortar mixture-design, according to the experience gained at Brette-les-Pins. To reduce the thixotropic behaviour of the fresh mix, and given the fact that the mix appeared to be over-designed regarding the final strength, it was decided not to use silica fume in the mixture, but only Portland cement as a binder. After readjustment, the early stiffening, observed on the material while at rest, is much slower than in the original mixture. The final strength is also lower, but the level is probably still high enough to provide sufficient durability for the HPCM under traffic. Regarding the fibres, a new synthetic fibre (of the aramid type) was examined. It appeared that an amount of 2% was sufficient to prevent macro-cracking. However, these fibres were damaged during pumping, resulting in very poor workability. Additional tests were carried out with a second type of aramid fibre, where the coating of individual fibres was strengthened. Here the stability during prolonged mixing was satisfactory, but the bond with the cementitious matrix in the hardened material was too low, impairing the crack-control ability of the fibres. Finally the option of synthetic fibres was abandoned. For the second construction site, a new machine was designed and built to perform the compaction of chippings. Light aluminium panels are trailed over the freshly placed HPCM, with a back-and-forth longitudinal motion. Full-scale tests performed in September 2010 gave encouraging results. A second trial was carried out in May/June 2011, on a 150 m section of the road RD117, near Saint-Philbertde-Granlieu, in the Loire-Atlantique department. This road carries around 16 000 vehicles per day.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

60 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

Since this trial section was to be carried out on a highway with no bicycle traffic, it was decided to go back to using steel fibre reinforcement. Another change from the Brette-les-Pins trial was that, as mentioned previously, the use of silica fume in the mix was omitted. The mix designs for the two French trials are summarised in the following table, together with the original mix used in Phase 2 of the long-life pavement (LLP) project. Table 3.1. HPCM mix-composition Constituents in kg/m3

Mix A

Mix B

Mix C

0.2/1 mm rounded siliceous sand

429

533

-

0.08-0.315 mm rounded siliceous sand

429

533

-

0/1 crushed quartz sand

-

-

883

CEM I Portland cement

985

814

982

Silica fume

197

163

-

Superplasticizer (dry powder)

4.4

2.26

2.21

Retarder (in liquid form)

4.95

2.44

2.65

13x0.2 mm steel fibres (3%-vol.)

235

-

226

-

13

-

Water

207

183

294

w/c

0.21

0.22

0.30

Slump (cm)

21

20-24

20-26

Chippings

4.5 kg/m2 (4/6 calcined bauxite)

4.5 kg/m2 (3/6 calcined bauxite)

3 kg/m2 (4/6 porphyric)

PVA fibres (1%-vol.)

A: Mix used by the partners of the LLP Phase 2 working group. B: Mix used for the Brette-les-Pins site. C: Mix used on the St. Philbert site.

From the first trial, two major issues were identified regarding the HPCM technology: a difficulty to control the embedment of chippings in the fresh mortar and a rough texture, generating a high level of rolling noise under traffic. These findings led to the re-design of this hydraulic solution, to avoid the incorporation of chippings and rather create a mortar macro texture. The first idea was to try to reproduce the texture of the best current asphalt-wearing courses, e.g. the BBTM 0/6 (very thin asphalt concrete with 6 mm maximum size of aggregate). Tests were carried out with polymeric matrix, but this process was found difficult to mechanise on site. It was finally preferred to use a known technique, which is well mastered in the United States (Wiegand, 2006), consisting of sawing narrow longitudinal groves in the hardened mortar.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 61

The test section at St. Philbert was therefore constructed with this solution, called grooved fibre-reinforced ultra-high performance concrete (GFRUHPC), over half of the length, and the initially developed HPCM surface with embedded bauxite chippings on the other half. With the GFRUHPC technique, three types of risks had to be assessed prior to proceeding with a full-scale application: •

The “shimmying” phenomenon for two-wheeled vehicles. This problem is well-known in the motorcyclist community and, in the past, made some concrete pavements very unpopular. However, according to recent American experience, it seems that shimmying only appears above a certain critical dimension of the grooves. With the ones specified for GFRUHPC, this risk seems to be minimised.

•

Tyre puncture. Since steel fibres are used in the mortar, one could expect that some fibres would come out of the top surface and could damage the vehicle tyres. Tests were carried out on a 25 metre long test section at IFSTTAR, Nantes (Nguyen et al., 2011). The risk was confirmed for bicycles, but not for motorcycles, cars or trucks. Since the GFRUHPC technology is dedicated to highways, the choice of steel fibre was retained.

•

Hydroplaning. Any wearing course must pass requirements dealing with water drainage under tyres. Based on a study published at TRB (Ong et al., 2006) and confirmed by theoretical calculations (Nguyen, 2011), this risk could be eliminated, provided that the grooving specifications are met.

