Optimization of Thin Asphalt Layers – State-of-the-Art Review

Optimization of Thin Asphalt Layers – State-of-the-Art Review
Optimization of Thin Asphalt Layers
– State-of-the-Art Review
By Ulf Sandberg* (ed.), Jørgen Kragh**, Luc Goubert***, Hans
Bendtsen**, Anneleen Bergiers***, Krishna P. Biligiri*, Robert Karlsson*,
Erik Nielsen**, Erik Olesen**, Stefan Vansteenkiste***
*Swedish National Road and Transport Research Institute (VTI)
** Danish Road Institute (DRI)
*** Belgian Road Research Centre (BRRC)
Photo by Hans Bendtsen
Deliverable No. 1 – Final version – 30 April 2011
ERA-NET ROAD Project "Optimization of thin asphalt layers"
ERA-NET ROAD is a consortium comprising national European road administrations. Its
purpose is to strengthen European road research by coordinating national and regional
research programmes and policies.
In 2009 ERA-NET ROAD issued a call for tenders on a transnational research project titled
“Optimization of thin asphalt layers”. The project is coordinated by a Project Executive Board
with representatives of six European road administrations:
Mats Wendel (chair), Swedish Transport Administration, Sweden
Thomas Asp (secretary), Swedish Transport Administration, Sweden
Tony K. Andersen, Ministry of Transport, Danish Road Directorate, Denmark
Jostein Aksnes, Norwegian Public Roads Administration, Norway
David Lee, Department for Transport, Highways Agency, United Kingdom
Christian Pecharda, FSV; Austrian Association for Research on Road - Rail –
Transport, Federal Ministry of Transport, Innovation and Technology, Austria
Christiane Raab, Empa, Swiss Federal Laboratories for Materials Testing and
Research, Swiss Federal Roads Authority, Switzerland
The Project Consortium consisting of the Danish Road Institute, the Belgian Road Research
Centre and the Swedish National Road and Transport Research Institute won the tender and
the project was initiated 1 July 2009. The researchers carrying out the project are the authors
of the present report with support from colleagues with special expertise.
The project acronym has been "OPTHINAL", derived from the project title “Optimization of
thin asphalt layers”.
The present report documents the results of a State-of-the-Art Review which is the first
project deliverable. This is the second edition of the report, the first edition was delivered in
May 2010.
The Final Report of this project was delivered 2010-12-17, and updated (edited) 2011-03-15.
Authors: Jørgen Kragh (ed.), Erik Nielsen, Erik Olesen (DRI), Luc Goubert, Stefan
Vansteenkiste, Joëlle De Visscher (BRRC) and Ulf Sandberg, Robert Karlsson (VTI).
Title: Optimization of Thin Asphalt Layers - Final Report
It will be published by ERA-NET Road.
ABSTRACT ............................................................................................................................. IX
ABBREVIATIONS AND ACRONYMS ................................................................................. X
EXECUTIVE SUMMARY................................................................................................... XIII
INTRODUCTION .......................................................................................................... 1
PURPOSE AND LIMITATIONS .................................................................................. 2
Purpose ....................................................................................................................... 2
Limitations ................................................................................................................. 2
BASICS AND HISTORY OF THIN ASPHALT LAYERS .......................................... 3
What characterizes a thin asphalt layer? .................................................................... 3
Terminology and standards ........................................................................................ 7
3.2.1 General ................................................................................................................. 7
TAL in this project ............................................................................................... 7
Terminology and standardization in CEN ............................................................ 8
Terminology in PIARC ...................................................................................... 10
Historical review ...................................................................................................... 10
METHODS AND SOURCES OF DATA COLLECTION.......................................... 12
Overview .................................................................................................................. 12
Questionnaire and interviews ................................................................................... 12
Literature searches and databases ............................................................................ 12
4.3.1 Transportation research database ....................................................................... 12
Personal networking and contacts with colleagues ............................................ 13
Essential conferences ......................................................................................... 13
Work with TAL within the consortium .............................................................. 14
SUMMARY OF THE INVENTORY .......................................................................... 15
Overview of responses ............................................................................................. 15
TAL usage according to the inventory ..................................................................... 16
THE USE OF THIN ASPHALT LAYERS ................................................................. 20
Time trends ............................................................................................................... 20
The drivers for the use of thin asphalt layers ........................................................... 21
Where thin asphalt layers may not be the optimum choice...................................... 24
Words of caution related to wear resistance and bearing capacity..................... 24
Words of caution related to grooved rumble strips ............................................ 24
Types of roads and traffic where thin asphalt layers are useful and popular ........... 25
Families and categories of thin asphalt layers .......................................................... 26
Introduction ........................................................................................................ 26
6.5.2 Product families according to the CEN 13108 series of standards .................... 27
Proprietary pavements ........................................................................................ 28
Asphalt rubber thin layers .................................................................................. 28
Remixing and other recycling into thin layers ................................................... 31
Effect of climate on the preference for thin asphalt layers....................................... 32
Dry versus wet climate ....................................................................................... 32
Hot versus cold climate ...................................................................................... 33
Thin asphalt layer performance in cold region – Example from Japan ............. 35
Interaction with studded tyres ............................................................................ 36
Countries or regions with policy-driven use of TAL ............................................... 38
Use in ERA-NET ROAD countries compared to other countries ............................ 40
PERFORMANCE: SPECIAL KEY PROPERTIES .................................................... 42
Macrotexture ............................................................................................................ 42
Skid resistance .......................................................................................................... 42
International survey ............................................................................................ 42
French experience .............................................................................................. 42
Danish measurements ......................................................................................... 43
Experience in the United Kingdom .................................................................... 44
Experience in the Netherlands............................................................................ 45
Influence on road traffic noise ................................................................................. 45
Noise optimization of thin layers ....................................................................... 46
The Kastrupvej example..................................................................................... 47
French experience .............................................................................................. 49
Influence on rolling resistance ................................................................................. 50
Thickness and weight advantages ............................................................................ 52
Economy................................................................................................................... 53
The special properties of asphalt rubber thin layers ................................................. 55
Cost Considerations............................................................................................ 55
Energy Savings Considerations .......................................................................... 57
Reduced Thickness Design Considerations ....................................................... 57
Life Cycle Cost Considerations.......................................................................... 58
Main conclusions from Swedish seminar in 2010 ............................................. 58
Overall assessment of TAL performance in the U.K. .............................................. 59
Other performance properties................................................................................... 60
SUMMARY OF PERFORMANCE............................................................................. 61
MIX DESIGN AND OPTIMIZATION OF TAL ........................................................ 63
CONCLUSIONS .......................................................................................................... 64
ACKNOWLEDGEMENTS ......................................................................................... 67
REFERENCES ............................................................................................................. 68
APPENDIX A: QUESTIONNAIRE AND INTERVIEWS..................................................... 76
Questionnaire ........................................................................................................... 76
List of addressees for questionnaires ....................................................................... 78
Results of the interviews .......................................................................................... 79
A.3.1 Summary of interview with an expert from the Netherlands ............................. 79
A.3.2 Summary of the interview with a Norwegian panel of specialists ..................... 79
A.3.3 Summary of the interview with a Swiss specialist ............................................. 83
A.3.4 Summary of the interview with a panel of Austrian specialists ......................... 86
A.3.5 Summary of the interview with an Italian specialist .......................................... 89
A.3.6 Summary of interviews with Danish specialists ................................................ 90
APPENDIX B: OTHER ASPECTS OF PERFORMANCE .................................................... 94
DURABILITY ......................................................................................................... 94
B.1.1 Ageing ................................................................................................................ 94
B.1.2 Acoustical durability .......................................................................................... 95
B.1.2.1 Danish long time experiment on acoustical durability ....................................... 95
B.1.2.2 Norwegian long time experiment on acoustical durability ................................ 98
B.1.2.3 Acoustical durability of Dutch thin layers ......................................................... 99
B.1.3 Wear by traffic ................................................................................................. 100
B.1.3.1 General ............................................................................................................. 100
B.1.3.2 Danish project HOLDA ................................................................................... 100
B.1.4 Typical lifetimes ............................................................................................... 101
B.1.5 Possibilities of restoring a deteriorated TAL ................................................... 102
MISCELLANEOUS PERFORMANCE PROPERTIES ....................................... 103
B.2.1 Other surface characteristics ............................................................................ 103
B.2.1.1 Megatexture and evenness ............................................................................... 103
B.2.1.2 Light and glare properties ................................................................................ 104
B.2.1.3 Drainage, splash and spray ............................................................................... 105
B.2.2 Emission of inhalable particulate matter into the air........................................ 108
B.2.3 Structural strength and bearing capacity .......................................................... 109
B.2.4 Use of natural resources ................................................................................... 110
B.2.5 Construction-related aspects ............................................................................ 111
B.2.5.1 Paving equipment ............................................................................................. 111
B.2.5.2 Laying time and traffic closure ........................................................................ 112
B.2.5.3 Maintenance and rehabilitation ........................................................................ 114
B.3. SUSTAINABILITY ASPECTS ............................................................................. 115
B.3.1 Sustainable construction ................................................................................... 115
B.3.2 Criteria for determining end-of-life.................................................................. 115
B.3.3 Recycling properties ......................................................................................... 115
B.3.4 Energy consumption and emissions during transportation .............................. 118
B.3.5 Potential effects of climate change................................................................... 119
CE-marked products according to the EN 13108 series ........................................ 120
Proprietary products ............................................................................................... 120
Noise-related classification .................................................................................... 121
C.3.1 HAPAS in the United Kingdom ....................................................................... 121
C.3.2 C road in the Netherlands .................................................................................... 122
C.3.3 SRS system in Denmark ................................................................................... 123
C.3.4 Dutch COP testing system................................................................................ 124
C.3.5 CEN Noise Classification System .................................................................... 124
ERA-NET ROAD initiated a transnational research project titled “Optimization of thin
asphalt layers”. The DRI-BRRC-VTI Consortium was trusted with carrying out the project
and began with a State-of-the-Art report covering, among other things, a literature study and
an inventory of experience with using thin asphalt layers (TAL). The results of this phase of
the project are given in the present report.
This study was limited to thin asphalt mixtures with a maximum thickness of 30 mm, which
means that surface dressings or slurry seals were outside the scope of the project. Neither
were top layers of double-layer porous pavements considered as TAL, even though such top
layers often are 20-30 mm thick. Mix design and optimization was the subject of another
study in this project and is therefore not treated here.
The main conclusions are that the application of TAL is certainly worthwhile, in particular as
a renewable “skin” of a stable road construction having sufficient bearing capacity. The skin
serves road users’ need for sufficient skidding resistance and other important functions.
The use of TAL seems to be increasing due to the needs of road administrations for cost
effective maintenance of the road infrastructure which, in many ways, are consistent with the
needs for lower traffic noise levels in residential areas near major roads, which may be one of
the positive effects when a TAL is applied.
The environmental impact of road transport CO 2 emission is currently widely discussed. Road
surface characteristics are one of the parameters that influence rolling resistance and hence
energy consumption and CO 2 emission. TAL offer relatively low rolling resistance because of
their favourable surface texture. Therefore they may have a positive impact on the reduction
of CO 2 emissions.
The report attempts to evaluate the various properties of TAL, comparing with more conventional and traditional surfacings such as dense asphalt concrete 11 or SMA 11. TAL in general
comes out somewhat better than the references in most respects; for example concerning cost,
use of nature resources, rolling resistance, and traffic noise emission.
However, there are also problems, for example concerning durability under severe traffic, and
bearing capacity (very little extra capacity provided by TAL). If studded tyres are used the
wear of TAL is usually significantly worse than the wear of thicker pavements with larger
chippings although there are special TAL with larger chippings than is usual in TAL, which
are well adapted for this purpose.
There are several special types of TAL; not the least a huge variety of commercial products
offered on the market; so-called proprietary TAL. A special type of thin layer is the asphalt
rubber surfaces, presently under trials in Sweden for adaptation to north European climate and
conditions. In USA, some of these layers are paved as thin as half an inch (approximately 12
mm) and yet they provide very good performance.
In the table below, acronyms and abbreviations used in the report are explained.
Annual Average Daily Traffic
Asphalt Concrete
Attestation of conformity
Abrasion value according to EN 1097-9 (percent fragmented material in
Nordic abrasion test)
Asphalt rubber (binder which contains 15-20 % by weight of rubber
American Society for Testing and Materials
Very thin asphalt concrete (used in CEN, abbreviation from the French
name Beton Bitumineux Tres Mince)
Beton Bitumineux Ultra Minces (ultra thin asphalt concrete)
European Committee for Standardization
Close Proximity (method) (tyre/road noise measurement close to a test tyre,
often using a trailer)
DAC 11
Dense(-graded) asphalt concrete, with maximum aggregate size 11 mm
decibel, unit for sound pressure level, re. 20 μPa
dB, the sound signal has been filtered by the standard A-weighting filter
Dutch Centre for Transport and Navigation
C road
Dutch correction for road surface influence on traffic noise
European Organization for Technical Approval
European Technical Approval
Highway Agency Product Approval Scheme (UK)
Health, Safety and Environment (Occupational consideration etc. etc.)
International Organization for Standardization
Life Cycle Cost
Life Cycle Assessment
Mean profile depth according to ISO 13473-1
Nominal Maximum Aggregate Size (typically the smallest sieve size which
allows all the aggregate to pass the sieve).
Porous Asphalt, sometimes also called drainage asphalt
Porous asphalt concrete, pervious asphalt, drainage asphalt
Polycyclic Aromatic Compounds (often used to describe cancerogenic
compounds originating from coal tar products).
World Road Association (comes from the original name Permanent
International Association of Road Congresses; the latter not used any more)
Polymer modified bitumen (typically related to EN 14023)
Pavement management system
Reclaimed Asphalt (European term linked to EN 13108-8)
Reclaimed Asphalt Pavement (US term)
Rolling resistance
Rolling resistance coefficient
SoA or SotA
Stone mastic asphalt (Europe), or Stone matrix asphalt (USA)
Statistical Pass-By (method) (statistical analysis of individual vehicle noise
levels from several vehicles measured 7.5 m from the centre of the lane)
Swedish Transport Administration (formerly SRA = Swedish Road Adm.)
Thin asphalt layer or Thin asphalt layers
Terms of Reference
Transport Research Laboratory in the UK (nowadays called TRL Limited)
Technical University of Gdansk in Poland
Ultra mince (ultra thin), from family BBUM
Ultra Thin Layer Asphalt Concrete, according to EN 13108-9, EOTA
Guideline, or proprietary product
Ultra Thin Bonded Wearing Course
Working Group (such as in CEN or ISO)
Very open asphalt concrete, porous asphalt
ERA-NET ROAD initiated a transnational research project titled “Optimization of thin
asphalt layers”. The DRI-BRRC-VTI Consortium was trusted with carrying out the project
and began with a State-of-the-Art report covering, among other things, a literature study and
an inventory of experience with using thin asphalt layers (TAL). The results of this phase of
the project are given in the present report.
The main conclusions are that the application of TAL is certainly worthwhile, in particular as
a renewable “skin” of a stable road construction having sufficient bearing capacity. The skin
serves road users’ need for sufficient skidding resistance and other important functions.
The use of TAL seems to be increasing due to the needs of road administrations for cost
effective maintenance of the road infrastructure which, in many ways, are consistent with the
needs for lower traffic noise levels in residential areas near major roads. This may be one of
the positive effects when a TAL is applied.
The environmental impact of road transport CO 2 emission is currently widely discussed.
Road surface characteristics are one of the parameters that influence rolling resistance and
hence energy consumption and CO 2 emission. TAL in general offer relatively low rolling
resistance because of their favourable surface texture. Therefore they may have a positive
impact on the reduction of CO 2 emission. Furthermore, since TAL only requires a thin skin of
material, superior materials can be used in lesser quantities, thus reducing the road
administrator induced CO 2 emissions associated with extraction, manufacturing and transport
of these materials.
The use of TAL in Europe seems to increase although available statistics make it difficult to
distinguish between TAL and other hot-mix asphalt.
In the "perpetual pavement" concept the philosophy is that the pavement base has eternal
bearing capacity and is paved with a thin long-lasting "skin" of surface layer which eventually
– due to water, ageing and other climatic action – must be renewed from time to time.
TAL provide a “skin” with favourable functionalities, such as noise reduction potential, relatively low rolling resistance, relatively good anti-spray properties and efficient light reflection.
This has accelerated the use of general product categories and proprietary products addressing
these demands, also implying relatively high sustainability and low construction as well as
maintenance costs. The fast laying of TAL implies shorter closure to traffic and this favours
the use of TAL. Provided the pavement base is of appropriate quality TAL offer solutions to
many of the functionalities mentioned above and this is probably why there is immense
interest in products of this nature.
Despite this, one shall not forget the problems and limitations associated with TAL. For
example, bearing capacity is only marginal in many cases, and resistance to wear from
studded tyres is poor, unless one uses large maximum aggregate sizes, in which case the
thickness needs to be relatively high. The open-textured or even porous kinds of TAL may
offer very good noise properties, but at the expense of limited durability under heavy traffic
load; for example in sharp curves or at large gradients. The porosity will also quickly get
clogged by dirt. Another problem worth mentioning is that it is not possible to dismantle TAL
by milling with the techniques we have at hand without downgrading the material.
The project group sent out a questionnaire to key experts. Unfortunately, the project team
received rather limited response and an interview round was not very successful either. Most
respondents mentioned noise reduction as their primary motivation to apply TAL. Also cost
reduction and fast paving operations seem to be important motivation, like good resistance of
TAL to skidding and rutting. A few respondents mentioned durability problems as a disadvantage.
Policies on applying TAL vary substantially from country to country. Countries with extensive usage of TAL include the UK, Switzerland, Sweden, Norway and the Netherlands. Also
Denmark and Austria use TAL relatively extensively. In Sweden, to some extent also in
Norway and Switzerland, TAL is used on the highway network, while in the Netherlands and
the UK usage is limited to municipal roads or city streets as well as provincial or trunk roads.
The authors have tried to evaluate the potential advantages and disadvantages of applying
TAL compared with standard DAC 11 or SMA 16. The three most important advantages are
Noise reduction
Lower cost
Less required working space
Other advantages include, for example, higher skid resistance (at low and medium speeds),
improved sustainability in most respects, better rut resistance and faster laying.
The three most important disadvantages are
Weather conditions while laying TAL are more critical
Dismantling by milling with present techniques downgrades the material
Susceptibility to cracking related to substrate deficiencies
Other disadvantages include, for example, susceptibility to ravelling, delamination and frost
damage, manual laying is not possible, shorter lifetime, and rather low skid resistance in wet
weather for some TAL variants.
Mostly, TAL were found to have fine skid resistance properties, although a few exceptions
were reported. Durability of skid resistance was excellent according to a UK study, but
several studies reported poor acoustic durability. Still, experience of TAL is too short and
much longer time series need to be studied.
The sensitivity of TAL to the weather conditions during laying has been mentioned as a major
disadvantage. Road administrations and contractors are often forced - by numerous factors to apply TAL during cold weather and then their durability may be low. Perhaps this can be
counteracted by optimizing the laying process.
TAL must be CE-marked in order to be marketed as complying with an EN 13108-series
product standard. These standards specify asphalt mixes, not their final application on the
road. The ETAG 16 guideline on ultra thin layers intends to deal with the entire process,
including paving operations and the final application. Products complying with this guideline
will probably be an additional route for CE-marking in the future. The impact of this CEmarking on the market still has to be seen in the daily practise of procuring asphalt materials.
At present, classification of pavement acoustic characteristics is limited to declaring product
properties in Denmark, the Netherlands and the UK. CEN work on this is at an initial stage.
No system exists for checking pavement product conformity of production concerning its
noise characteristics.
At least two countries represented in the PEB are highly interested in the effect on TAL of the
exposure to traffic with vehicles using studded tyres. The present review concludes that
aggregate quality and the proportion of large aggregate are the main parameters determining
wear resistance of dense and gap-graded asphalt concrete wearing courses.
The report also discusses the concept of using Asphalt rubber (AR) pavements as thin layers
in the various pavement systems. In a broad context, a multitude of benefits of using an AR as
a pavement preservation strategy were enlisted, including less reflective cracking, reduced
maintenance, excellent durability, less raveling, good rut resistance, good skid resistance and
smooth ride, better drainage facilities, reduced tyre/road noise, cost effectiveness, beneficial
engineering use for old tyres, and higher energy efficiency.
However, it must be noted that these are the merits of AR typical for the conditions in the
USA. In Europe so far, there has been a different scenario when one takes into account the
derived benefits of AR, as observed in relation to a few similar pavement strategies of
comparable quality. Nevertheless, ongoing research and practical applications, the results of
which so far are reasonably positive, will determine whether the AR concept could be a
success in Europe as well.
The report indicates that the actual achievement of both excellent functional properties and
good durability (lifetime) is nothing which comes easily. In practice, it is often difficult to
realise both these requirements simultaneously since they are frequently in conflict with each
other. The information made available through this report should, therefore, serve as a basic
guideline for achieving the best compromise between the goals.
Thin asphalt layers have been used in several European countries and countries outside
Europe for more than 15 years with promising results. They seem to be cost effective pavements, fast to build and may have good surface properties. Development in recent years
shows that thin asphalt layers can reduce noise, increase traffic safety (skid resistance and
forward visibility during wet condition) and be durable.
In the frame of ERANET ROAD II, a call was issued in 2009 for a comprehensive study of
this type of road surface. The overall purpose of the study should be to optimize thin and very
thin asphalt surfacing with thickness up to 30 mm.
The first phase of the project to study such wearing courses consists in gathering detailed
information on the use of thin layers and the experience obtained in Europe and elsewhere, if
possible. A literature review has been carried out for this purpose. This review of the literature
has been supplemented by an inventory amongst distinguished specialists because the experience of Project Consortium partners is that although knowledge and experience can be found
in regular literature, essential information may often remain hidden for example in unpublished research reports from institutes and contractors.
The present State-of-the-Art review has been organized so that after an initial overview of the
character and history of thin asphalt layers, the method for searching literature and supplementary information is described and the result of the inventory is summarised. The use of
thin asphalt layers and the main reasons for preferring them is dealt with in Chapter 6 while
Chapter 7 describes performance characteristics such as skidding resistance, noise and rolling
resistance, characteristics which are incentives for future application of thin layers. Durability
is a major concern and is dealt with in Annex B. This Annex also describes various surface
characteristics such as evenness and the ability to reduce splash and spray, but also
constructional aspects such as laying time and traffic closure are mentioned.
Annex B also contains information about sustainability aspects, including recycling of
materials and the influence on climate change of constructing and maintaining road
infrastructure with thin asphalt layers. Methods and systems for checking the performance of
thin asphalt layers are described in Annex C.
A concise summary of findings concerning all performance aspects is presented in Chapter 8.
An essential challenge is to investigate how asphalt technological aspects and performance
characteristics are interrelated (some are conflicting) and how they can be optimized. The
questions as to how, why and where to use thin asphalt layers is discussed in the present
report, and a preliminary overview of advantages and disadvantages (risks) involved when
applying thin asphalt layers is attempted.
Page 1 of 124
The general purpose of the project is to collect, analyse, summarize and report information on
asphalt surface layers 10 - 30 mm thick, including all types of hot-mix design and application
methods. Proprietary and special products, like types with rubber-modified bitumen shall be
dealt with. Focus shall be on asphalt technology aspects and on performance characteristics
assessed to be important for future application of thin layers.
The present report on the State-of-the-Art Review carried out by the DRI-BRRC-VTI
Consortium aims at documenting the results of literature reviews and an inventory of expert
knowledge not yet documented in regular literature.
In the second phase of the work the Project Consortium will attempt at recommending
improvement and optimization. Gaps in available knowledge shall be identified and directions
shall be pointed at for research needed to fill these gaps.
The study was limited to thin asphalt mixtures, which means that surface dressings or slurry
seals were outside the scope of the project.
Focus was on hot mix asphalt. In this connection, so-called warm mix asphalt was considered
a special hot mix application.
The maximum thickness was more or less arbitrarily defined to be 30 mm.
Double layers are composite constructions and they have not been considered as being TAL.
Thus, double layer wearing courses, even the top layer of such pavements, were outside the
scope of the present project.
Mix design and optimization was the subject of another study in this project and is therefore
not treated here.
Page 2 of 124
What characterizes a thin asphalt layer?
A "Thin Asphalt Layer" (TAL) is characterized by three main features:
Feature number 1 – Various gradations:
The layer consists of a mix of aggregate particles which follow a size distribution called a
gradation, and the aggregates are bonded together by a bituminous binder to a homogenous
plant produced mixture. In some cases the application of the material on the road intentionally
leads to a heterogeneous layer structure.
Feature number 2 – Surface characteristics in focus:
The Thin Asphalt Layer is the surface layer of the road which means that the functionality –
in broad terms – is directed towards the interface between the road structure and tyre or shall
display certain properties towards the drivers of the vehicles or other road users.
Feature number 3 – Typical thickness 10-30 mm:
By the Terms of Reference (ToR) for this project the Project Executive Board (PEB) has
defined it geometrically as asphalt layers with a thickness between 10 and 30 mm.
To solve some of the "interface issues" mentioned under feature 2 several distinct asphalt
material solutions have been developed over the years – sometimes with quite different
approaches. The geometrical definition in the ToR cuts horizontally through many separate
product types that normally will not be considered close together. The project team has tried
to cope with this problem in this State-of-the-Art report and decided to use the abbreviation
TAL for "Thin Asphalt Layer(s)" (both singular or plural) to accommodate and abide to the
requirement of the PEB. For this reason it must be understood, that TAL will not generally be
defined or found in literature outside the scope of this project, as technical literature will not
use thickness in the same distinct way as in the ToR of this project.
The abovementioned three features have further consequences or implications that can be
derived as follows:
Regarding Feature number 1 – Various gradations:
A homogeneous plant produced mixture means that neither surface treatments nor slurry
surfacings will be covered by this project as they have other characteristics. A surface
treatment is not a product but the result of an "in-situ" process and does not when applied
have a homogeneous structure with a well defined layer thickness. Slurry surfacings are also
the result of an "in-situ" process and even though a well defined layer thickness can be
Page 3 of 124
accomplished slurry surfacings are more seen as a maintenance solution rather than as a
normal "construction element" in the tool box for new construction.
Regarding Feature number 2 – Surface characteristics in focus:
When Thin Asphalt Layers are identical with surface layers a lot of technical and functional
requirements become important in different situations. Texture, skid and wear resistance,
noise reducing capability, light reflection can be highlighted just to mention a few items that
emerge because Thin Asphalt Layers act as interface to the road users and the surrounding
environment. Many of the chapters in this report give details on these aspects.
Regarding Feature number 3 – Typical thickness 10-30 mm:
The geometrical definition in Terms of Reference is presumably intended to give a description
of Thin as opposed to Thick. But it will be demonstrated in the next subparagraph and
throughout the project that the "thin-thick" issue can have different meaning due to variations
in national historical tradition – often linked to traffic intensity level. An example: Some years
ago what in France would be considered as a thin asphalt concrete surface layer would at the
same time in Denmark be seen as a thick asphalt layer.
The general description of the asphalt materials or pavement solutions covered by TAL will
involve characteristics such as:
Particle size distribution – normally called "aggregate gradation" – which like in Figure
3.1 on the Y axis will show the amount (mass percentage) passing different square sieves
as a function of the sieve size plotted on a logarithmic scale in mm
Binder properties which can be a range of grade of binder, binder percentage and in
some cases whether or not the bituminous binder is modified with polymers, crumb
rubber or other chemicals to enhance the material properties (the rheology) of the
resulting binder
TAL can be "standard asphalt" just applied in layers between 10 and 30 mm but for some
products the pavement solution is unavoidably linked to special features during the
paving operation. This is the case for a special family of asphalt materials that requires
special conditions for tack coating during paving
Normally, a tack coat shall be applied on the underlying layer, with the intention to
enhance the adhesion of the TAL to the underlying layer. Thus, paving of TAL is a "twocomponent" procedure, where first a tack coat is applied, followed by applying the TAL
Feature 2 defines TAL as a surface layer and in order to define the role of TAL in the total
pavement structure Figure 3.2 shows a typical cross-section on a normal pavement structure
used for medium to high traffic intensity. Figure 3.3 and 3.4 show typical surface of two types
of TAL. The surface in 3.4 was selected essentially based on its very favourable noise
Page 4 of 124
SMA 8, BBTM 8 Class 1, BBTM 8 Class 2, UTLAC 8 and AC 11
AC 11 dense
Percentage passing [%]
BBTM 8 Class 1
BBTM 8 Class 2
16 [mm]
Figure 3.1: Grading curves for TAL from test sections in Denmark (noise-reducing pavements
and reference surface AC 11d from the first block of test sections near Herning, Denmark).
Figure 3.2: Cross-section of a typical asphalt pavement designed for 10 years of traffic.
Page 5 of 124
Figure 3.3: Close-up picture of a two years old thin UTLAC pavement with 8 mm maximum
aggregate size. The size of the black and white squares is 10x10 mm.
Figure 3.4: A close-up view of the surface of a proprietary thin asphalt layer called
“Microflex”, as paved on Kasteelenlaan in Ede; four years old when the photo was shot. The
aggregate size is 2-6 mm. The coin is 23 mm in diameter.
Page 6 of 124
Beginning from the top, TAL is the upper layer serving the interface to traffic. The next layer
is typically an asphalt binder course or upper bituminous base layer followed by a lower
bituminous base. These three layers are the bituminous bonded part of the pavement where
the last two provide the bearing capacity. TAL may or may not contribute to the bearing
capacity. Further down come the unbound bases of gravel and sand an in the end the soil
foundation. Cement or lime treated base layers can also in some countries be used, but in
relation to TAL in the pavement structure in this report the main question is: Will TAL
contribute to the bearing capacity or not? This will be largely independent of whether the
sufficient bearing capacity originates from a stabilized base or from the bituminous layers.
The important thing is that sufficient bearing capacity is available at the interface between
TAL and the underlying structure.
Terminology and standards
3.2.1 General
In a survey of performance of UK "thin surfacings", the following categorization was used
[Nicholls et al, 2006]:
– Paver-laid surface dressing (PLSD)
Ultra-thin surfacings developed in France
– Thin asphalt concrete (TAC)
Generally with polymer-modified binder
– Thin stone mastic asphalt (TSMA)
Generally unmodified bitumen with fibres
– Multiple surface dressing (MSD)
Binder and aggregate applied separately
– Micro-surfacing (MS)
Thick slurry surfacing, generally with modified binder
3.2.2 TAL in this project
TAL or Thin Asphalt Layers, as defined within this project, are a large family of in-plant
produced hot asphalt mixes which can be paved with nominal layer thickness between 10 and
30 mm (see also section 3.1). Consequently, TAL may meet the specifications set out in one
of the product standards of the EN 13108-series (‘Bituminous mixtures – Material Specifications’) or criteria described in the ETAG 16 guideline.
As already mentioned, cold applications such as slurry surfacing as defined in EN 12273
‘Slurry surfacing – product standard’ are not considered within this project.
It should also be mentioned that the top layer of double-layer porous asphalt, where the top
layer is usually 20-30 mm thick, is not considered as TAL, since it is considered as a
composite structure together with the bottom layer.
Of the five UK categories presented in the previous section, only the three first would fit the
definition in this project.
Page 7 of 124
3.2.3 Terminology and standardization in CEN
With respect to the European normalization framework (CEN) the following documents are of
EN 13108-1 ‘Bituminous mixtures – Material specifications – Part 1: Asphalt concrete’
EN 13108-2 ‘Bituminous mixtures – Material specifications – Part 2: Asphalt concrete for
very thin layers’
EN 13108-3 ‘Bituminous mixtures – Material specifications – Part 3: Soft asphalt’
EN 13108-4 ‘Bituminous mixtures – Material specifications – Part 4: Hot rolled asphalt’
EN 13108-5 ‘Bituminous mixtures – Material specifications – Part 5: Stone Mastic
EN 13108-6 ‘Bituminous mixtures – Material specifications – Part 6: Mastic Asphalt’
EN 13108-7 ‘Bituminous mixtures – Material specifications – Part 7: Porous Asphalt’
prEN 13108-9 ‘Bituminous mixtures – Material specifications – Part 9: Bituminous
mixture for Ultra-thin Asphalt concrete (UTLAC)’
ETAG 16 “Guideline for European Technical Approval of Ultra Thin Layer Asphalt
The European standard EN13108-1 constitutes the widest product standard in the EN 13108series of ‘Bituminous mixtures – Materials specifications’ (see also 6.5.2). The standard sets
out the specifications for many types of asphalt concrete for both surface, binder and base
layers. In some countries (e.g. Denmark) this standard also embraces the majority of surface
layers defined in the ToR as TAL.
The European standard EN13108-2 constitutes a second part of the EN 13108-series of ‘Bituminous mixtures – Materials specifications’ (see also 6.5.2). The standard sets out the specifications for asphalt concrete for very thin layer applications. Asphalt concrete for very thin
layers is to be used for surface courses with a thickness of 20 to 30 mm.
The European standard EN13108-3 constitutes a third part of the EN 13108-series of ‘Bituminous mixtures – Materials specifications’ (see also 6.5.2). The standard sets out the specifications for asphalt concrete with very soft binders. The gradations are similar to Asphalt
Concrete in EN 13108-1, but several traditional used gradations could not be fitted in EN
13108-1. As many countries did not have experience with these types in the more traffic
intensive part of middle Europe the product standard has a dominant Nordic influence. The
majority of the mixes in this standard fall within the definition of TAL.
The European standard EN13108-5 constitutes a fifth part of the EN 13108-series of ‘Bituminous mixtures – Materials specifications’ (see also 6.5.2). The standard sets out the
specifications for stone or split mastic asphalt or SMA. Such mixes are characterized by a
discontinuous grading, a high mastic content and an open surface texture. The thickness of an
SMA layer may vary between 25 and 50 mm. Therefore, some SMA-C mixes 1) (nominal
SMA-C and SMA-D refer to Belgian tender specifications, where SMA-C is essentially an SMA 10 (maximum aggregate size is 10 mm), and SMA-D is essentially an SMA 0/6.3.
Page 8 of 124
layer thickness between 30 and 40 mm) and SMA-D mixes (nominal layer thickness between
25 and 30 mm) may be considered as thin asphalt layers.
The guideline for European Technical approval of ultra thin layer asphalt concrete ETAG 16
is at present still a document being drafted within EOTA, but in parallel product specifications
for UTLAC are presently subject to CEN Enquiry as prEN 13108-9 after having been drafted
in CEN/TC 227/WG 1 (the working group for asphalt materials under CEN Standardization
Technical Committee 227, working on road materials in general).
There are important differences within the European standardization between on one hand the
asphalt mixes such as very thin layers or SMA-D and on the other hand the ultra thin asphalt
layers as defined in the guideline:
With respect to very thin layers or SMA-D mixes there are test methods available as
described in the EN 12697-series: ‘Bituminous mixtures – Test methods for hot mix
asphalt’. The 12697-series, consisting of about 45 parts, provides the standardization (and
harmonization) of methods to test asphalt mixes. The series does not set out any material
specification. Moreover, one disposes of product standards (EN 13108-series) which set
out the product specifications. Finally, the latter standards are also linked to EN 13108-20
which defines the ‘initial type testing’ or ITT-study and to EN 13108-21 which sets out
the ‘Factory Production Control’ or FPC (internal production control assessment). There
is nothing available with respect to paving operations or a quality control of the asphalt
mix following the road works.
With respect to ultra thin asphalt layers or UTLAC:s only a guideline document (ETAG
16) is being drafted within EOTA. The document does not contain any strict specifications but a guide how to characterize the materials in order to fulfil the essential requirements of the European Community. In contrast with the EN 13108-series, where the
responsibility for the product standard stops when the “loose mix on the lorry” leaves the
asphalt plant, the EOTA guideline describes the entire process, including the paving
operations. This is due to the fact that the paving operations for ultra thin layers are more
critical in comparison with thicker asphalt layers (>20 mm) in order to guarantee a good
quality and durability of the application.
Quite recently (March 2010), a prEN 13108-9 ‘Bituminous mixtures – Material Specifications
– Part 9: Bituminous mixture for Ultra Thin Layer Asphalt Concrete (UTLAC)’ has been
drafted. The new product standard will deal with the specifications of ultra thin asphalt layers
characterized by a nominal thickness of 10 to 20 mm. In analogy with the other standards
within the EN13108-series no requirements related to paving operations or quality control
following the road work will be included. The latter topics remain within the scope of the
ETAG guideline.
It is important to remember that the route through EOTA ends up in an ETA which assures
the asphalt producer/asphalt contractor that the company can market a proprietary product not
fulfilling a European product standard but through the ETA is capable of having the product
CE-marked according to the Road Construction Product directive. Applying for an ETA is a
company and product specific application route that is costly. It will perhaps in many cases be
more economical for the companies if they can reach the CE marking status through the EN
13108-9. The reason is that they will presumably already have third party inspection on the
Page 9 of 124
premise with regard to CE marking activities linked to other asphalt materials. So, in practise,
the ETAG will only have importance for UTLAC products falling outside the framework of
EN 13108-9.
The above standards are of direct interest in relation to a large part of the TAL family since
their technical properties or characteristics (such as surface texture, skid resistance, noise
reducing capacity, etc) to a large extent can meet the functional properties set out especially
for this kind of application. Nevertheless, other types of asphalt mixes are also applied with a
nominal layer thickness of 30 mm or even less. Such asphalt mixes are described in other
parts of the EN 13108 series and will be discussed later in this report.
3.2.4 Terminology in PIARC
Within the PIARC (World Road Association) two terms are related to thin asphalt layers:
‘thin asphalt surfacing’: a bituminous surface course with an average laying thickness of
30 to 50 mm (this seems to be related to the UK definition; see 3.2.1)
‘very thin asphalt surfacing’: a bituminous surface course with an average laying thickness of 20 to 25 mm
As a synonym for surfacing the term overlay is used in both cases. It is unclear if the term
surfacing or overlay covers hot mix asphalt applications or also includes cold surfacing
techniques. Moreover, a layer thickness up to 50 mm for a thin application seems rather
Historical review
This paragraph describes in very general terms the historical evolution of TAL, as a detailed
historical background is beyond the scope of this report. Even though it may be interesting, it
is difficult to cover all the different asphalt products as the evolution is influenced by national
developments and regional conditions for traffic intensity, climate and material availability.
MacAdam designed a number of farm road constructions which at that time meant a vast
improvement of the bearing capacity but the surface was in general an unbound surface.
As society evolved the population grew and the need for road transport increased. This meant
that people living close to the roads were becoming annoyed by dust problems and the traffic
intensity demanded some form of capping layer with a better load spreading ability and better
riding comfort. This started an era for bituminous solutions. In some cases macadam with a
penetrating oil or tar product or oil gravel road was sufficient to combat the dust and bearing
capacity problem on the rural roads, but when higher demand was imposed from traffic more
sophisticated products were needed. Inspiration of the present day concept of Soft Asphalt
evolved from the early developed materials in this period.