Photos from the construction of the St. Philbert site are shown in the following Figure 3.5. As for the HPCM, the Wehner & Schulze machine was used to assess the potential skid resistance of the GFRUHPC layer (Figure 3.6). The smooth material, in the absence of surface texturing, showed a rapid decrease in its friction coefficient. However, the use of crushed, angular quartz sand, together with the application of grooves in the fresh mortar, led to a large improvement of the skid resistance. After the rapid wear of the superficial cement paste, there seems to be a surface regeneration which creates an increase of friction coefficient. The remaining level of friction after 500 000 cycles is even higher than the best level achieved by the control asphalt material after only some thousands of cycles.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

62 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

Figure 3.5. Construction of the HPCM section in St. Philbert (France)

Fibre addition in the mixer

Paving train

HPCM paving

Manual correction of the HPCM surface

Concrete surface before grooving

Grooving

Grooving with a diamond saw

Final appearance, grooving distance 10 mm, width 4.5 mm, depth 3 mm

Source: Clean Project.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 63

Figure 3.6. Skid-resistance of GFRUHPC, as measured with the Wehner & Schulze machine

Source: CLEAN project.

The potential noise generation is a parameter that cannot be readily measured in the laboratory within the current pavement material technology. The noise generation was assessed through the CPX technique (Weigand, 2006), which consists of measuring the noise level with microphones fixed near the tyres of a reference car. The results are displayed in Figure 3.7. HPCM confirms its noisy character, with an increase of 3-4 dBA as compared to the BBTM 0/10 control asphalt material. However, the GFRUHPC displays the same type of noise generation as that of asphalt, both in terms of total energy and frequency range. Figure 3.7. Noise generation as measured by the CPX method on the St. Philbert test section after two weeks of traffic

Source: CLEAN project.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

64 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

Apart from the testing of the two surface structures, a test involving the photo catalytic air purifying effect of cement with Ti02 was also carried out in connection with the St. Philbert site. Laboratory testing showed that the photo catalytic air purifying effect was better for the GFRUHPC than for a conventional concrete pavement with Ti02 cement, whereas the HPCM surface exhibited lower performance due to the screening effect of the incorporated chippings. After the construction site completion, cores were periodically taken, in order to monitor the evolution of the depolluting capability of the material. As for conventional concrete, it appears that a periodical cleaning process is necessary to maintain a certain level of this function.

Assessment of performance With the original HPCM solution as developed in Phase 2, the main problem for both of the trial sites was the difficulty in achieving a good and uniform embedment of the chippings. This has led to high noise levels and a relatively poor aesthetic appearance. Some cracking has appeared at the first site at Brette-les-Pins, partly as a result of the special geometry of the roundabout but mostly as a result of the use of a low amount of PVA fibres, which was not sufficient to control the cracking. Between the constructions of the two sites, a 25-metre long pad was built at IFSTTAR and was found to be nearly free of any cracking, which shows that with a proper fabrication and laying, the goal of avoiding shrinkage-induced cracking is attainable. At the second site at St. Philbert (with 3% steel fibres) some cracking developed. The main cause probably lies in the difficulty in obtaining a homogenous dispersion of the steel fibres throughout the mixture (fibre balls were visible from place to place at the pavement surface). Also, the evenness of the base course was not satisfactory, leading to poor control of thickness of the wearing course. When compared against the first test section, cracks were less numerous but much more open, leading in some parts to delamination and loss of the material. After some months of service, the new wearing course of St Philbert test section had to be removed, given the risk for the safety of road users– mainly motorcyclists – induced by punch-outs. Skid resistance values are good for both the original HPCM design and the GFRUHPC. The noise generation is still not satisfactory for the HPCM owing to the difficulty in achieving a good chipping embedment, whereas the GFRUHPC shows a good potential to be comparable to an asphalt pavement regarding noise properties.

Recommendations A comprehensive review of the HPCM and GFRUHPC solutions, encompassing technical, economic and environmental aspects, can be found on the CLEAN internet site: http://clean.ifsttar.fr/. Based on the two French trials, the following main points can be summarised: •

The material requires a long mixing time, unless special mixers, similar to the ones used for more conventional ultra-high performance concrete, are used.

•

Attention must be paid to the risk of fibre cluster formation.

•

When conveying the fresh material with a pump, piston pumps are preferred.

•

A consistent application of the material cannot be carried out to a thickness less than 19 mm, given the precision of application techniques and the evenness of the base courses.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS – 65

•

A thorough curing process should be applied just after casting in order to avoid the appearance of cracks.

•

Placement of GFRUHPC sections should commence and terminate with slightly greater thicknesses, and measures must be taken to ensure good bond with the base course.

•

The material should be applied in full-lane width (3.50 m according to the French standard), avoiding inlay application.

•

The typical material specifications could be as follows: use of angular, hard sand with a maximum size of 1 mm, amount of steel fibre 2.5-3% by volume, slump of (20-24 cm), compressive strength of 40 MPa at traffic opening, to reach 120 MPa at 28 days, total shrinkage less than 0.5%.

A further research and development effort is needed to allow the application of this technique with a sufficient level of safety. However, most difficulties were identified during the CLEAN project and solutions are visible to reach a full degree of maturity.