In approximately the first third of the 20th century, in larger towns mastic asphalt was applied.
Sometimes on cobble stone pavement or on top of cement bound materials 2 .
If sufficient bearing capacity is present, thin mastic asphalt layers might still be a possibility for TAL. For road
applications, price considerations often prohibit the use of mastic asphalt as other solutions are more economical.
Page 10 of 124
Then the development in society called for even higher bearing capacities and the Marshall
dense-graded asphalt concrete entered the scene. The concept of TAL was gradually forgotten.
In the later part of the 20th century the demands from road users increased on parameters such
as evenness, skid and wear resistance etc., which called for more sophisticated products and
applications. Especially in countries or regions where good quality aggregate was scarce a
technical-economical incentive for producing a cheaper asphalt material for the bituminous
base quickly developed, and the concept of a thinner surface layer (TAL) was reintroduced.
This layer could be accepted to be more expensive but was more tailor-made to the needs of
the traffic. Due to the oil crisis in the 1970's, wearing courses requiring less material
(especially bitumen) became more interesting.
The Novachip process is a surface treatment process that places a thin (12–20 mm) 3 and
course aggregate hot mix wearing course over a polymer modified tack coat/membrane, using
only one piece of special equipment for the paving process. It was developed in 1986 by
Screg Routes STP in France, intended to increase skid resistance and seal old pavement
surfaces and was suggested as a better alternative to surface dressing, with no loose chippings
after the paving process. The following years, a French patent (1989) and a US patent (1990)
were registered. The NOVACHIP TM is a registered trademark of Societe Internationale
Routiere, which is a subsidiary of Screg Routes STP.
Novachip was then used widely in Europe and USA. However, it never became popular in
Denmark as the patent rights were too expensive, while it became popular in the next-door
neighbour Sweden. At the present time the patent rights seem to have been released.
By and by in the 1990’s, several other proprietary TAL appeared on the market, mainly in
France, Netherlands, and the U.K. In the U.K., a system called HAPAS (Highway Authorities
Product Approval Scheme) for type approval of proprietary road surfacings was established in
the late 1980’s, which appeared to be useful as an acceptance system for TAL in the U.K..
The first TAL 4 certificate was issued in 2000. At the present time, there are 37 "thin
surfacings" certified by HAPAS in the U.K.
In the last two decades, other functionalities like anti-spray properties, light reflection, noise
reducing capabilities, low rolling resistance, etc., have come along and accelerated both
general product categories and proprietary products that address these demands. In the
"perpetual pavement" concept it is taken in its extreme form, where the philosophy is that the
base of the pavement, having eternal bearing capacity, is paved with thin long-lasting "skins"
of surface layer which eventually – due to water, ageing and other climatic action – must be
renewed from time to time.
TAL offers the solution to certain or many of the functionalities mentioned above and this is
why there is immense interest in products of this nature.
In Sweden it is commonly applied as a 20-30 mm thick layer
The commonly used UK term for thin asphalt layer (TAL) is ”thin surfacing”
Page 11 of 124
In order to collect information on the use of thin asphalt layers and the experience obtained in
applying such surfacing, a search for and review of literature was carried out. The transportation database was searched for terms containing keywords assumed to cover the field of
interest here while proprietary products were searched via Google. In parallel with this search,
an inventory was made among institute and contractor specialists who were requested to
reveal experience not yet published in regular literature.
The inventory is described in Section 4.2 and the literature search is described in Section 4.3.
Questionnaire and interviews
A questionnaire was made comprising a list of eight questions about thin layers in order to
obtain information which cannot be found in literature. The questionnaire can be found in
Annex A.1. The number of questions was deliberately kept low in order to have a reasonable
response ratio. The following topics were treated:
Types of TAL used
Motivations to use TAL
Basic documents
Tests on TAL
Experience with TAL
Research needs
The questionnaire was sent to a selection of 23 persons which were expected to have
experience in the field (see Annex A.2). These persons originate from eleven different
European countries. Seven addressees had responded when this was written.
At the same time, a number of experts were contacted and requested to take part in an indepth interview about the subject. In this way, experience from six European countries was
The results of the answers to the questionnaire and the interviews are woven into the
following chapters and annexes, while a summary is given in Chapter 5.
Literature searches and databases
Transportation research database
A literature search was carried out for information on TLA in the Transportation Database
(OECD, ITRD, International Transportation and Research Documentation). Products/trade
names for TAL were searched in the Google Product or Google.
Page 12 of 124
The keywords listed in Table 4.1 were used as search terms in the Transportation Database.
The trade/products names listed in Table 4.2 were used as search terms in Google
The keywords and phrases listed in Table 4.1 were used as search terms in the Transportation
Database. The keywords/phrases listed are those that came out with a positive search result.
The question marks ("?") are used as a truncation. This means that when searching for “Thin
asphalt layer?” the search results will also include references on thin asphalt layers.
Table 4.1: Keywords and key phrases for the search in the Transportation Database.
Thin asphalt layer?
Ultra thin process? and asphalt layer?
Asphalt rubber technology
Very thin asphalt layer?
Ultra thin asphalt layer?
Modern thin asphalt layer?
Thin hot mixture asphalt layer
Thin application and wearing course? asphalt
Cost effective and wearing course?
Modified thin asphalt layer?
Very thin surface layer?
Ultra thin overlayer?
Thin stone mastic asphalt or Thin SMA
Table 4.2: Trade/product names for the search in Google Products/Google.
TbK or TB K
Personal networking and contacts with colleagues
A search for TAL was carried out through personal contacts to key persons from asphalt
companies in Denmark. These companies have been working on research and development of
TAL for many years. Some of the asphalt contractors have an international parent company
and key persons were asked to promote contacts.
Also a search for information about TAL was carried out among known proprietary TAL
products to obtain updated information on the properties. The companies are known to
improve their products regularly to keep up with international development of TAL.
Essential conferences
Two of the authors attended a BAFU-OFEV Tagung (seminar) about implementation of
Swiss low-noise road surfaces (of which many were thin layers) held in Olten, Switzerland, 9
September 2009.
Page 13 of 124
The Dutch Road Research Institute (DVS) in June 6th 2007 organized a seminar in Delft with
the title “Experience with very thin noise reducing layers” as a part of the IPG project on
optimizing thin layers for noise reduction. DRI participated in this interesting seminar.
Work with TAL within the consortium
The three members of the consortium have been active during the latest 10 years in various
aspects of research, development and full scale testing and demonstration of TAL products.
This has been in the framework of EU projects like SILVIA [Thomsen et al, 2006], SILENCE
[Nielsen et al, 2006] and in national and international projects. DRI has focused on
developing TAL pavements with optimized noise reduction without compromising aspects
like traffic safety (skid resistance), durability, etc [Bendtsen & Raaberg, 2007].
In the SILVIA project what might be called a first generation of noise-reducing thin layers
were developed and tested in full scale on three urban roads [Thomsen et al, 2006]. In the
SILENCE project a second generation of noise-reducing thin layers were developed and
tested in full scale on an urban road [Thomsen et al, 2008].
In cooperation with the Dutch Road Research Institute (DVS), DRI has developed and tested
first and second generation noise-reducing thin layers for application on highways [Bendtsen
et al, 2008; Bendtsen et al, 2009]. As a crucial part of this research information on international experience with TAL has been collected and integrated in the research [Bredahl Nielsen
et al, 2005; Bendtsen & Raaberg, 2005].
Page 14 of 124
Overview of responses
The following persons responded to the questionnaire:
Ian Walsh, Jacobs Engineering (UK) Ltd, Maidstone (United Kingdom)
Jacob Groenendijk, KOAC.NPC, Apeldoorn (the Netherlands)
Peter J. Andersen, Vejdirektoratet, Copenhagen (Denmark)
Jostein Aksnes, Statens vegvesen Vegdirektorat, Oslo (Norway)
Olivier Ripke, Bundesanstallt für Straßenwesen, Bergisch-Gladbach (Germany)
Erik Van den kerkhof, Colas Belgium, Brussels (Belgium)
Cliff Nicholls, TRL Ltd, Wokingham, Berkshire (United Kingdom)
Berwich Sluer, BAM Wegen nv, Bunnik (the Netherlands)
Kenneth Olsson, Skanska, Farsta (Sweden)
The following persons were interviewed:
Wim van Keulen (van Keulen advies), Vlijmen, the Netherlands
Jostein Aksnes (Norwegian Public Road Administration - NPRA, Road Directorate),
Geir Refsdal, Rolf Johansen and Jan Lindahl (NPRA, Eastern Region), Olle R. Larsen
(Kolo Veidekke), Norway
Alain Jacot (Société d’Analyses & Contrôles Routiers, SACR), Zürich, Switzerland
Johann Litzka (Austrian Association for Research on Road - Rail – Transport, FSV),
Jürgen Haberl (PMS-Consult, Engineering Office for Traffic and Infrastructure), Peter
Riederer (BPS, Oberösterreichische Boden- und Baustoffprüfstelle GmbH), Michael
Kostjak (Swietelsky BaugesmbH), Ronald Blab (Vienna University of Technology),
Gaetano Licitra (ARPAT, Agenzia Regionale per la Protezione Ambientale della
Toscana, and University of Pisa), Pisa, Italy
Bjarne Bo Jensen (NCC Roads A/S), Lars Ladehoff (Colas Danmark A/S), Uffe
Mortensen (Pankas A/S), Niels Christoffersen and Uffe Mortensen (Inreco A/S),
The Danish experts were interviewed by telephone calls by Erik Olesen, the others were
interviewed at personal meetings by Luc Goubert.
Most respondents mention noise reduction as motivation for using TAL. Other important
motivations are cost reduction and speed of laying. Reduced working space, good skid resistance and good rut resistance are also mentioned as advantages. Few respondents mention
durability problems as a disadvantage.
Various ranges of thickness are mentioned, ranging from “15 – 25 mm” up to “25 – 35 mm”.
Voids contents vary from 1 % up to 24 %.
The most common bad experience with TAL is ravelling (Figure 5.1 left part) and delamination (Figure 5.1 right part). Also mentioned are substrate-related cracking and frost susceptibility. The good experiences are the noise reduction and the excellent skidding resistance. Various respondents mention specific solutions which have been developed to prevent or at least
to reduce the problems with TAL, such as modified bitumen and/or the use of CaO (hydrated
lime) to reduce ravelling susceptibility, a special tack coat to prevent delamination, etc.
Page 15 of 124
Figure 5.1: Most common problems with TAL: ravelling (upper left) and delamination (upper
right) (photo courtesy of Ian Walsh, Jacobs Engineering (UK) Ltd) compared to TAL in very
homogene and good condition (N201 in Heemstede, the Netherlands).
All respondents see an important role for TAL in the future, mainly because of the high
priority given to the traffic noise problem. Annex A presents the details of the interviews.
TAL usage according to the inventory
Figure 5.2 shows the area of TAL in absolute figures for some European countries, intended
to give an impression of to what extent TAL are used in the countries, and the large variation.
The areas have to be considered as approximations and in some cases also as minima; e.g. for
Belgium only the area built by Colas is taken into account and in Austria only the TAL on
highways is accounted for, as no other values were available. For Sweden, only the national
roads are included; while it is known that also communal roads use TAL. Countries not
indicated in Figure 5.2 may also have TAL; for example, in France and Spain there is a
substantial amount of TAL.
Figure 5.3 shows the percentages of the main road (national) network covered with TAL for
some countries (no data available from GB and NL).
Page 16 of 124
Area TAL as estimated by experts (x million
Thin layers
Very thin layers
Figure 5.2: Area in millions of square metres of thin and very thin asphalt layers in some
countries. Not all roads in the countries concerned are included in these estimates, so the
estimated areas must be considered minimum values. Note that according to the interviewed
Italian expert, there are no TAL at all in Italy.
Percentage main road network covered
with TAL
Figure 5.3: Percentage of the main road network covered with TAL; estimates based on
interviewed experts. No data are available for GB. NL has a large percentage of TAL on
communal roads, but nothing on the main roads.
Page 17 of 124
An uncertainty factor in these diagrams is the different definitions used for TAL. For
example, in the U.K. a "thin surfacing" is defined as “a proprietary bituminous product with
suitable properties to provide a surface course that is laid at a nominal depth of less than 50
mm” [Nicholls et al, 2006]. This is quite different from the TAL that this report is intended to
cover. Other countries may have corresponding national definitions or common terminology
which is inconsistent with the TAL definition in this project, and this might have influenced
the collected data.
An illustration of how the data in Figure 5.2 is distributed across Europe appears in Figure
5.4. Note that the lack of data for many countries may distort the picture.
Figure 5.4: The data in Figure 5.2 illustrated on a European map, with estimated total TAL
area in million m2 indicated for each country. Colour codes: Red = Very high usage, yellow =
high usage, blue = low to moderate usage, white = no usage or data missing. Note that, for
example, France and Spain use TAL extensively, but quantitative data are missing and thus
the colour is white.
Page 18 of 124
The Swedish data contain both what is named TSK (tunnskikt), i.e. "real" TAL, but also socalled remixing, which also meets the definition of TAL. It is worth noting that as much as 5
million m2 of TSK is laid annually in Sweden; most of it is TSK with NMAS = 16 mm, amd
the trend is increasing [Olsson, 2010].
See further Annex A.
Page 19 of 124
Time trends
In the latest decade many countries have moved into a direction where the end-user (the road
user) has been put in focus. It has been a challenge both to provide different functionalities of
the pavement to the road user and to maintain the achievements from former optimization of
other functionalities.
In general the road user has no insight in very important pavement properties like bearing
capacity because the road user naturally focuses on visible aspects and experience gained
from driving on the road (texture, riding comfort, evenness etc.). This means that for the road
users’ perception of the road it is virtually the road surface as an "interface" to traffic that
For the road owner/road administration achieving the desired bearing capacity in addition to
the road users' demands in the most economical way is normally also linked to optimizing the
cost through the use of higher specialized surface layers over thick lifts of stabilized bases or
bituminous bases. Due to the very specialized nature of the surface layers they are normally
more expensive per tonne than standard dense graded asphalt concrete which means that in
order to remain in a competitive market they need to function in smaller layer thicknesses.
On the other hand if a specialized surface layer provides the majority of the needs of the road
user, the road owner/road administration can obtain the necessary bearing capacity by using
materials and techniques that even though they may have to be used in thicker lifts over the
total pavement structure will give a more economical solution.
In the lower part of the construction many options can come into play depending on the local
situation and especially material availability. At the bottom soil with low bearing capacity can
be upgraded by stabilization with lime. Unbound base layers can also be treated with Portland
cement or foam bitumen again in order to upgrade materials (perhaps locally available) with
inferior bearing capacity to a higher level.
The thickness of the bituminous base can in these situations be reduced. If surplus amounts of
reclaimed asphalt materials are present, the application of bituminous bases with a high
amount of recycled material can also be a manner to optimize the total pavement economy.
Combining the incentive of the road users and the road administrations both groups have a
common interest towards using TAL for surface layers, so it is a general trend that will only
become more and more pronounced.
Even though the general trend is towards TAL it is important to highlight one exception and
to give a word of caution.
The exception is when a high level of noise reduction is desired and the pavement structure is
demanded to provide as much noise reduction at the source as possible. In this case thick lifts
of porous asphalt is needed. And yet again even here TAL shows its possibility. The concept
of two layer porous asphalt includes a thin layer of small aggregate porous asphalt as a kind of
Page 20 of 124
filtering layer over a thick lift of coarse aggregate. This combination has shown excellent
noise-reducing ability.
The word of caution concerns the fact that even though we have a great variety of highly
specialized TAL at our disposal the TAL concept has not solved all problems yet.
As a "Rule of Thumb" a layer thickness is typically between 2.5 and 4 times the nominal
maximum aggregate size (NMAS). The general explanation for this is:
If you pave a layer thickness smaller than 2.5 times NMAS you will experience problems
in achieving the sufficient and desired level of compaction and thereby endangering the
durability of the pavement.
If you pave a layer thickness larger than 4 times NMAS you will risk that the traffic
loadings will generate permanent deformation (rutting) in the layer.
This means that TAL defined in the project with the defined layer thickness of 10 – 30 mm
normally have a NMAS of approximately 12 mm or smaller.
Wear by studded tyres and snow chains is still a major problem in many regions. If your
winter conditions call for and allow the extensive use of studded tyres then you might find
that TAL with small or medium-sized aggregates is not the optimum surface layer. A "Rule of
Thumb" associated to resistance against wear from studded tyres is "The larger the aggregate
the better"; typically starting from 16 mm and up. In Norway and Sweden, TAL with 16 mm
max aggregate (such as Novachip or similar) are frequently and increasingly used despite
severe studded tyres exposure and works fine. These have thicknesses of 20-30 mm [Olsson,
2010], which then "violates" the first explanation to the "Rule of Thumb" above.
Although, the time trend is an increasing use of TAL in most countries, it is not a consistent
trend. For example, in the U.K. where TAL and SMA have largely replaced the much noisier
and traditional hot rolled asphalt (HRA), especially in municipalities and cities, the trend is a
return to HRA at the expense of TAL and SMA, since HRA requires even less of the highquality aggregate that TAL needs, following the problems to get access to such material in the
U.K. [Lee, 2010].
The drivers for the use of thin asphalt layers
There are many drivers for the use of TAL. This can be deduced from the initial part of the
present report. This paragraph will sum up drivers, some of which are generally applicable
while others are linked to either certain traffic situations or pavement materials. Depending on
your national or regional view you can prioritize them or rank their importance in different
order. Some of the drivers act in combination. Because of this some overlapping in the
explanations may occur.
Road user's demands (in general):
The demands from road users, the transport sector and the society constitute an ever increasing incentive to optimize the functionalities towards the "interface" to traffic. For some
functional properties it means that only the "contact surface" is of interest. As the product
becomes more and more sophisticated the price increases, and when only the contact surface
is important the layer thickness is minimized in order to stay competitive. An example is
Page 21 of 124
colour of the roads surface where light colour aggregates reflect the light from street lamps in
order to save energy (typically electrical energy in urban areas).
Differentiate materials in surface layer and bituminous base layer
Many functional properties are linked to "contact area" or the texture depth of the surface
layer. When the properties are not associated to "bulk" properties of the surface layer there is
a very strong technical and economical incentive to minimize surface layer thickness as the
surface layer normally is more expensive that underlying layers. Examples of properties in
this respect can be aggregate characteristics like polish and wear resistance.
Saving natural resources
As natural aggregates of premium quality for road surfaces are either scarce or have to be
hauled in over large distances at the expense of fuel consumption (energy and CO 2 ) and cost
TAL can provide a reduction in cost and a better management of natural resources by a slow
exploitation rate.
Local/inferior materials for base layers
The TAL-concept leads to the possibility to use an increased amount (relatively) of local
material for base layers (less energy consumption and transport cost). Through upgrading
local inferior materials to provide an enhanced contribution to the bearing capacity the TAL
concept can permit technical-economical pavement structure solutions. Upgrading can take
place either through lime stabilization of the soil or through treatment of the unbound
pavement layers with Portland cement or foam bitumen.
TAL – functionalities following from small maximum aggregate sizes and open texture
The TAL-concept means that the nominal maximum aggregate size is 12 mm or less.
Lowering the NMAS to 8 or 6 mm (or even smaller) when you traditionally have used 12 – 16
mm (or even larger) you will experience improved functionality with respect to noisereducing capability (if open texture types are chosen) and also reduced rolling resistance. The
last point at the moment is the subject for several projects worldwide.
Some of the proprietary products within TAL that have been optimized for their noise
reducing capability have shown excellent initial noise-reducing effects without having the
same clogging problem as porous asphalt. Open textured TAL are virtually "self-cleaning" by
the action of the passing traffic.
At some point in the development of noise reducing surface layers many road engineers had
serious reservations towards lowering the NMAS as it was expected to bring down the skid
resistance to an unacceptable level. This has been proven not necessarily to be the case;
according to French experience, where the thinner layers have increasing macrotexture and
better skid resistance than their conventional counterparts [Bendtsen & Raaberg, 2007].
TAL integrated in the "Perpetual Pavement" concept
The rationale behind the concept of "Perpetual Pavement" is that roads with very high traffic
intensity are constructed "once and for all" with respect to bearing capacity while the surface
layers are renewed from time to time. TAL is unavoidably an integrated part of this concept.
Even though the surface layers are intended to be repaved, extremely high durability is needed
because high traffic intensity makes road closures due to maintenance unacceptable on the
grounds of traffic delay costs etc. In these cases even very costly artificial aggregate (like
Page 22 of 124
bauxite) may be technically economical due to the high polishing and wear resistance. With
such expensive aggregates it is important to minimize the amount used per m2.
TAL – speeding-up paving operations
A major favourable feature of TAL is that paving operations may be performed relatively fast,
which is very important in order to reduce obstruction to traffic. Fast paving operations also
mean cost savings. Part of the reason is the faster cooling of thin layers.
TAL - versatile concept in urban areas due to layer thickness
Resurfacing city streets with kerbs between the road and the pedestrian areas or bicycle lanes
will normally involve a lot of manual labour in order to raise the kerbs and adjust the gutters.
This part of the rehabilitation job is an expensive part relative to the paving operation. With
sufficient kerb height TAL perhaps provide two or even three resurfacings before the curb has
to be adjusted for height.
In urban areas flyovers and bridges with roads beneath can cause the authorities extra maintenance burdens. If the road passing under the bridge shows fretting (loss of bituminous
mortar) and ravelling (loss of stones) but has sufficient bearing capacity a new surface layer
can be paved without having very much influence on the allowed clearance under the bridge.
Necessary operations like either excavating the road or elevating the bridge are very costly
parts of a rehabilitation project if clearance height is on the brink of being insufficient.
TAL – when mass is important
The report will not deal with specific surfacing solutions for bridges, as especially bridge
decks for orthotropic steel bridges are more or less tailor-made in combination with protective
membrane built-up. But for surfacing cement concrete bridges asphalt materials developed for
the use in TAL may provide a possibility due to their lower mass per square unit compared to
traditional pavement solutions.
Now some special drivers for TAL will be mentioned which are associated to specific asphalt
material solutions and therefore not applicable to all TAL.
UTLAC – a combination of protective membrane and surface layer
If maintenance shall be performed on an old, worn road surface showing initial fretting and
ravelling and perhaps even hairline surface cracks due to ageing (not due to insufficient
bearing capacity) a special type of TAL can be an optimum solution. The application of
UTLAC means than a large amount of unbroken polymer modified bitumen emulsion is
spread just before the hot open graded asphalt mix is placed creating an extremely good bond.
As the amount of residual binder from the emulsion is approximately twice the amount from a
normal tack coat operation the UTLAC will provide an almost perfect protective membrane
shielding the underlying layers from water from above.
Soft asphalt – in case of insufficient frost heave protection
It can not be stressed often enough that the TAL-concept will only provide durable solutions
if the surface layer is placed on a structure with sufficient bearing capacity. But on the tertiary
road network and in domestic areas where the overwhelming majority of traffic is passenger
cars insufficient protection against extreme frost heave can be the case. Either due to the fact
that the rural road has developed over time and is not properly constructed or the domestic
area was only meant for lighter traffic and a certain risk for frost heave was acceptable from
Page 23 of 124
an economical point of view. In these cases with low to medium traffic the soft asphalts
provide a technical-economical solution. Due to the soft binder the material follows the
movements of the road structure. In case of extreme frost heave cracks may occur but during
the first 3-5 years the binder is capable to close the minor cracks. This is done due to a
combination of elevated summer temperature and "massaging" (after compaction) from the
traffic until the binder is aged to an extent where the soft asphalt more or less turns into an
asphalt concrete.
Where thin asphalt layers may not be the optimum choice
6.3.1 Words of caution related to wear resistance and bearing capacity
The previous section describes a number of reasons why thin asphalt layers may be preferred.
However, some words of caution may be justified: thin asphalt layers may be a poor choice of
pavement type in some cases.
First of all, when wear by tyres is exceptional, such as on high-volume roads in countries and
regions with a substantial proportion of studded tyres, given that aggregate quality is the
same, TAL will generally be worn faster than the thicker pavements do, in case they use small
aggregate sizes (and TAL usually do so). See further Section 6.6.4.
A problem exists towards open-textured surfaces which is not unique for TAL but is true for
all open-textured mixes irrespectively of their layer thickness. Since TAL often are opentextured, the problem applies to TAL somewhat more than to conventional AC or SMA. In
domestic and urban areas where parking or sharp turns occur the surface may have some
problems to endure the shearing forces under the tyres. This is especially pronounced in
parking areas when the drivers use the servo steering ability available in many cars.
When relatively high bearing capacity of the wearing course is needed, thin layers will not
provide this. See further Section 9.3.
6.3.2 Words of caution related to grooved rumble strips
A recently noticed problem in Denmark and perhaps elsewhere, that contains some potential
problems for TAL in several respects, is a conflict between different departments within the
road administrations. TAL can be produced with excellent noise reducing capability even in
very thin layers. Therefore TAL is often chosen as an optimum solution by road engineers
considering taking technical, economical and environmental (noise) aspects. Then when the
traffic markings shall be placed (thermoplastic stripes between opposing lanes and at the kerb)
another department focusing on traffic safety and activities to avoid collisions and drivers
falling in sleep demands that rumble strips shall be grooved into the pavement. See an
example in Figure 6.1.
Page 24 of 124
Figure 6.1: Grooved rumble strips in TAL on a two-lane highway near Skive, Denmark.
This is of course important, but has two major disadvantages:
1. The grooves increase the noise level tremendously when the traffic occasionally hits them
and thereby render the investment in a noise reducing pavement of less value, unless the
rumble strips are milled with a sinusoidal shape, see [Kragh, 2007a]
2. Making grooves in a TAL takes away a large percentage of the layer thickness that was
intended to protect a perhaps leaner asphalt binder course beneath. This is especially a
technical point when the grooved traffic markings are made to prevent head-on collisions,
because this could be just over the longitudinal joint of the asphalt materials (both surface
layer and for the layers beneath) decreasing the durability of the pavement.
Types of roads and traffic where thin asphalt layers are useful
and popular
The concept of thin asphalt layers is very versatile when applied correctly and when the
requirements for its use are fulfilled. The requirements for a durable TAL are:
The underlying road structure has to provide sufficient bearing capacity. Depending on
the nature of the asphalt material may or may not contribute, but if the surface
characteristics are open textured or contain a soft binder no additional contribution can
be expected.
Page 25 of 124
The underlying structure shall have the correct evenness and longitudinal and transversal
profile, as there is virtually no levelling possibility in TAL without endangering the
durability of the layer.
A tack coat should be applied on the underlying layer before the TAL mix is applied. All
asphalt materials need to be tack coated to the substrate and this is also true for TAL. For
UTLAC the polymer modified tack coat is especially good.
If the conditions are met different types of TAL can cover all traffic categories and all
climates (wet/dry and hot /cold), but the resistance against studded tyres is limited due to the
small aggregate size.
Many of the TAL applied today are used for their noise reducing and "anti-splash" capability
which is linked to the open texture of the surface.
It is a problem to extract meaningful figures for the use and application of various types of
asphalt as many countries interpret the European standard differently. As an example the
Road Standards in Denmark (the interpretation national guideline of the European product
standards) specify several asphalt types like AC (EN 13108-1), BBTM (EN 13108-2) and
UTLAC (prEN 13108-9) according to asphalt concrete. As asphalt concrete include both
surface layers and base layers it is not possible to extract data covering the use of TAL. By
engineering judgement 95 % of all hot mix surface layers applied in Denmark are in
accordance with the TAL definition of this project.
Figures 5.2-5.4 show the total area of applied TAL in a few EU countries.
Families and categories of thin asphalt layers
Based on the geometrical definition of TAL used in this report (hot mix asphalt with a nominal layer thickness < 30 mm), a particular TAL may correspond to one of the European
product standards of the EN 13108-series or the ETAG 16 guideline as listed in Section 3.2.2.
It should be noted that based on the functional properties frequently associated with TAL
some hot mix asphalt types are generally of less importance such as those described in EN
13108-1 or EN 13108-6. Other types are only used in particular countries or regions within
Europe. For example: soft asphalt mixes in Scandinavia or hot rolled asphalt in Ireland or UK.
Reclaimed asphalt or RA is described in EN13108-8 ‘Bituminous mixtures – Material specifications – Part 8: Reclaimed Asphalt’. Since this standard doesn’t deal with a new asphalt
layer as such, no further attention will be paid to this standard. Moreover, the use of RA in
TAL throughout Europe is probably either very restricted or even not permitted.
In contrast to the situation in Denmark, a large part of all hot mix surface courses in Belgium
are not in accordance with the definition of TAL in this report. In Belgium, only about 5 % of
all surface layers meet the criteria of EN 13108-2. The use of porous asphalt according to EN
13108-7 is also very limited (about 1 % of all surface courses).
Page 26 of 124
Product families according to the CEN 13108 series of standards
In general, the hot mix asphalt types as described in the EN 13108-series and the ETAG
guideline can be divided in two major categories (except for mastic asphalt):
Hot mix asphalt with a sandy skeleton
Hot mix asphalt with a stony skeleton
The major difference between the two families is the grading curve. Asphalt mixes with a
sandy skeleton are characterized by a continuous grading curve while mixes with a stony
skeleton are typically based on a discontinuous grading.
The standards for Asphalt Concrete and Soft Asphalt embrace both types of grading curves
even though the discontinuous ones are not as open as some of the other types. The important
distinction between Asphalt Concrete and Soft Asphalt is not the gradation but the binder
type. Asphalt Concrete is produced with a penetration grade bitumen 160/220 or harder and
gives a contribution to the bearing capacity. On the other hand Soft Asphalt uses softer grades
of binder than 160/220 (either penetration graded or viscosity graded) and the materials will
not contribute to the bearing capacity of the pavement.
Hot mix asphalt with a sandy skeleton is described in EN13108-1 (Asphalt concrete). Such
mixes contain a large amount of sand and are characterized by a low void content (typically
between 2 and 7 %). Therefore, asphalt concrete is a closed mixture which doesn’t meet the
functional properties for TAL. Nevertheless, asphalt concrete is used both in base courses and
top layers.
Hot mix asphalt with a stony skeleton is described in EN13108-2, EN13108-5, EN13108-7,
prEN13108-9 or the ETAG 16 guideline. Such mixes are characterized by a high (>70 %)
stone contact and are gap graded (discontinuous grading). Generally, such mixes do contain a
high void content (up to 25 % for porous asphalt). SMA, as described in EN13108-5, contains
a high mastic (filler + binder) content and is characterized by lower void content, although it
still maintains an open surface texture. The latter mixes, due to their favourable surface
properties (such as open texture) are frequently chosen as an appropriate top layer.
Mastic asphalt according to EN 13108-6 is characterized by very high mastic (filler +
bitumen) content. It is produced at very high temperatures (typically between 180 – 240 °C).
Mastic asphalt is a very dense mixture (void content < 3 %). In contrast to all other types of
asphalt, mastic asphalt does not require any compaction. In Belgium, mastic asphalt is not
used as a top layer (only on bridges, roof parking decks, foot paths or for quick maintenance
works during wintertime).
Due to the normally used size of the embedded chippings in Hot Rolled Asphalt (EN 131084) and the type of heterogeneous structure from top to bottom of the pavement layer, the
material has only theoretical and not practical possibility to fit within the definition of TAL in
the ToR.
Page 27 of 124
Proprietary pavements
A commercial market for thin asphalt layers has been established in Europe for more than 10
years now; especially in the Netherlands, France and the UK, and especially focused on the
local and urban communities, such as the road and street departments of medium-sized and
large cities. Road construction companies have been very active in producing products that
answer to this market. Table 6.1 presents a list of such products. These TAL focus on a
certain property that they are optimized for; most commonly low noise.
Figure 3.4 shows the surface of a typical advanced proprietary TAL, named Microflex.
Table 6.1: Example of proprietary TAL offered by European road construction companies.
These examples are from France, the Netherlands and the UK.
Micro-Top 0/6
Micro-Top 0/8
Stil Mastiek
Konwé Stil
Nobelpave HS
Microflex LS
Microflex HS
Microville HS
Asphalt rubber thin layers
Asphalt Rubber (AR) is a mixture of 80 % hot paving grade asphalt (bitumen) with 20 %
ground tyre rubber produced from waste tyres; also referred to as crumb rubber. Figure 6.2
shows typical crumb rubber granules used in AR pavements. The rubber is added to bitumen
at a high temperature and mixed usually in 45-60 minutes with a high shear mixer. After the
mixing interaction, the hot AR product acquires unique elastomeric properties. The hot AR is
then pumped into a conventional hot
plant and mixed with aggregate and
placed like a conventional asphalt
concrete (in USA called Hot Mix
Asphalt or HMA), except for a few
significant differences. The differences
relate to the gradation of the mineral
aggregate and the percent binder.
Figure 6.2: Crumb Rubber Granules –
used in Swedish AR binders in 2007;
free of wire and other contaminants.
Photo by courtesy of Mats Wendel,
Swedish Transport Administration.
Page 28 of 124
The AR hot mix is generally either a gap graded or an open graded mix. The total rubber
content is approximately 1.8 % by weight of the total mix in both types. The gap graded mix
contains about 7-8 % AR binder and is generally placed as the final structural course 50-75
mm in thickness. The AR gap graded mix (often referred to as ARAC) is a volumetric mix
design using the Marshall procedure and designed for 5.5 % air voids. This mix is fairly
similar to the SMA pavements used in Europe, except for the AR binder.
A second type of AR mix called the asphalt rubber open graded friction course (simply
referred to as asphalt rubber friction course, ARFC) was developed to improve the long term
durability and provide good wet weather skid resistance, and less splash and spray. In addition
to these good properties it has also been observed (in US conditions) that the ARFC reduces
cracking, provides a very good ride quality, and is noise-reducing. The open graded version
contains generally 9 % binder, has 18-22 % air voids, and is placed as the final wearing
course with a thickness ranging from 12.5 to 25 mm. The maximum aggregate size is typically 9.5 mm. Figure 6.3 shows typical pavement cores of two different sections: one that has
three different asphalt concrete (AC) layers, namely, ARFC, ARAC and an existing/new
conventional AC layer along with the corresponding typical range of thicknesses as shown in
Figure 6.3(a). In Figure 6.3(b), one can observe a two-layer system that comprises an ARFC
overlay upon a Portland cement concrete (PCC) structure with corresponding typical
thicknesses for each layer. Figure 6.4 shows the typical appearance of the surfaces of ARAC
and ARFC.
Thus, a combination of lower aggregate maximum size in the AR, and inclusions of crumb
rubber in the mix, consequentially incrementing an increase in the percentage of AR binder
content, has made it possible for AR mixtures variants (ARFC and ARAC to a large extent) to
be used as thin asphalt layers in the USA and elsewhere. In the USA, AR has been
successfully used for over 40 years. Besides being a standard material, it has offered many
benefits including less reflective cracking, less maintenance with more durability, less
ravelling, good rut resistance, good skid resistance, smooth ride, less splash & spray and
better drainage, cost effectiveness, and is environment-friendly. The film thickness of an AR
binder is about 19-36 micrometers compared to about 9 micrometer for a typical HMA [Way
et al, 2010].
12½ – 25 mm
12½ – 25 mm
50 – 75 mm
150 – 250 mm
Existing / New
Figure 6.3: Corresponding thicknesses of: (a) Triple-layer ARFC, ARAC, AC, and (b) Twolayer ARFC, PCC.
Page 29 of 124
Figure 6.4: Typical Field Sections of (left) Asphalt Rubber Gap Graded (ARAC), (right)
Asphalt Rubber Open Graded (ARFC) [Source: Arizona Department of Transportation].
Between 1989 and 2004, a total of 135 projects were constructed using the ARAC and ARFC
mixes, totalling over 33333 lane-km in Arizona. All these projects were built on interstates
(motorways) and major arterial roads. As an extra benefit, the ground tyre crumb rubber from
over fifty million tyres has been recycled into the Arizona in-service pavements.
Following the successful experience of the Arizona projects, the technical know-how was
adopted by the California Department of Transportation (CALTRANS) to develop a reduced
thickness design guide for AR mixes using AR gap graded mix as the standard [Kirk &
Holleran, 2000]. The various designs have shown that the multi-layer systems using AR
binders such as shown in Figure 6.3 have proven to be the most cost-effective. In California,
AR is specified to include 18-22 % crumb rubber by total mass of the AR blend. The most
commonly used asphalt rubber products in California are RAC-O (Rubberized Asphalt
Concrete – Open graded) and RAC-G (Rubberized Asphalt Concrete – Gap graded) with a
thickness of 30-60 mm, based on limited experience with thicker layers and economic considerations.
Many other states in the USA, such as Florida, Texas, South Carolina, Washington, Louisiana
and New Mexico are also using AR. Furthermore, it is also worth mentioning that a few states
in Canada have tried AR pavements.
AR using the wet process, was introduced in southern Europe many years ago; primarily in
Portugal and Spain. Currently, the length of roads paved with AR pavements in Portugal and
Spain is substantial. The grading of the aggregate and terminology in the Portuguese
pavements seems to be a little different than that used in USA, which also is the case for the
rubber content. The gap-graded version used in Portugal has a void content of around 14-16
%, which in USA would be similar to an open-graded type. Other countries that have been
using AR pavements are Austria, Italy and the Netherlands. Slovenia is starting-up a project
on AR in 2010.
Page 30 of 124
In 2007, the Swedish Road Administration (SRA), now reorganized into the Swedish
Transport Administration (STA), started a development project to investigate the potential of
rubber asphalt pavements for Swedish highways. The objective of the project was to investigate if the AR concept developed and tried in USA could be an interesting alternative for
asphalt pavement construction and maintenance in Sweden. It was / is hoped that the longterm effect of the project would be reduced annual and life cycle costs, noise reduction and
lower emission of particles as well as improved friction.
The types of AR pavements tested in Sweden so far are listed in [Sandberg, 2010]. The
Swedish term for AR is GAP for the dense-graded version and for GAÖ for the open-graded
version. Reference pavements have been either SMA 16 or SMA 11.
The thickness of the Swedish AR pavements is generally 30 mm for 11 mm aggregate size
and 25 mm for 8 mm aggregate size. Technically, this means that the AR pavements with
NMAS 8 or 11 mm are thin layers. Test sections have been laid on various highways, ranging
from arterials in cities with 50 km/h posted speed, over regional and national highways with
70-90 km/h posted speed, up to motorways with 110 km/h posted speed. Traffic volume
ranges from very small on regional highway to very high on motorways.
Trial sections monitored annually were laid in 2007, 2008 and 2009 [Sandberg, 2010]. Significant noise reductions compared to conventional SMA were noted only for the open-graded
variant. Conclusions from a recent seminar focused on the Swedish experience of asphalt
rubber pavements suggest that asphalt rubber is a very promising concept also for European
conditions, and even in winter climates where studded tyres are used. For seminar
presentations and other information, see [Gummiasfalt, 2010]. The main conclusions from the
seminar are presented in Section 7.7.5.