Summary and conclusions Several countries planned to participate in the field testing of the high-performance cementitious material pavement that was developed through laboratory testing in Phase 2 of the project. However, for various reasons, field tests were eventually only carried out in France, and the French CLEAN project therefore constitutes the main input to Phase 3 of the current project. The UK did some testing of large scale mixing and pumping equipment, performed manual laying trials, and reached promising results in this respect, but no test section was realised. France used the HPCM solution developed during Phase 2 as the starting point for further small-scale testing and two trial sections. In order to decrease the level of thixotropy of the fresh mortar, and given a certain overdesign of the original recipe, it was decided to use Portland cement as the only binder. Various alternative fibres were rigorously tested, namely PVA, glass or aramid fibres, but steel fibres appeared to be the only ones allowing, at the same time, a safe pumping placement and efficient crack prevention in the hardened state. It turned out that under site conditions the original HPCM was delicate to apply and the noise generation under traffic was quite high (as with any surface-dressing). Therefore, an alternative to this option was studied, where no coarse aggregates were embedded in the fresh mortar. Instead a texture was provided to this material, with the idea of controlling both the skid resistance and the noise generation. This GFRUHPC appeared to be the most promising way of producing a cement-based long-life surfacing with good surface properties and low noise generation. The laying technique was validated but fine tuning remains to be done for a better thickness and evenness control and to avoid cracking, which is a mandatory condition for the objective of durability. Once those advances have taken place, the technology should be affordable and very sustainable in terms of energy and resource consumption.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

66 – 3. FIELD TRIALS WITH HIGH-PERFORMANCE CEMENTITIOUS MATERIALS

References CLEAN (n.d.), http://clean.ifsttar.fr (accessed December 2013). Méthode d'essai des lpc n°63, Version 2.0 : "Mesure en continu du bruit de contact pneumatiquechaussée", édition LCPC. Nguyen, T.B. (2011), “Formulation de l’EHFG pour la planche N°2”, Rapport projet ANR CLEAN, Livrable R5, June. Nguyen, T.B., T. Sedran, O. Garcin, P. Maisonneuve, Y. Pichaud, J. Cesbro and F. De Larrard (2011), “Planche d’essai IFSTTAR”, Rapport projet ANR CLEAN, R8bis deliverable, http://clean.ifsttar.fr/ Ong, G.P. and T.F. Fwa (2006), “Analysis of effectiveness of longitudinal grooving against hydroplaning”. TRB Annual Meeting, January. Wiegand, P. (2006), “Concrete Solutions for Quieter Pavements on Existing Roadways”, National Concrete Pavement Technology Center, October.

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

WORKING GROUP MEMBERS – 67

Working Group members Chair: Mr. Francois DE LARRARD (France) Belgium

Ms. Anne BEELDENS, Belgian Road Research Centre

Denmark

Mr. Finn THØGERSEN, Danish Road Directorate

France

Mr. Francois de LARRARD, IFSTTAR and Lafarge Mr. Ferhat HAMMOUM, IFSTTAR Mr. Thanh-Binh NGUYEN, IFSTTAR

Germany

Mr. Oliver RIPKE, BASt Mr. Lars NEUTAG, BASt

Hungary

Ms. Katalin KARSAI, KTI

Israel

Mr. Leonid SUSSKIN, National Roads Company

New Zealand

Mr. Dave ALABASTER, New Zealand Transport Agency Mr. Phil HERRINGTON, Opus International Consultants Ltd

Norway

Mr. Rabbira SABA, Norwegian Public Roads Administration

Spain

Mr. Angel MATEOS, CEDEX

United Kingdom

Mr. John CHANDLER, TRL Mr. Richard ELLIOTT, URS Infrastructure & Environment UK Limited Mr. Wyn LLOYD, UK Highway Agency Mr. Simon RICKETTS, TECROC

United States

Mr. Jack YOUTCHEFF, Federal Highway Administration

ITF/OECD Secretariat

Ms. Véronique FEYPELL

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

68 – WORKING GROUP MEMBERS

Members of the Editorial Committee The report was primarily written by Mr Finn Thøgersen (Danish Road Directorate, Denmark) and Mr Richard Elliott (URS Infrastructure & Environment UK Limited), respectively co-ordinators of the HPCM and epoxy-asphalt trials, and by Mr François de Larrard (IFSTTAR and Lafarge, France), Chairman of the Working Group.

Other contributors Thierry Sedran (IFSTTAR, France).

LONG-LIFE SURFACINGS FOR ROADS: FIELD TEST RESULTS — © OECD/ITF 2017

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT The OECD is a unique forum where governments work together to address the economic, social and environmental challenges of globalisation. The OECD is also at the forefront of efforts to understand and to help governments respond to new developments and concerns, such as corporate governance, the information economy and the challenges of an ageing population. The Organisation provides a setting where governments can compare policy experiences, seek answers to common problems, identify good practice and work to co-ordinate domestic and international policies. The OECD member countries are: Australia, Austria, Belgium, Canada, Chile, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Latvia, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States. The European Union takes part in the work of the OECD. OECD Publishing disseminates widely the results of the Organisation’s statistics gathering and research on economic, social and environmental issues, as well as the conventions, guidelines and standards agreed by its members.

OECD PUBLISHING, 2, rue André-Pascal, 75775 PARIS CEDEX 16 (74 2017 02 1 P) ISBN 978-92-82-10810-9 – 2017