As asphalt rubber has been demonstrated in USA to be a very cost-effective pavement in a
very thin version (<20 mm thick) and some of the Swedish tests sections were only 30 mm
thick and could well be made thinner, this may be one of the most interesting types of TAL.
Therefore, a closer examination of present results would be justified in a later follow-up
Remixing and other recycling into thin layers
Hot in-place recycling is a process in which an existing, often rather worn, asphalt course is
carefully heated with infrared radiation, in order that a certain thickness of the layer is pickedup and the material disintegrated, then mixed again, perhaps adding some fresh material and
finally relaying the mix in a new binder or wearing course. This is made in an integrated
operation, in one machine or a train of machines. During repaving the properties of the asphalt
surface course are changed. The new layer will be plane and even, as after laying a new
asphalt layer.
A normal TAL is not created in this way, as it requires high-quality aggregate and binder.
However, the resulting layer of hot recycling may serve to give an existing pavement an extra
lifecycle; albeit not having the quality of a regular TAL. As the layer thickness often is 30
Page 31 of 124
mm or somewhat lower, however, one must consider these recycled surfaces as a special type
of low-quality TAL, which is the justification for mentioning them here.
There are four different methods for hot in-place recycling:
1. Reshape = repaving without changing the composition of the asphalt (no new material is
2. Repaving = the surface profile is restored as in the reshaping method and a new add-mix
is applied and both layers are compacted at the same time
3. Remix = hot in-place recycling with a change to the composition of the asphalt (some
new material is added and mixed with the old material)
4. Remix plus = hot in-place recycling with modified asphalt composition and laying of a
new surface layer (made of fresh material) with a second screed
During hot in-place recycling the asphalt course is gently heated up by an infrared heater. A
special construction machine, the remixer, picks up the heated asphalt material, mixes it and,
depending on the method used, immediately relays it with a variable screed (reshape) or
thoroughly mixes it with hot bitumen and/or additional mix and then lays it (remix). In the
"remix plus" method (also called "remix compact") the loosened asphalt material is thoroughly mixed with hot bitumen and/or additional mix and is laid by the first screed of the remixer.
A second screed in the remixer, immediately behind the first, lays an additional new mix 5 .
The method is useful especially for long homogene road sections. After laying the asphalt
surface course, chippings may be applied to increase the initial grip and are finally compacted
with rollers while hot. For more information; see [Wirtgen, 2003].
In some countries, such as Germany and Sweden, the remix method has become popular and
large areas have been laid in Sweden at a thickness of around 30 mm. One version reconditions the pavement at a temperature of only 100 oC, using an SBS modified emulsion 6 , with
good results. In "remix plus" the top layer is often around 15 mm. In all four hot in-place
recycling methods, the relaid or new layer(s) is/are compacted by rolling.
Effect of climate on the preference for thin asphalt layers
Dry versus wet climate
The preference of TAL in dry and wet climate, respectively, is unavoidably linked to the
surface texture of the chosen pavement with regard to what is acceptable considering all other
conditions being equal. In order to assure the road users safe driving condition TAL with an
open surface textures will be preferred under wet climate conditions as these types of
materials initially were commended for their ability to reduce splash and spray and maintain
wet skid resistance – which are important factors for overtaking lorries and braking during
rainy conditions. This becomes even more important when the road consists of several lanes,
irrespectively of whether the transversal profile is sufficient or not.
Part of the information in this section is from: http://www.kutter-dsk.de
See for example: http://www.peab.se/Miljo--framtid/Innovativ-teknik/Eco-Paving/
Page 32 of 124
When the local conditions and the climate change imply the possibility of extreme weather
occurring from time to time this consideration must be taken seriously. Possible candidates
range from open graded AC over SMA and BBTMs and UTLACs to PA. Driver safety is the
main concern and the wet conditions for the open textured road surfaces may have a
recognised drawback on the durability. Here it is important to focus on material types that
have thick bituminous mortar films on the aggregates. This highlights candidates like SMA
and UTLAC – the first often includes bituminous mortar stabilized by the addition of
cellulose fibres and the second which has its unique sealing capability in form of the thick
applied polymer modified tack coat which is ”boiled up” into the structure of the surface layer
creating a good and virtually impermeable bond.
In climates where precipitation hardly occurs the range of materials is larger. Under these
conditions dense graded materials can also be acceptable candidates. The dense grade
materials offer better side support to the larger aggregates in the mix which may in the long
run improve the durability of the pavement- This increased side support may also increase the
surface resistance against shearing forces.
But if the dry climate is combined with a hot climate (which is often the case) then some word
of caution are appropriate with regard to the dense graded ACs. The high service temperatures
in the surface layers accentuate rutting (permanent deformations) as a likely failure
mechanism. In order to avoid a potential rutting problem the asphalt producers may be
tempted to use a too lean mix design. This creates two problems for the durability of the
In spite of the dry conditions water or moisture can easily penetrate thin bituminous mortar
films and reach the aggregate which may be moisture susceptible. If de-icing salts are used at
winter times moisturized microclimatic conditions in the pores of the pavement can have an
effect even when you think the pavement is dry and no moisture action can occur.
The second problem with lean dense graded mixes in hot, dry climates is that the thin
bituminous mortar films may show accelerated hardening over time leading to premature
failure due to fretting (loss of mortar) and ravelling (loss of aggregates) because the binder
becomes brittle.
Hot versus cold climate
In the asphalt concrete mix the most climate sensible component traditionally is the bitumen
part. The viscoelastic properties of the bitumen depend on the temperature – and on the traffic
load. This means that during hot climatic conditions the bitumen is like a viscous liquid and
permanent deformation or rutting of the asphalt mix can occur when the temperature is high.
In cold conditions the critical property of the bitumen in the mix is elastic stiffness. The
bitumen can become very brittle, resulting in cracks during the loading by the traffic when the
temperature is low.
Page 33 of 124
Traditionally during the pavement design and the design of the asphalt mix a proper selection
of the type of bitumen can minimize the climatic influences of very high, very low temperature or a combination of both. If necessary an alternative is to choose polymer modified
bitumen. An example of a design method with focus on the climatic behaviour is the Superpave performance grade binder selection.
Ageing or oxidation of bitumen is the result of the reaction with oxygen in the air. The
reaction results in bitumen which becomes harder and more brittle whereby the pavement is
more exposed to cracking. Normally the rate of the ageing process is slow but hot climate
with high temperature speeds up the process.
In the asphalt pavement the most critical parameter for the ageing is the air void content. A
high air void content results in a higher rate of ageing. Other parameters affecting the ageing
are the bitumen film thickness and the aggregate grading (dense- or gap-graded).
Because the ageing of bitumen happens faster on the surface of the pavement, TAL is more
critical to ageing than thicker layers of wearing course asphalt. The open structure also gives
higher exposure to oxygen than denser structures, which may increase oxidation and thus
decrease the durability.
Since most of the ageing occurs during production, transportation, laying and compaction of
the asphalt mix, the ageing process can be minimized with optimum plans for the production
of asphalt and an optimum choice of bitumen type.
In the process of laying asphalt – depending on the grade of bitumen – the temperature of the
asphalt during the paving is 150 – 180 °C and during the compaction with selected types of
rollers the temperature is 130 – 170 °C. When the temperature is 90 – 100 °C the compaction
must be finalized because the viscosity of the bitumen at this temperature is so high that
further compaction work is useless.
The decrease of the temperature depends on the thickness of the asphalt layer. The thicker the
layer the slower it cools down and the longer the asphalt mixture can be compacted, the less
weather sensitive the laying and compaction process. That means that in cold climate the
laying of TAL is restricted, difficult or impossible. See further Section 6.6.3.
Aggregates in asphalt concrete are not sensitive to hot or cold climate in the same way as the
bitumen part. The most critical parameter in relation to the climate or weather is the ability to
resist degradation caused by cycles of freezing and thawing when the aggregates are in
saturated conditions.
Setting up proof specifications with requirements to
Los Angeles Abrasion
Crushed Content
minimises the climatic influence from aggregates in TAL.
Page 34 of 124
Thin asphalt layer performance in cold region – Example from
The Hokkaido Island in Northern Japan has long and cold winter seasons. The annual
snowfall in Sapporo is 5 m per year and the average temperature in January is -5 °C. Winter
maintenance of roads is carried out by spreading wet salt and by the use of snow ploughs
equipped with steel. Rotary snow removers are also used. The use of studded tyres has been
prohibited from 1990 and nowadays only snow chains are used [Bredahl Nielsen et al, 2005].
It has been a challenge in the Hokkaido area to develop pavement types effectively reducing
splash and spray, having high skid resistance and at the same time providing a high structural
resistance to the wear of snow ploughs and snow chains. Ordinary SMA provides excellent
durability but is limited in its technical performance, while porous asphalt has excellent
technical performance but poor durability; and especially so in cold climate. For this purpose,
a new class of very open SMA pavements has been developed which intends to combine the
best properties of the two types of pavement and which will work well in cold climate.
These pavements are constructed as a "hybrid" of stone mastic asphalt (SMA) and porous
asphalt pavements (see Figure 6.5 and 6.6). The mix is dense with a very open surface texture
in the upper 10-13 mm, equivalent to the maximum aggregate size; something which is rather
typical of many TAL. The binder is standard SBS modified binder. The pavement is laid by
an ordinary asphalt paver in one pass and the mix requires severe quality control during
production as it is quite sensitive to the amount of mortar. Bleeding during compaction is
often observed. These pavements are new to Japan, and to a cold climate in general, and the
practical experience therefore still were limited in 2005, but they had so far shown good
performance [Bredahl Nielsen et al, 2005].
Probably, the Japanese have not constructed these as TAL, i.e. with thickness less than 30
mm, since they generally use 13 mm maximum aggregate. But, if these very open SMA
pavements were constructed with a maximum aggregate size of 8 or 10 mm they could be laid
as thin layer pavements.
Sieve Size in mm
Figure 6.5: Material gradation curves for porous asphalt pavement, new "hybrid" pavement
and regular SMA pavement [Bredahl Nielsen et al, 2005].
Page 35 of 124
Figure 6.6: Porous asphalt (right) and the very open SMA pavement (left), both with
maximum 13 mm aggregate size [Bredahl Nielsen et al, 2005].
In fact, this type of pavement is nothing new to Europe. Compare with Figure 3.1 where it
appears that the BBTM class 1 and UTLAC pavement types have grading curves which very
much resemble those of the Japanese open SMA. The only difference is that the Japanese use
larger aggregate (up to 13 mm versus up to 8 or 10 mm). Yet, the Japanese experience is of
interest to this project since they have demonstrated a possible adaptation of such a pavement
type (as BBTM 1 and UTLAC) to a cold climate. That is, essentially they have confirmed that
a type of pavement with grading similar to that of BBTM or UTLAC has better durability in
cold climate than a porous asphalt pavement (which is not surprising). The big question-mark
is, however, how such a pavement would work under the action of studded tyres. It may be
worthwhile to try the UTLAC and BBTM class 1, taking the Japanese experience into
account, in a cold region in Europe, both with and without the use of studded tyres.
Interaction with studded tyres
VTI has developed an extensive model for the prediction of wear due to studded tyres
[Jacobson and Wågberg, 2004], which summarises the Swedish experience. The model is
based primarily on tests using VTI Circular Test Track (CTT) in which 28 samples can be
tested at the same time. These results have been correlated to ten field test sites where surface
wear has been followed up since the early 1990's. The laboratory tests include testing
materials comprising different gradations, stone size maximum as well as aggregate properties
such as petrology, density, shape, resistance to fragmentation and resistance to wear. The
parameters shown to have significant impact on surface wear resistance were (based on CTT
and field tests):
Page 36 of 124
Aggregate quality
 On a descending quality scale: Porphyry, quartzite, granite, gneiss
Proportion of coarse aggregate (> 4 mm)
 The larger, the better. Range 45-67 %
Maximum aggregate size
 The larger, the better. Range 8-16 mm
Mixture type (aggregate gradation: Dense or gap graded)
 Gap graded better than dense graded. Factor is partly a result of “proportion of
coarse aggregate”
Use of mixed aggregate qualities
 Wear resistance is given by proportional contributions from each aggregate
Binder type (modification)
 Dense asphalt concrete seems dependent on binder while gap graded is not
Degree of compaction and air void content
 Inadequate compaction is detrimental in many ways, which are of greater
 Range 94-96 % degree of compaction shows clear detrimental effects while
97 % nominal degree of compaction only shows indicative effects
 Indicative effects between high (worse) and low air voids. Range within
specifications (range varying for different mixes)
Wet or dry surface
 Wet surfaces are generally worn at a greater rate compared to dry
Type of stud and stud force
 Properties of studded tyres are very important and subject to legislation
 Greater vehicle speed results in greater wear. Range 60-110 km/h
Climate zone has earlier been shown to influence wear resistance, but the rationale behind it is
still unclear. The comments given to the factors influencing wear resistance given above can
of course not be interpreted as general conclusions but valid within the ranges of observations
reported. Furthermore, the factors influencing wear resistance are probably interacting.
To model the contribution of aggregate quality on wear resistance, aggregate resistance to
wear is assessed using the Nordic abrasion test, EN 1097-9. The Nordic abrasion test is a ball
drum test, similar to Micro deval, aimed at assessing fragmentation of coarse aggregate fractions. The test reports percentage fragmented material (A N ). For gap graded wearing courses,
the wear relative to a reference is reported based on A N and maximum aggregate size
(NMAS) as equal to 1,547 + A N *0,143 - NMAS*0,087. For dense graded mixes, the model is
different in the way that NMAS is exchanged with parameters considering the fraction of
large aggregate and Marshall air voids.
Page 37 of 124
To summarise, aggregate quality and proportion of large aggregate can to a large extent
explain and predict wear resistance of dense and gap graded asphalt concrete wearing courses.
However, a number of other parameters, such as climate conditions and material properties
will interfere.
Countries or regions with policy-driven use of TAL
As appears in Figures 5.2-5.4, there are some countries where TAL are used more systematically, driven a little harder than in other countries by certain policy-related issues.
In Sweden and Norway the main use is of TAL with NMAS of 11-16 mm on the main road
network, with 16 mm most common in Sweden and 11 mm most common in Norway,
although also to a lesser extent on local roads. The reason is above all, is in Sweden that these
TAL appear to have good resistance to wear of studded tyres and are relatively inexpensive to
repave, while in Norway it is mainly a matter of volume of material used, as studded tyre use
has decreased. As described elsewhere; moreover, it has become popular to produce relatively
thin layers by remixing, due to environmental concerns and economy.
Switzerland uses a lot of TAL on the main road network; presently around 50 % (see Section
A.3.3). This is entirely driven by a policy to reduce noise. The main surface for this purpose is
AC MR8 which is dense asphalt concrete with 8 mm NMAS and a thickness 20-30 mm, with
a grading giving it a much rougher macrotexture than normal AC8. MPD values are more like
those of an SMA8; in fact its gradation curve is very similar to that of Swedish SMA8.
In the Netherlands, TAL have become very popular on the municipal (mostly 50 km/h) and
provincial (mostly 80 km/h) road networks. Cities have turned their interest from DAC and
SMA to the thin layers as the low noise solution on low-speed roads. In order to encourage
the use of quieter road surfaces in urban areas, an economic bonus was given by the Dutch
authorities to communal and regional road administrations which applied low-noise surfaces.
For 2001 and the next three years there was available about € 50 million (in total) for the quiet
roads compensation scheme. More than 15 million € were paid in two years in such bonuses.
What the Dutch government did was to provide a compensation for the extra costs of low
noise pavements compared to the costs of a conventional asphalt pavement. This included
normal maintenance for 15 years and a new top layer after 7-8 years. The contribution was
fixed and depended on the kind of road surface: one or two porous layers, open- or semi-open
or any other noise-reducing elements. The prime target was to stimulate the local authorities
to gain experience. The second target was to stimulate research by the road builders and to put
together a lot of information about the behaviour of these kinds of roads.
The surfaces had to be tested according to the C road procedure (see Section C.3.2) and provide
a certain noise reduction in new condition in order to eligible for the bonus. Even if this bonus
system is now abandoned, it was considered to be a success and had indeed the effect of
opening the eyes of the street authorities for the advantages of using thin layers.
The DWW (now RWS) a few years ago issued advice for road authorities that they are now
allowed to use thin layers at up to 80 km/h when noise reduction is needed.
Page 38 of 124
In the U.K. hot rolled asphalt (HRA), usually with rather course stones rolled into the mastic,
was the dominating surfacing on trunk roads as well as on local roads and streets. However,
for economic reasons and because of noise issues, cities became interested in TAL in the
period 1995-2005 (approximately), as illustrated by the rapidly expanding HAPAS certification of TAL systems (see Section C.3.1). As shown in Section 7.8, the experience of the
first decade of use of TAL was largely positive.
Even London has started to use TAL extensively. It is interesting to note that the 2007
publication "London Asphalt Specification", which is intended to give the street authorities in
Greater London guidance regarding selection of surfacings, recommends an extensive use of
TAL. The first edition from 2007 was followed by a second edition in 2009 [Walsh, 2009].
The major table specifying wearing course for roads where traffic may cause noise disturbances is shown in Table 6.2. In the document it is pointed out explicitly "It is strongly
recommended that only TSCS 7 should be specified in preference to generic Stone Mastic
Asphalt [SMA]".
Table 6.2: Preferred materials in the "London Asphalt Specification", edition 2009. The
surface courses designated as ST6 and ST10 are TAL (6 and 10 are NMAS in mm), while
ST14 with thickness 35-50 mm is thicker than TAL considered in this project. S1H is an HRA
surface. Road Type 1 means >10msa 8 [<600cv/day] and Road Type 2 means 2.5 to 10msa.
TSCS = Thin Surfacing Course Systems
msa = Millions of equivalent standard axles (exerting or applying a force of 80kN and passing during the
pavement design life)
Page 39 of 124
Another interesting note is the following: "Fretting of the surface of roundabouts and sharp
bends can be the result of too large an aggregate size being used in the Thin Surfacing.
Normally a 10mm aggregate size material is recommended. Thin Surfacing laid in winter is
more prone to early fretting as the surface is more likely to have been insufficiently
compacted. The greater the texture depth, the more prone to fretting is the surface" [Walsh,
2009]. There are several more interesting recommendations pertaining to TAL in the
Use in ERA-NET ROAD countries compared to other countries
In order to reflect both the national and regional practices, the figures published each year by
EAPA (European Asphalt Pavement Association) are a very useful source [EAPA, 2007a].
Those figures give an overview, from a number of European countries, about the asphalt
production, asphalt application and the use of binders. The data have been established with
the assistance of the members of EAPA, and are at present the best available and reliable for
the asphalt industry.
With respect to surface courses, the EAPA overview of 2008 gives detailed information about
the production of each type of asphalt based on the European product standards (EN 13108
series). Data are expressed as a percentage of the total amount of surface courses.
Consequently, no precise data is available with respect to the nominal layer thickness of a
particular application (except for mixes defined by EN13108-2). The overview does not contain any data for mixes defined in either prEN13108-9 or the UTAG 16 guideline. In Table
6.2 an overview is given with respect to countries both within and outside the ERA-NET
ROAD framework.
It should be noted that the policies within Europe with respect to the use of TAL may differ
substantially. As already mentioned TAL in Denmark represent about 95 % of all surface
courses. In Belgium, this figure is probably much lower and is dependent of the region
(Flemish or Walloon region). For example, at present the Flemish tender doesn’t specify TAL
as defined in EN13108-2 but the Walloon region does. Moreover, in Flanders SMA-C 9) (40
mm) is the standard choice for a top layer on a motorway. Another example is given by
porous asphalt. Porous asphalt (either in a single or twin layer application) is the standard
choice on motorways in the Netherlands, while in Belgium any use of porous asphalt is very
rare. The main reason is problems encountered during winter maintenance although climate
conditions can be considered similar to those in the Netherlands.
It appears in Table 6.2 that there is not much resemblance to Figures 5.2 and 5.3, if any at all.
This is explained by the sometimes confusing definitions of TAL and by generally poor or
even missing statistics. However, at least one thing is worth noting in Table 6.2; namely that
Spain is using TAL extensively.
See footnote p. 8
Page 40 of 124
Table 6.2: Overview of production of TAL within Europe (% of the total production of hot
mix and warm mix asphalt), where the column 13108-2 would be within the TAL definition
used in this project. However, there should also be at least a TAL-relevant column for 131089 (UTLAC) but such data is missing. Source: data published in 2008, see the main text.
Outside FR
Page 41 of 124
In order to use a thin layer, the maximum aggregate size (NMAS) must not be large. Small
NMAS means low macrotexture. This might cause problems with wet skid resistance and
hydroplaning. In order to counteract this, TAL are often designed with an open texture (gapgrading) which gives high macrotexture even with small NMAS. Such macrotexture is
favourable to noise emission and rolling resistance, since it is related to low megatexture.
Both megatexture and macrotexture are critical for the environmental properties noise and
rolling resistance; the latter being proportional to fuel consumption and thus CO 2 emission.
Most of the proprietary surfaces listed in Table 6.1 have been designed with substantial efforts
to optimize megatexture and macrotexture; some using advanced methods and materials to
achieve such properties.
Skid resistance
International survey
The Danish Road Institute conducted a literature study in 2005 on thin layers where the conclusions regarding traffic safety were the following [Bendtsen et al, 2005]:
Generally, measurements show that thin noise reducing layers have a high skid
No special information about the performance of thin layers during winter conditions
was found.
French experience
The Danish Road Institute (DRI) and DVS in the Netherlands in 2005 conducted study tours
to France in order to collect information on noise reducing thin layers. The main findings on
traffic safety issues were [Bendtsen & Raaberg, 2005]:
The thin layer pavements have an excellent skid resistance
The skid resistance is better for TAL than for ordinary dense asphalt concrete. This is
due to the surface texture of the thin layers being rougher
Skid resistance increases with decreasing maximum aggregate size. An explanation
could be that the number of contact points between the pavement and the tyre is larger
with smaller aggregate. TAL with 6 mm aggregate show better skid resistance than
pavements with 10 mm aggregate
The polished stone value of the TAL is good due to high quality aggregate
Skid resistance can be improved on newly laid TAL by applying sand during
Reduction of splash and spray is better when using 10 mm aggregate than when
applying 6 mm aggregate.
Page 42 of 124
Danish measurements
Thin noise reducing pavements have an open surface structure in order to reduce noise generated from air pumping and at the same time having a smooth even surface in order to reduce
the vibrations generated in the tyre which also induces tyre/road noise. It is generally not
considered that there is a correlation between skid resistance and tyre-road noise. Skid
resistance is related to the pavement micro- and macro texture where as noise is mainly
related to macro texture.
The skid resistance of a series of Danish test sections with TAL has been measured by the
ROAR device operated by DRI. The measurements are carried out using 20 % slip and prewetting the pavement surface with a 0.5 mm water film, meaning that the measurements are
performed under wet conditions. The results from measurements on a series of thin layers
with different aggregate size can be seen in Figure 7.1 where there is a quite clear tendency
that the skid resistance is increasing when the maximum aggregate size is decreasing. A
physical explanation for this could be that the sharp edges of the aggregate improve the skid
resistance and when the aggregate size is reduced the number of sharp edges pr meter road
surface is increased.
Skid resistance
Max aggregate size
Figure 7.1: Maximum aggregate size and the skid resistance measured on Danish test
sections with thin layers [Bendtsen & Raaberg, 2007].
On the basis of the available data and the analyses carried out the following tendencies can be
seen [Bendtsen & Raaberg, 2007]:
All pavements included in the test show skid resistance fulfilling the requirements set
by the Danish Road standards
Generally the noise reducing thin layers have high skid resistance when they are new
Page 43 of 124
There is a little reduction in skid resistance as pavements (dense reference pavements
as well as noise reducing thin layer pavements) get older
The skid resistance of the noise reducing thin layer pavements is marginally higher
than that of dense reference pavement in Denmark
No correlation has been found between macro texture (MPD) and skid resistance
The general tendency seems to be an increase in skid resistance if the maximum
aggregate size is reduced
The inclusion of oversized aggregates did not influence the skid resistance.
Figure 7.2 shows the results of skid resistance measurement 2004 and 2006 on noise reducing
TAL test sections and a reference section on the Danish motorway M10 at Herning [Bendtsen
& Raaberg, 2007]. There are not long enough time measurements available to give any
conclusion on the long time performance of skid resistance.
Measurements of the friction at M 10 test site
TP 8c
SMA 6+
AC 8o
August 2004
AC 11d
September 2006
Figure 7.2: Comparison of skid resistance measured in 2004 and 2006 at test sections with
noise-reducing thin layers on Highway M 10 in Denmark [Bendtsen & Raaberg, 2007].
Experience in the United Kingdom
Please refer to Section 7.8 and Figure 7.13.
Page 44 of 124
Experience in the Netherlands
Skid resistance measurements were performed on 30 thin asphalt layers similar to SMA
pavements. These were the main conclusions [Groenendijk, 2004]:
Most surfaces show a wet skid resistance (test at 50 km/h with 86 % slip) after
installation of more than 0.4. Some values are initially around 0.3 but after two
weeks of traffic the skid resistance already increases significantly. The end report
of IPG also confirms these findings [Bennis et al, 2008]
There is a clear decrease of skid resistance within the first week after opening for
traffic. After this week skid resistance increases again. This evolution continues
the first 6 months and is dependent on traffic density.
Eventually skid resistance decreases again. Nine thin noise reducing layers were
monitored by KOAC-NPC for CROW. The wet skid resistance of eight surfaces
decreased. One surface had a large increase probably due to ravelling [Bijma,
2006]. Ravelling may occur and may compensate this effect of decreasing skid
resistance which is why the intervention level at the end of lifetime may be
reached later.
Very locally, a lack of roughness may be caused by faults in production / transport
/ installation (demixing), visible as a permanent bituminous layer covering the
microtexture of the aggregate [Groenendijk et al, 2006].
Texture measurements show MPD values equal or a bit smaller than DAC. It is
advised to require a minimal MPD value.
Due to lack of skid resistance measurements for ZOAB similar thin layers, measurements of
double layer ZOAB are used as estimation [Groenendijk, 2004]:
Most surfaces show a wet skid resistance after installation of less than 0.38, which
is the requirement. Within 2 weeks this increases significantly.
Eventually skid resistance decreases again. Ravelling may occur and may
compensate this effect which is why the intervention level at the end of lifetime
may be reached later.
Influence on road traffic noise
Tyre/road noise is generated by four major processes [Sandberg & Ejsmont, 2002] with minor
contributions from other mechanisms:
1. Tyre vibration: Vibrations are generated by the contact between the pavement surface
and the tyre tread pattern blocks. Tyre vibration dominates tyre/road noise in the
frequency range 300 to 1500 Hz; the rougher the surface, the higher the noise levels. A
larger maximum aggregate size leads to higher tyre/road noise levels.
2. Displacement of air - The air pumping effect: When the tyre tread pattern rubber
blocks hit the road surface, air is pressed out of the cavities between the rubber blocks
and between the tread and the road surface. When the blocks leave the road surface air
is sucked back into the cavities. This air pumping to the surroundings generates noise
Page 45 of 124
at frequencies above 1000 Hz. If the road surface is open or porous the air pressure
gradients will be lower since the porosity "short-circuits" the pressure gradients, and
the noise is reduced.
3. The horn effect: The curved tread pattern of the tyres and the plane road surface act
as an acoustical horn which amplifies the noise generated around the contact point
between the tyre and the road surface. If the road surface is porous (and therefore
sound absorbing) there will be less amplification.
4. Propagation effects: The propulsion noise and tyre/road noise propagate from the
vehicle to the road surroundings. During this propagation noise is reflected from the
road surface. If the road surface is porous sound may be attenuated due to phase
effects on the reflected sound field.
5. The effect of stiffness: If the pavement is almost as soft as the tyre, tyre deflections
will be much smaller and less noise will be generated.
Noise optimization of thin layers
The basic concept of using open textured thin layer pavements for noise reduction is to create
a pavement structure, with as big cavities at the surface of the pavement as possible in order
to reduce the noise generated by air pumping, and at the same time ensuring a smooth surface
so the noise generated by tyre vibration will not be increased. Such a noise reducing open
textured pavement can be thin, as the mechanisms determining the noise generation only
depend on the surface structure of the pavement.
ure 7.3 illustrates two types of pavements with an open surface texture. The pavement texture
having a “positive profile” will give rise to higher noise levels from tyre vibration than the
pavement texture having a “negative profile”. Good compaction and cubic aggregate shape
generate a pavement surface with “negative” texture.
“Positive profile”
“Negative profile”
Figure 7.3: Sketch of pavements with “positive” and “negative” profile of the surface
texture [Bendtsen et al, 2008]
A qualitative model has been suggested for the surface texture influence on the generation
tyre/road noise, see Figure 7.4. X is the difference in height between the highest points of a
road surface profile. H is the average distance between the highest points of the road profile.
The Mean Profile Depth (MPD) is an indicator for the openness of a pavement surface.
To obtain as little tyre/road noise as possible:
1. Reduce X and H to secure a smooth surface to reduce tyre vibration
2. Increase MPD to reduce the noise generated by air pumping.
Page 46 of 124
Figure 7.4: Illustration of road surface texture parameters influencing tyre/road noise
generation [Fujikawa, 2004], [Bendtsen et al, 2008]
The Kastrupvej example
As a part of the EU project SILENCE the Danish Road Institute initiated a project to test
optimized noise-reducing thin SMA layers in the laboratory [Nielsen et al, 2006]. A full scale
experiment with seven test pavements has then been launched in 2007 in Copenhagen. The
aim was to establish and even surface with a “negative texture profile” with open cavities.
The noise optimization was done by using:
small maximum aggregate size NMAS (4, 6 and 8 mm) to reduce noise generated from
tyre vibration
high built-in air void content to reduce noise generated from air pumping. The SMA
pavements have air void contents of 8.8 to 15.3 % 10). These SMA pavements do not have
communicating pores through the whole thin layer, but they have an open surface texture
a small proportion of oversized aggregate (8 mm aggregate) added to the mix to obtain a
more open surface structure; referred to as “4+8” and “6+8”
as cubic aggregate as possible to reduce noise from tyre vibration. Triangular and round
aggregate is expected to result in a rougher surface giving rise to more noise from tyre
Table 7.1 and Figure 7.5 give key information on the mix design and Figure 7.6 shows one of
the surfaces.
Table 7.1: Mix characteristics for the thin layers on Kastrupvej [Thomsen et al, 2008] (a
minus sign "-" means that no data are available).
DAC 11
SMA 6+8-a
SMA 6+8-b
SMA 4+8
SMA 6+8 (Opt.)
Max. aggregate size [mm]
6 + 5/8
6+ 5/8
4+ 5/8
6+ 5/8
Binder [%]
Air void [%], geometric
This is much higher than normally would be considered an SMA, but it was an adaption made in this Danish
Page 47 of 124
Cumulative % passing
SMA 6+8 (Opt.)
SMA 4+8
SMA 6+8-a
DAC 11
4 5,6 8 11 16
Sieve (size of particles in mm)
Figure 7.5: Grading curves for the SMA pavements [Thomsen et al, 2008].
Figure 7.6: Optimized noise reducing SMA 4, Kastrupvej, when the pavement was three
months old. The black and white squares are 10 x 10 mm [Thomsen et al, 2008].
Also an optimized open-graded asphalt concrete (AC 6o = OGAC 6) was constructed on the
test road. Figure 7.7 shows the noise reduction for passenger cars relative to the dense graded
asphalt concrete reference DAC 11 when the pavements were a few months old. The OGAC 6
Page 48 of 124
yielded the best noise reduction for passenger cars of 4.3 dB followed by the SMA 6+8 (Opt.)
of 3.7 dB.
SMA 6+8-b
SMA 6+8-a
SMA 4+8
SMA 6+8 (Opt.)
Noise reduction [dB]
Figure 7.7: Roadside noise reduction for passenger cars relative to the reference DAC 11 of
the same age when the pavements were a few months old at reference speed 50 km/h and
reference temperature 20 °C [Thomsen et al, 2008].
French experience
Trends in the development of thin asphalt layers in France optimized for noise reduction are
presented in [Bendtsen & Raaberg, 2005]. Table 7.1 shows the average noise reduction
measured at thin layers with 6 mm nominal maximum aggregate size relative to dense asphalt
concrete with 10 mm aggregate, AC 10d. Results are given for two types of thin layers, with
larger built in air void content in type 2 than in type 1. Table 7.2 shows the average noise
reductions relative to dense asphalt concrete with 14 mm aggregate, AC 14d, which is
sometimes used as a reference pavement. The noise reductions were 3 to 4 dB for passenger
cars and 2 to 3 dB for trucks relative to AC 10d, and 1 dB higher than this relative to AC 14d.
Table 7.1: Average noise reduction at French thin layers with 6 mm maximum aggregate size
relative to AC 10d [Bendtsen & Raaberg, 2005].
Type of TAL
Passenger car (90 km/h)
Multi-axle trucks (80 km/h)
Type 1
2.7 dB
1.9 dB
Type 2
3.7 dB
3.1 dB
Page 49 of 124
Table 7.2: Average noise reduction at French thin layers with 6 mm maximum aggregate size
relative to AC 14d [Bendtsen & Raaberg, 2005].
Type of TAL
Passenger car (90 km/h)
Multi-axle trucks (80 km/h)
Type 1
4.2 dB
3.0 dB
Type 2
5.2 dB
4.2 dB
Influence on rolling resistance
In recent years VTI has studied the relation between rolling resistance coefficient (RRC) of
tyres on various road surfaces and its relation the Mean Profile Depth (MPD) of the surface
texture. An extensive measurement program was conducted in 2009 in Sweden and Denmark.
The results are shown in Figure 7.8 and 7.9, with all results plotted as RRC versus MPD.
Figure 7.10 shows the test equipment in action in Sweden.
RRC for average tyre
MPD [mm]
Figure 7.8: Rolling resistance coefficient (average of three tyres) normalized to 80 km/h and
16 °C, measured on road surfaces in Sweden and Denmark in 2009. The blue symbols are thin
layers, the pink symbols are thicker pavements. Measurements made with RR trailer of the
Technical University of Gdansk, as ordered by VTI and DRI.
Page 50 of 124
RRC for average tyre
y = 0,0016x + 0,0079
R2 = 0,7712
M PD [mm]
Figure 7.9: Rolling resistance coefficient (RRC) as a function of MPD. The RRC values are
averages for the 3 test tyres; test speed was 80 km/h. AR pavements are shown with symbols
filled with green colour; grey symbols are conventional asphalt pavements (AC and SMA).
Figure 7.8 shows the results of measurements on motorway M10 at Solrød and on local road
Kongelundsvej, both in Denmark, and on various highways near VTI in Linköping, Sweden.
Pavements of thin layer type are indicated as blue symbols and conventional pavements (> 30
mm thick) as pink symbols. Measurements were made with three test tyres of passenger car
type, mounted on the RR trailer of the Technical University of Gdansk in Poland.
Caution shall be observed in interpreting the results since one cannot exclude the possibility
that there is a small bias between the Swedish and Danish measurements, which were made
two months apart and at quite different temperatures. However, a temperature correction to
16 oC has been made, utilizing the correction factor suggested in ISO 28580.
The results in Figure 7.8, with the caution expressed above, suggest that the TAL give
somewhat lower RRC than the thicker pavements, given the same texture (MPD). This could
occur due to the possibly more favourable macrotexture of TAL with a more "negative" 11)
profile than the other pavements.
More results of measurements and more on relations between RR and MPD are presented in
[Sandberg, 2011].
Negative here refers to the profile peaks being directed downwards rather than upwards, which is an
advantage for RR but a disadvantage for wet skid resistance.
Page 51 of 124
Figure 7.10: The Rolling Resistance Trailer of the Technical University of Gdansk conducting measurements of RRC for one of the test tyres on one of the test sections near VTI.
Figure 7.9 shows the results of measurements on highways and motorways in Skåne in
southern Sweden, with the purpose to see any differences or similarities between asphalt
rubber (AR) and AC/SMA surfaces. The AR surfaces were thin layers (25-30 mm thick). AR
pavements are shown with symbols filled with green colour; grey symbols are conventional
asphalt pavements (AC and SMA > 30 mm thick). Measurements were made with three test
tyres of passenger car type, mounted on the RR trailer of the Technical University of Gdansk
in Poland.
The results in Figure 7.9, again with the caution expressed above, suggest that there is no
significant RRC difference between AR and AC/SMA surfaces. Further discussion of these
results and the tests can be found in [Sandberg, 2010].
It should be noted that the main advantage of TAL with regard to rolling resistance (RR)
would be that TAL are often made with a little lower texture (MPD) than most of the thicker
surfaces; mainly due to the use of smaller NMAS. It is clear that texture has a substantial
influence on RR.
Thickness and weight advantages
Pavements of TAL are often used on concrete bridges and steel bridges. For concrete and
steel bridges in the Nordic countries the TAL is a part of the waterproofing systems protecting
the bridge deck against water and de-icing salts [Wegan, 2001]. In these systems with
medium and light type of traffic the type of the wearing course is SMA, UTLAC and asphalt
Page 52 of 124
concrete in a thickness from 20 mm. On larger bridges the thickness of the surfacing is up to
50 mm.
On bigger roads and on motorways paved with TAL, due to the requirement of noise
reduction, the adjacent bridge deck is usually paved with the same TAL, retaining the
properties of noise reduction and visual appearance.
When waterproofing smaller and older bridges with reduced weight capacity, the use of TAL
becomes apparent due to the reduced stress on the bridge system. In the design of the 3690
meter long Messina bridge, which spans the Strait of Messina in Italy, with a main span of
3300 m length, one critical design criterion is a thin wearing course to reduce the weight of
the bridge deck [Tækker, 2009].
On these bridges an alternative to TAL is thin pavements with synthetic binders based on
epoxy, polyurethane or acrylate. The same principle applies for overpasses and elevated
Assessment of investment decisions in economic terms can be made from a society perspective, as well as from a road owner perspective. Holistic society perspectives can make use of
Socio-economic cost benefit analysis in which costs of society, road users and road owners are
taken into account. In this way, the society can ensure the most value for money to the tax
payers and society as a whole. However, decisions cannot be made solely on the socioeconomic costs due to budget limitations. Therefore, road owner costs are also of great
interest in the decision making. In both cases, it is preferred to analyse costs occurring during
all stages from planning until far ahead in the future. Life-Cycle Cost Analysis (LCC or
LCCA) is the common term used for Cost Benefit Analysis including all stages in a product
life cycle. LCC requires a life cycle period and a discount rate. In the case of pavement
wearing courses, the life cycle period can be either the period in which the road is maintained
at the present standard or the expected life of the pavement. The period needs to be at least as
long as that of the longest lasting wearing course. A discount rate is needed to consider the
positive effects of postponing expenditures and allow capital to generate benefits elsewhere.
Costs and benefits are usually calculated, discounted (to present values) and summed to a
base year, often referred to as net present value. For more detailed information on socioeconomic project assessment in Europe, the reader is referred to the HEATCO report [Bicket
et al., 2006].
LCC can be done for many purposes. In the case of thin overlays, the following examples are
Decide on strategy to maintain pavement at given standard
Decide on whether a functional or environmental improvement corresponds to expectations on value for money
Relate different options to each other regarding improvement of functional or environmental properties of a road (e.g. to meet noise or particle mitigation requirements)
Road owner costs are mainly related to planning, design, construction, maintenance and
operation activities. Thin overlays could be an option in the construction phase but certainly
Page 53 of 124
in the maintenance phase. Then, the cost of thin overlays and their expected impact on future
needs for maintenance is determining their contribution to net present value. The impact on
future needs for maintenance is often simplified to just stating the estimated life length of a
treatment. However, success in a pavement life cycle perspective is also dependent on maintaining structural capacity over time and preferably increases it if needed to meet increasing
traffic volumes. For the same reasons, selecting the proper maintenance option should also be
related to the historic evolution of performance and the corresponding deterioration
mechanisms. Table 7.3 refers to costs found in literature and public sources. The costs used in
LCC should be representative of the present values during the whole calculation period.
However, since our knowledge about future costs is limited, it is generally accepted to use
present costs if no reliable information is available about future costs.
Table 7.3: Examples of costs reported from public sources.
Thin overlay cost
Expected life
Reference cost
$140.000 /mile
Thin lift w spray paver
$85.000 /mile
Thin overlay
$75.000 – 86.000 /mile
Thin overlay
€4,3 /m2
Thin overlay, 25 mm
Equal to ref.
$210.000 /mile
Cold plane w 2” overlay
$120.000 /mile
Thick overlay
$127.000 - 148.000 /mile
Thick overlay
€6,8 /m2
Overlay, 40 mm
2-4 years
Expected life
6-8 years
Abadie C “Louisiana Bituminous Surface Preservation Program” Louisiana Asphalt Pavement Association Conference
2009. Baton Rouge. USA
Choi J-H, Bahia H “Life-Cycle Cost Analysis-Embedded Monte Carlo Approach for Modeling Pay Adjustment at State
Departments of Transportation” TRR 1900. 2004. pp. 86-93.
Gransberg D “Life-Cycle Cost Analysis of Surface Retexturing with Shotblasting as an Asphalt Pavement Preservation
Tool” TRR 2108. 2009. pp.46-52.
Karlsson R. LCC tool for investment decisions. Commission for Swedish Transport Administration.
The expected life and costs reported in the table above differ. Life expectancy and costs will
differ from object to object depending on the conditions present. For a more detailed
discussion on life expectancy or maintenance intervals, the reader is referred to later sections
in this report. Regarding costs, one can conclude that many mechanisms behind costs of
overlays are related to the weight of materials used; e.g. the cost of bitumen and transport of
materials. Cost of aggregate probably varies both between regions and within regions for a
large number of reasons such as availability, quality, volumes, market, etc.
Overlay properties may influence road users in a number of ways such as:
 Friction may influence accidents
 Macro texture may influence friction, fuel consumption, tyre wear and noise levels
 Longitudinal evenness may influence fuel consumption, vehicle condition and goods
 Rutting may influence friction by aqua planning
Page 54 of 124
Furthermore, speed adaption and comfort is related to all these road surface properties. The
consequences of road surface properties inflicted on road users may in turn be priced,
transformed into road user costs, to allow cost benefit analysis of investment alternatives.
Road user costs may also be inflicted by the treatment itself, for example delay time costs due
to queuing during maintenance. Pricing of time, accidents, noise etc. differ between countries.
The reader is referred to the HEATCO report [Bicket et al, 2006] for a review of pricing
principles and levels in Europe.
The special properties of asphalt rubber thin layers
In a broad context, the following provides a summary of the derived benefits of using an
Asphalt rubber (AR) as a pavement preservation strategy, referring to common US motorway
Successful field performance history (over 40 years)
Less reflective cracking
Reduced maintenance
More durable
Less ravelling
Good rut resistance
Good skid resistance
Smooth ride
Good in hot & cold climates
Allowance for higher binder contents
Less splash & spray: better drainage
Less tyre/road noise
Cost effective
Beneficial engineering use for old tyres
Reduced induced stresses in PCC pavements due to thermal gradients
Reduced climate change impacts
Environment-friendly and energy efficient
Transferred to European conditions, most of the benefits listed above would be comparable to
those of European SMA pavements, and the benefits of AR in comparison would be limited,
as far as we know presently. Nevertheless, it is hoped that ongoing experiments will show that
some of the benefits mentioned above will count as positive also in comparison to SMA. If
this will be the case the future for AR in Europe is bright.
These sections document a few special properties of AR mixtures that earmark and establish
the potential use of AR as a thin pavement layer. Evaluation of AR based on life cycle cost,
energy efficiency, and structural mix designs are detailed with notable examples from various
research studies in the United States.
Cost Considerations
Gap-graded AR (generally referred to as ARAC), generally 50 – 75 mm thick when laid in
USA and 25-30 mm when laid in Sweden, is primarily placed to address cracking on cracked
pavement sections. An ARFC (the open-graded variant) may be placed as an overlay depending upon the traffic volume and type of the highway. As mentioned previously (Section
Page 55 of 124
6.5.4), ARFC as applied in the US is one of the thinnest asphalt pavements ever applied as it
may be only 12 mm thick. The finished AR product is generally 25 to 50 % more expensive,
but is actually more cost effective when considered in a life cycle cost analysis [Kaloush et al,
2008]. Figure 7.11 (a) shows maintenance costs in dollars per lane-kilometre for a
conventional overlay and AR-ACFC (ARFC or AR open graded mix) over time in years. As
observed, AR requires less maintenance costs, and a corresponding low percent cracking as
shown in Figure 7.11 (b), and therefore has a longer service life. Figure 7.12 shows two other
comparative accounts between conventional and AR overlays with regard to the rutting
pavement distress and smoothness properties. As observed, AR pavements provide better
resistance to rutting during the design life: furthermore, smoothness is lower than the
conventional pavements over the years to provide better skid resistance and ride.
Overlays / Inlays
% C racking
Maintenance Cost $/lane-Km
Percent Cracking
9 10 11 12 13 14 15
Overlays /
0 1 2 3 4 5 6 7 8 9 101112131415
Figure 7.11: Conventional Overlays (AC or SMA) versus Asphalt Rubber, referring to US
conditions: a) Maintenance Costs; b) Percent Cracking [Kaloush et al, 2008].
mm / Km
Rut Depth, mm
Asphalt Rubber
9 10 11 12 13 14 15
Figure 7.12: Conventional Overlays versus Asphalt Rubber: a) Rut Depth in mm; b)
Smoothness [Sousa, 2006].
Page 56 of 124
Energy Savings Considerations
Various research studies conducted by Sousa documents that the energy savings by using AR
pavements (both gap graded and open graded designs) is quite impressive [Sousa et al, 2006].
Table 7.3 represents the heat of combustion values for crumb rubber modifier used in AR.
The table has been reproduced from another published article by Kaloush et al, 2008 [Kaloush
et al, 2008]. The 310267 kJ/Kg represents energy savings in terms of less AC overlays needed
over the life of the pavement. It is also the result of using less than ½ of the thickness of the
conventional HMA with equal or better field performance. The 310267 kJ/kg energy savings
refers to a 50 mm ARAC (AR gap graded) overlay used as an alternative to a conventional
100 mm asphalt pavement overlay. This has been a common practice of Arizona DOT for
over 135 pavement preservation projects built to date. Furthermore, 566109 kJ/kg energy
savings refers to a 25 mm ARFC (AR open graded) mix used on the top of a PCC pavement
in comparison to a 125 mm conventional HMA overlay. The 107860 kJ/kg energy savings
refers to the mining and transport energy components associated with using thinner AR
pavements compared to the thicker conventional HMA pavements.
Table 7.3: Energy Utilization (kJ/kg) for Asphalt Rubber [Sousa et al, 2006; Kaloush et al,
Energy Gain / Loss (kJ/kg)
Tyre Shedding
Shred Transportation
CRM Transportation
Steel Recovery
Asphalt Saved
Aggregate Saved
Total Gain / Loss
to 46516
to 566109
Reduced Thickness Design Considerations
Owing to the successful field performance of AR sections in the various states of the USA,
California Department of Transportation (CALTRANS) developed a reduced thickness design
guide for AR mixes that can provide the same service life as any thicker conventional HMA
pavements [Kirk and Holleran, 2000]. The historical studies and mix designs indicated that
the AR gap graded mixes needed a significant increase in binder content over dense graded
mixes. However, the AR gap graded mixtures had the least percentage of reflective cracks,
one third less than a 100 mm conventional overlay and less than one half a 200 mm overlay.
Hence, AR gap graded mix was chosen as the standard in the reduced thickness design guide.
Owing to its higher binder content in comparison with a conventional dense graded mix, an
AR mix would provide higher confidence in the design, apart from being more conservative.
The various designs have shown that AR mixes have proven to be very cost-effective, and the
multi-layer systems using AR binders have proven to be the most cost-effective. Also, high
binder content ARFC mixes used in combination have provided superior field performance.
Page 57 of 124
Life Cycle Cost Considerations
Life cycle cost analysis (LCCA) is recognized by public agencies as an effective tool to assist
in the selection of construction, rehabilitation, and maintenance treatments. Hicks and Epps
utilised life cycle cost process to compare results to evaluate LCC for pavement containing
conventional binders with similar applications containing AR binders [Hicks and Epps, 2000].
It was found that AR is cost effective in many of the applications used by the state highway
agencies of Arizona, California, and Texas in the United States. The results of the study
indicated that the use of asphalt rubber products is a cost effective solution in using AR as a
mix and/or chip seals applications. In addition, AR allows a thickness reduction, eventually
increasing the cost effectiveness of AR applications. It was recommended that AR is most
cost effective when reflection cracking is expected and it was emphasized that AR binders
will not be cost effective unless the thickness of the layer is reduced or extended life is
achieved. Table 7.4 presents total mix cost for different pavement scenarios and cost per
square meter per 25 mm normalized thickness for each pavement type [Carlson, 2009]. As
observed, cost per square meter per 25 mm thick of each pavement type for an ARFC open
graded mixture is about 9 % lower than a conventional HMA pavement while ARAC gap
graded mixture is about 16 % more expensive than the conventional HMA pavement. It must
be noted that a typical ARFC mix is placed as a 25 mm lift, a pavement which is a reduced
thickness one. This accounts for a huge savings in the cost of the AR pavement in comparison
to the conventional HMA pavement.
Table 7.4: Total Mix Cost in US Dollars of Different Pavement Types [Carlson, 2009].
Bid Price per Ton
Cost per Square
Meter/25 mm
HMA Dense
$94 (+16%)
$5.42 (+16%)
$82 (+9%)
$4.05 (-9%)
Main conclusions from Swedish seminar in 2010
In September 2010 an international invitational seminar was held in Sweden [Gummiasfalt,
2010]. Experience of asphalt rubber in North America and Europe was shared, with a focus on
an ongoing Swedish research program. The conference Chairman made the following conclusions at the end of the seminar (copied from [Andersson, 2010]):
In Sweden 45 % of recycled tyres goes to ”other than burning”
In Arizona 80 % of recycled tyres goes into Asphalt rubber pavements
No future problems in getting access of rubber granulate in Sweden
The working conditions were not significantly worse regarding PAH, Anilin, Naftalen
than conventional…
Benzotiatol, emanating from the vulcanization process, is the probable cause for
observed irritation and ”illness”
Asphalt rubber seems to generate lower concentration of PM10 than conventional…
Page 58 of 124
No substantial increase regarding dangerous leakage was observed for the asphalt
rubber compared to the conventional…
Better performance for wear resistance and crack propagation
Good friction results, no problem
Marginal effect regarding lower noise, 0-1 dB(A)
Were overall satisfied with the performed project
• From a technical and environmental point of view
• From a productivity point of view (”no big differences from other…)
Focused efforts in development of alternative PMB.
It should be added that long-term performance must be studied more and applications made
also in other European countries until asphalt rubber surfaces can be recommended for wide
applications in Europe. C/B ratio is so far uncertain, but present situation seems promising.
Overall assessment of TAL performance in the U.K.
During the 1990s, various categories of thin surfacing were introduced into the UK. They
rather quickly gained a substantial share of the surface course market in all parts of the UK
network. When the first of the thin asphalt surfacings laid in the UK were approaching their
expected service lives around 2005 it was decided to review the evidence on performance and
update the earlier assumptions about their service life [Nicholls et al, 2006].
The data base collected included maximum aggregate sizes of 6, 10 and 14 mm on a total of
128 sites. The thickness is not given, and one can wonder how thin a layer with NMAS of 14
mm may be, but at least some 14 mm surfacings were of the "ultra-thin paver-laid surface
dressing" developed in France. All five categories of thin surfacings listed in Section 3.2.1
were represented. Parameters monitored were skid resistance, texture and visual condition.
Data had been gathered on the various categories of thin surfacing system after periods in
service of up to 12 years for paver-laid surface dressing (PLSD), 13 years for thin asphalt
concrete (TAC), 10 years for thin stone mastic asphalt (TSMA), 10 years for multiple surface
dressing (MSD) and 6 years for micro-surfacing (MS).
Nicholls and his co-authors made the following conclusions, slightly edited by this Editor
[Nicholls et al, 2006]:
The results collected from the monitoring showed that there is a minimal risk of either skid
resistance or texture depth reducing significantly. See Figure 7.13. The main failure mode for
all the categories appeared to be fretting 12 , which tended to occur towards the end of the
serviceability life. From a visual assessment of the performance of each of the systems, the
following conclusions on the typical behaviour were drawn:
PLSD systems can be expected to give a 10 year service life.
TAC systems can be expected to give a service life of more than 13 years.
TSMA systems can be expected to give a service life of more than 10 years.
MSD systems can be expected to give a 7 to 8 year service life.
There were insufficient data on MS systems to estimate the typical service life.
Fretting and ravelling are essentially the same property, namely loss of aggregate
Page 59 of 124
There was no clear difference in the durability for any category of thin surfacing systems
whether the aggregate size is 0/10 mm or 0/14 mm. There were insufficient data on 0/6 mm
aggregate size systems to determine whether a change to that size will have an influence on
the durability.
The findings from the monitoring were that thin surfacing systems can routinely be constructed successfully to provide a safe surfacing with adequate skid-resistance, texture and visual
condition and that these properties are maintained in service. Therefore, the evidence supports
the acceptance by the Highways Agency of these systems for use on trunk roads in England
[Nicholls et al, 2006].
Figure 7.13: Mean Summer SCRIM Coefficient (MSSC) measured on a range of UK thin
surfacings plotted as a function of surface age. From [Nicholls et al, 2006].
Other performance properties
Other aspects of performance are treated in some detail in Annex B. See also the summary in
the next chapter.
Page 60 of 124
Table 8.1 summarizes the major advantages and disadvantages of TAL compared with DAC
11, based on the responses from the questionnaires. Compared to SMA 16, a reference surface
commonly used in Nordic countries, the table would look almost the same. Some advantages
of TAL would even be more pronounced, like the noise reduction and the cost reduction. On
the other hand TAL does not have a better rut resistance than SMA 16.
Table 8.1: Major advantages and disadvantages of TAL compared with DAC.
Major advantages
Major disadvantages
Noise reduction may be from marginal to
very high depending on TAL type
Ravelling susceptibility
Skid resistance is generally good at low and
medium speeds
Delamination susceptibility
Cost reduction in general, due to thinner
layer and fast paving (exceptions exist)
Frost susceptibility
Most features improve sustainability
Susceptibility to cracking related to substrate
Some types have potential for lower rolling
Some specially optimized proprietary TAL
may be expensive
Reduced working space needed
Weather conditions during laying are critical
Good rut resistance
Manual placement not possible
Speed of laying is high
Intrinsic shorter lifetime
Aggregate quality must be high; may be
difficult to find quality aggregates
Tack coating needed between TAL and
High-speed skid resistance and/or
hydroplaning may be a problem
Noise reduction decreases over time
Table 8.2 compares the performance of TAL with those of some other common road surface
types: DAC 11, SMA 11, single layer porous asphalt 0/8 (SLPA 0/8) and two-layer porous
asphalt (TLPA 8+16). Reference is the DAC 11 surface. “+” and “++” mean a better
respectively a much better performance for the given criterion; “0” means a similar
performance as the reference surface and “-“ and “--“ indicate a worse, respectively much
worse behaviour. This table is based on the answers received in the questionnaires, literature
and “expert judgement” by the authors.
Page 61 of 124
Table 8.2: Comparison of five common road surface types, including TAL. All evaluations
are made in comparison to DAC 11. The colours mean: Green = Better than DAC and SMA
11, Red = Worse than DAC and SMA 11, Yellow = about equal to DAC and SMA 11.
Skid resistance
Working space
Rut resistance
Speed of laying
Frost susceptibility
Weather conditions during laying
Manual placement
Noise reduction decreases over time
Splash and spray
Page 62 of 124
This important subject is treated in Annex C.
Mix design and optimization of TAL would normally be an important section in a State-ofthe-Art report. However, since this subject will be the main topic of a report in Stage 2 of this
project, this section has been cut out from the State-of-the-Art report and was instead
integrated into the Final Report for this project (see the Preface).
Page 63 of 124
The use of thin asphalt layers (TAL) in Europe seems to increase although available statistics
and inconsistent definitions make it difficult to distinguish between TAL and other hot-mix
In the "perpetual pavement" concept the philosophy is that the pavement base has eternal
bearing capacity and is paved with a thin long-lasting "skin" of surface layer which eventually
– due to water, ageing and other climatic action – must be renewed from time to time.
TAL provide a “skin” with favourable functionalities, such as noise reduction potential, relatively low rolling resistance, relatively good anti-spray properties and efficient light reflection.
This has accelerated the use of general product categories and proprietary products addressing
these demands, also implying relatively high sustainability and low construction and
maintenance costs. The fast laying of TAL implies shorter closure to traffic and this favours
the use of TAL. Provided the pavement base is of appropriate quality TAL offer solutions to
many of the functionalities mentioned and this is probably why there is immense interest in
such products.
Despite this, one shall not forget the problems and limitations associated with TAL. For
example, bearing capacity is often only marginal, and resistance to wear from studded tyres is
poor, unless large maximum aggregate is applied, in which case the thickness needs to be
relatively high. Open-textured or even porous TAL may offer good noise properties, but at the
expense of limited durability under heavy traffic load; for example in sharp curves or at large
gradients. Pavement porosities also tend to quickly get clogged by dirt. Another problem
worth mentioning is that TAL cannot be dismantled by milling with the techniques at hand
without downgrading the material.
The project group sent out a questionnaire to key experts. Unfortunately, the response was
rather limited and a round of interviews was not very successful either. Most respondents
mentioned noise reduction as a primary motivation to apply TAL. Also cost reduction and fast
paving seem to be important, like good resistance of TAL to skidding and rutting. A few
respondents mentioned durability problems as a disadvantage.
Policies on applying TAL vary substantially from country to country. For example TAL represents 95 % or so of all new Danish hot mix surface courses, while in Belgium this percentage is much lower and differs between regions. Also in Sweden, there is a substantial difference in the use of TAL between regions; not necessarily correlated with climatic conditions.
Countries with extensive usage of TAL include the UK, Switzerland, Sweden, Norway and
the Netherlands. Also Denmark and Austria use TAL relatively extensively. In Sweden, to
some extent also in Norway and Switzerland, TAL is used on the highway network, while in
the Netherlands and the UK usage is limited to municipal roads or city streets as well as
provincial or trunk roads.
The report gives general advice and a few examples of published life cycle costs (LCC)
compared with the cost of thicker overlays. This topic was further studied in another phase of
the project. The LCC of TAL cannot be assessed with any accuracy until TAL lifetime and
performance over time has been documented. Until then we must rely on calculation based on
engineering judgement concerning the TAL lifetime.
Page 64 of 124
Major advantages of applying TAL compared with standard DAC 11 or SMA 16 are
Noise reduction
Higher skid resistance (at least at low and medium speeds)
Lower cost
More sustainable
Less required working space
Better rut resistance
Faster laying
Major disadvantages of applying TAL compared with standard DAC 11 or SMA 16 are:
Susceptible to ravelling
Susceptible to delamination
Susceptible to frost damage
Susceptible to cracking related to substrate deficiencies
Weather conditions while laying TAL are more critical
Manual laying is not possible
Shorter lifetime
Noise reduction decreases rather rapidly with time
Some variants may offer rather low skid resistance in wet weather
Dismantling by milling with present techniques downgrades the material
The advantages and disadvantages listed above are not consistently valid across all types of
TAL and surface conditions, but are fairly representative of TAL is a family of wearing
courses. However, it is recognized that experience about long-term performance of TAL is
still insufficient; and consequently one needs to continue studying time series in the future.
The sensitivity of TAL to the weather conditions during the laying has been mentioned as a
major disadvantage. Road administrations and contractors are often forced - by numerous
factors - to build TAL during cold weather and then their durability may be low. Perhaps this
can be counteracted by optimizing the laying process; thus this is recommended for study.
TAL must be CE-marked in order to be marketed as complying with an EN 13108-series
product standard. These standards specify asphalt mixes; not their final application on the
road. The ETAG 16 guideline on ultra thin layers, however, intends to deal with the entire
process, including paving operations and the final application. Products complying with this
guideline will probably be an additional route for CE-marking in the future. The impact of
this CE-marking on the market still has to be seen in the daily practise of procuring asphalt
At present, classification of pavement acoustic characteristics is limited to declaring product
properties in Denmark, the Netherlands and the UK. CEN work on this is at an initial stage.
No system exists for checking pavement product conformity of production concerning its
noise characteristics.
At least two countries represented in the PEB are highly interested in the effect on TAL of the
exposure to traffic with vehicles using studded tyres. The present review concludes that
Page 65 of 124
aggregate quality and the proportion of large aggregate are the main parameters determining
wear resistance of dense and gap-graded asphalt concrete wearing courses.
TAL as defined in this project with layer thickness 10 – 30 mm normally have 12 mm NMAS
or smaller. When winter conditions call for extensive use of studded tyres and snow chains,
TAL may not be an optimum surface layer, as the rule in such cases says: “The larger the
aggregate the better". Therefore, in Sweden and Norway a special type of TAL, designated
TSK, with NMAS of 11 or 16 mm is widely used.
In the future it is necessary to look closer at the need for transportation of materials for
construction and maintenance projects, in order to reduce energy consumption and the
emissions following this. TAL may have both positive and negative influences on this,
depending on the mass of material needed and the availability of this material near the
construction or maintenance project. The need for high-quality aggregate in TAL is one issue,
while the lower mass of material is another one.
The report also discusses the use of asphalt rubber (AR) pavements as thin layers. A multitude
of benefits of using AR were listed, including less reflective cracking, less raveling, good rut
resistance, good skid resistance and smooth ride, better drainage facilities, reduced tyre/road
noise, cost effectiveness, beneficial engineering use for old tyres, and higher energy
efficiency. A combination of lower aggregate size in the AR, inclusions of crumb rubber in
the mix, consequentially incrementing an increase in the percentage of AR binder content, has
made it possible for AR mixtures to be used as thin asphalt layers that have provided substantial benefits such as reduced thickness design, higher durability, and lower maintenance
during the course of AR design life cycle.
However, it must be noted that these are the merits of AR typical for the conditions in the
USA. In Europe so far, there has been a different scenario when one takes into account the
derived benefits of AR, as observed in relation to a few similar pavement strategies of
comparable quality. Nevertheless, ongoing research and practical applications, the results of
which so far are rather positive, will determine whether the AR concept could be a success in
Europe as well.
The report indicates that the actual achievement of both excellent functional properties and
good durability (lifetime) is nothing which comes easily. In practice, it is often difficult to
realise both these requirements simultaneously since they are frequently in conflict with each
other. The information made available through this report should, therefore, serve as a basic
guideline for achieving the best compromise between the goals.
Page 66 of 124
It is gratefully acknowledged that Annex B.2.2 (about particulate matter) was written by Mr
Mats Gustafsson, VTI. As this was his only contribution to this report, he is not listed as a coauthor. Nevertheless, his contribution is valuable.
Mr Goubert would like to thank the following individuals for their willingness to filling out a
questionnaire and sharing their knowledge with the project team:
Ian Walsh, Jacobs Engineering (UK) Ltd, Maidstone (United Kingdom)
Jacob Groenendijk, KOAC.NPC, Apeldoorn (the Netherlands)
Peter J. Andersen, Vejdirektoratet, Copenhagen (Denmark)
Jostein Aksnes, Statens vegvesen Vegdirektorat, Oslo (Norway)
Olivier Ripke, Bundesanstallt für Straßenwesen, Bergisch-Gladbach (Germany)
Erik Van den kerkhof, Colas Belgium, Brussels (Belgium)
Cliff Nicholls, TRL Ltd, Wokingham, Berkshire (United Kingdom)
Berwich Sluer, BAM Wegen nv, Bunnik (the Netherlands)
Kenneth Olsson, Skanska, Farsta (Sweden)
Further, Mr. Goubert and Mr. Olesen would like to thank the following individuals for
agreeing to be interviewed:
Wim van Keulen (van Keulen advies), Vlijmen, the Netherlands
Jostein Aksnes (Norwegian Public Road Administration - NPRA, Road Directorate),
Geir Refsdal, Rolf Johansen and Jan Lindahl (NPRA, Eastern Region), Olle R. Larsen
(Kolo Veidekke), Norway
Alain Jacot (Société d’Analyses & Contrôles Routiers, SACR), Zürich, Switzerland
Johann Litzka (Austrian Association for Research on Road - Rail – Transport, FSV),
Jürgen Haberl (PMS-Consult, Engineering Office for Traffic and Infrastructure), Peter
Riederer (BPS, Oberösterreichische Boden- und Baustoffprüfstelle GmbH), Michael
Kostjak (Swietelsky BaugesmbH), Ronald Blab (Vienna University of Technology),
Gaetano Licitra (ARPAT, Agenzia Regionale per la Protezione Ambientale della
Toscana, and University of Pisa), Pisa, Italy
Bjarne Bo Jensen (NCC Roads A/S), Lars Ladehoff (Colas Danmark A/S), Uffe
Mortensen (Pankas A/S), Niels Christoffersen and Uffe Mortensen (Inreco A/S),
Page 67 of 124
AAB (2006): “Hot mix asphalt. General specification” (in Danish), Danish Road Directorate
November 2006.
Abbott, P. (2008): Personal communication between Mr P. Abbott, TRL Ltd in the UK and
Mr J. Kragh, Danish Road Institute/Road Directorate, June 2008.
Aksnes, Jostein; Hansen, Odd Durban; Haukland, Frode; Storeheier, Svein Å. (2010):
"CPX-målinger på forsøksdekker 2010". SINTEF IKT Notat # 90E354.05/TB, Trondheim,
Andersson, Per (2010): "Asphalt Rubber Seminar in Sweden september 29-30 – Closing
remarks". See further [Gummiasfalt, 2010] below.
Bahia, H.U.; Faheem, A. (2007): “USA Experience with OGFC and other thin layers”,
presentation during the Workshop on “Experience with very thin noise reducing layers”,
Technical University of Delft, the Netherlands, June 6th 2007.
BBA (2008): "Guidelines document for the assessment and certification of thin surfacing
systems for highways", British Board of Agrement (BBA), 2008. See (www.bbacerts.co.uk).
Bendtsen, H.; Raaberg, J. (2005): French Experiences on Noise Reducing Thin Layers.
Technical note 28, 2005, Danish Road Institute, Road Directorate (www.roadinstitute.dk).
Bendtsen, H.; Raaberg, R.; Thomsen, S. (2005): International Experiences with Thin Layer
Pavements Milestone – M2. Technical note 29, 2005, Danish Road Institute, Road Directorate
Bendtsen, H.; Raaberg, J. (2007): Safety and Noise Reduction, Thin Layers. Technical note
52, 2007, Danish Road Institute, Road Directorate (www.roadinstitute.dk).
Bendtsen, H. et al. (2008): DRI-DWW Thin Layer Project - Final Report. Danish Road
Directorate/Road Institute, DRI report 159, 2008 (www.roadinstitute.dk).
Bendtsen, H.; Ellebjerg, L. (2009): "Two-layer porous asphalt – lifecycle. The Øster Søgade
experiment". Report 165, 2009, Danish Road Institute, Road Directorate. Website:
Bendtsen, H.; Lu, Q.; Kohler, E. (2009): "Acoustic aging of asphalt pavements - A
Californian / Danish comparison". Report 171, 2009, Danish Road Institute, Road Directorate
Bennis, T.; Hofman, R. (2008): "Eindrapportage IPG wegdekken – Monitoring." Document
DVS-2008-046, 2008 of the Dutch “Noise Innovation Programme”, Rijkswaterstaat (RWS),
Dienst Verkeer en Scheepvaart (DVS), P.O. box 5044, 2600 GA Delft, the Netherlands.
Download at: http://vwww.stillerverkeer.nl/index.php?section=&page=actuelelijst
Page 68 of 124
Bicket, P.; et al (2006): “HEATCO – Deliverable 5, Proposal for harmonised guidelines”.
Contract No. FP6-2002-SSP-1/502481, Lead contractor: IER, Germany. 2006.
Bijma, R. W. (2006): "Textuurmetingen, stroefheidsmetingen en gedetailleerde visuele
inspectie op stille wegdekken (Uitvoering 2006)". Project number e0601378, KOAC-NPC, 10
November 2006, KOAC, Apeldoorn, the Netherlands.
Bredahl Nielsen, Carsten; Bendtsen, Hans; Andersen, Bent; Ertman Larsen, H.J.
(2005): "Noise reducing pavements in Japan. Study Tour Report." Danish Road Institute,
Road Directorate. Technical note 31, 2005. (www.roadinstitute.dk).
Brosseaud, Y. (1996): “Revetements Routiers: les betons bitumineux très minces: caractères
spécifiques, bilan des realisations et perspectives”, Proceedings of the Eurasphalt&Eurobitume Congress, Strasbourg, May 7-10, 1996.
Brosseaud, Y.; Adbadie, R.; Legonin, R. (1997): “Couches de roulement très minces et
ultra-minces en matériaux bitumineux à chaud, bilan d’emploi et comportement”, Bulletin des
Laboratoires des Ponts et Chaussées n°207, pp. 55-71.
Brosseaud, Y. ; Abadie, R. ; Legonin, R., (1997): “ Bétons bitumineux très minces et ultraminces”, Observatoire des techniques de chausses n°94 published by S.E.T.R.A., 92223
Bagneux Cedex, France, ISSN 1250-8683.
Carlson, D. D. (2009): "What's New with Asphalt Rubber". Arizona State University Pavement, Materials and Transportation Conference, Arizona State University, Tempe, AZ, USA,
February 2-4, 2009.
Cooee (2011): Cooee is a Danish project started in 2011 focusing on rolling resistance of road
surfaces; see the project website at: http://www.cooee-co2.dk/
CROW (2004): “The method Croad 2004 for traffic noise”. Publication n° 200 of CROW (in
Dutch), CROW, P.O. box 37, 6710 BA Ede, the Netherlands (www.crow.nl) (April 2004,
ISBN 90 6628 401 3).
CROW (2007): “Karakteriseren van dunne asfaltdeklagen”. Publication n° 239 of CROW (in
Dutch), CROW, P.O. box 37, 6710 BA Ede, the Netherlands (www.crow.nl) (March 2007,
ISBN 978 90 6628 486 9).
CROW (2010): Spreadsheet from CROW, P.O. box 37, 6710 BA Ede, the Netherlands
(www.crow.nl). Downloadable from the following website:
CRTN (1988): “Calculation of Road Traffic Noise (CRTN)”, Department of Transport,
Welsh Office, Her Majesty’s Stationery Office, 1988.
Delorme, J-L. (1993): “Formulation mise en oeuvre et adherence des enrobes bitumineux”,
Bulletin de liaison des Laboratoires des Ponts et Chaussées n°185, pp. 139-143.
Page 69 of 124
EAPA (2007a): “Long-Life Asphalt Pavements – Technical version”. European Asphalt
Pavement Association, Brussels, Belgium (http://www.eapa.org/).
EAPA (2007b): “Asphalt figures 2007“. EAPA, European Asphalt Pavement Association,
Brussels, Belgium (http://www.eapa.org/).
EAPA (2008): “Asphalt figures 2008“. EAPA, European Asphalt Pavement Association,
Brussels, Belgium (http://www.eapa.org/).
ECRPD (2010): “Integration of the Measurement of Energy Conservation in Road Pavement
Design, Maintenance and Utilisation (ECRPD)”, European project under the Intelligent
Energy Executive Agency (IEEA), acting under powers delegated by the European
Communities, see: http://www.roadtechnology.se/ecrpd.eu
Fujikawa, T.; Oshinoa , Y.; Mikami, T.; Tachibana, H. (2004): “Examination on Effects
of Road Roughness Parameters for Abating Tyre/Road Noise", CD-ROM Proceedings,
Inter.Noise conference 2004, Prague, the Czech Republic.
Graetz, B. (1998): “Langzeitwirkung von Duennen Schichten bezueglich der Erhaltung
relevanter Oberflaechenmerkmale”, Bitumen, vol 60(2), pp.67-70.
Groenendijk, J. (2004): Dunne deklagen als standaard maatregel voor groot onderhoud?.
Project number e0400766, KOAC-NPC, 11 November 2005, Apeldoorn, the Netherlands.
Groenendijk, J. (2011): "Thin asphalt layers, road surfaces of the future? Case study: TAL in
the Netherlands". Presentation at the final OPTHINAL seminar, Sterrebeek, Belgium.
Groenendijk, J.; Gerritsen, W. (2006): Vraagspecificaties voor Stille Wegdekken, Project
number e0500969, 2006, KOAC-NPC, Apeldoorn, the Netherlands. Downloadable from:
Gummiasfalt (2010): "Seminarium om gummiasfalt". Website for seminar on asphalt rubber
pavements, in Gothenburg, Sweden, 29-30 September 2010: http://www.gummiasfalt.se/
Gustafsson, M.; Blomqvist, G.; Gudmundsson, A.; Uthus, N. S. and Juneholm, M.
(2009): "Tyre induced particle emission from pavement abrasion". ETTAP09 - 17th Transport
and Air Pollution Symposium - 3rd Environment and Transport Symposium, Toulouse,
France, June 2009.
Hammoun, F.; Hamlat, S.; Terrier, J.P. (2008): “Laboratory evaluation of the resistance to
tangential forces of bituminous surfacing”. Proc of the Eurasphalt&Eurobitume Congress
2008, Copenhagen, Denmark.
Hanson, D.I. (2001): “Construction and performance of an ultrathin bonded hot mix asphalt
wearing course”, Transportation Research Record No. 1749, pp. 53-59, Transportation
Research Board, Washington, DC, USA.
Page 70 of 124
Hartjes, J.M.; Voskuilen, J. (2008): “Classificatie van asfaltmengsels met behulp van de
RSAT proef”. Proceedings of ‘Infradagen’, June 2008, Technical University of Delft, the
HAPAS (2011): “The Highway Authorities Product Approval Scheme (HAPAS)”. See:
Hatakeyama, K.; Makabe, T; Emukai, T. (2010): "Study on Ultra Thin Overlay Method
Applicable in Medium to Low Temperature Range". Proceedings of the 11th International
Conference on Asphalt Pavements, International Society for Asphalt Pavements (ISAP),
Nagoya, Japan, August 1-6, 2010.
Hicks, R.G. and Epps, J. A. (2000): Life Cycle Costs For Asphalt Rubber Paving Materials,
Proceedings of the Asphalt Rubber 2000 Conference, Vilamoura, Portugal, November 14-17,
2000, ISBN 972-95240-9-2.
Highways Agency (2006): “Highways Agency internal guidance on how to treat noise
reducing pavements in noise prediction”, 2006.
ISO/CD 11819-2: “Acoustics – Measurement of the influence of road surfaces on traffic
noise – Part 2: The close-proximity method”, ISO/CD 11819-2:2000, 2000-12-13
Jacobson, T.; Wågberg, L.-G. (2004): "Prediction models for pavement wear and associated
costs". Swedish National Road and Transport Research Institute, Linköping, Sweden.
Johnson, Ann M. (2000): “Best Practices Handbook on Asphalt Pavement Maintenance”.
Professional Engineering Services, Ltd. 213 Townes Lane, Wayzata, MN 55391.
Judycki, J.; Jaskula, P. (2003): “Testing of performance properties of asphalt mixes for thin
wearing courses”. Proceedings of the 6th RILEM Symposium, Zurich, Switzerland.
Kaloush, K., Biligiri, K., Rodezno, M., Belshe, M., Way, G., Carlson, D., and Sousa, J.
(2008): Asphalt Rubber Asphalt Concrete Friction Course Overlay as a Pavement
Preservation Strategy, the 4th International Gulf Conference on Roads, Doha, Qatar, 10-13
November 2008.
Kirk, J. V.; Holleran, G. (2000): Reduced Thickness Asphalt Rubber Concrete Leads to
Cost Effective Pavement Rehabilitation, Synopses for 1st International Conference World of
Pavements Sydney, Australia, February 20-24, 2000.
Kragh, J. (2007a): “Traffic noise at rumble strips – Internoise paper 2007”, Danish Road
Institute / Road Directorate, Report 156, 2008, www.roadinstitute.dk
Kragh, J. (2007b): “Noise Classification - Asphalt Pavement, Danish Road Institute / Road
Directorate, Technical Note 61, 2007, www.roadinstitute.dk
Kragh, J. (2008): “Road Surfacings - Noise reduction time history”, Danish Road Institute /
Road Directorate, Report 161, 2008, www.roadinstitute.dk
Page 71 of 124
Lee, David (2010): Personal communication with David Lee, Department for Transport,
Highways Agency (member of the PEB of this project).
Litzka, J.H.; Pass, F.; Zirkler, E. (1994): “Experiences with Thin Bituminous Layers in
Austria”. Transportation Research Record 1454, Transportation Research Board, Washington,
McDaniel, R. S.; Thornton, W. D. (2005): "Field Evaluation of a Porous Friction Course
for Noise Control." In Proceedings (CD-ROM) of the TRB 84th Annual Meeting,
Transportation Research Board of the National Academies, Washington D.C., 2005.
MIRIAM (2011): MIRIAM is an international project started in 2009 focusing on rolling
resistance of road surfaces; see the project website at: http://miriam-co2.net/
Nicholls, J.C.; Daines, M.E. (1992): “Spray Suppression by Porous Asphalt”, Highways
Resource Centre, Transport Research Laboratory, Second International Symposium on Road
Surface Characteristics, Berlin, 1992.
Nicholls, Cliff; Carswell, Ian; Gibb, Mike; Williams, John (2006): "Service Lives of Thin
Surfacing Systems in the UK". Paper at the conference Transport Research Arena Europe
2006, Gothenburg, Sweden.
Nielsen, C. B.; Nielsen, E.; Andersen, J. B.; Raaberg, J. (2004): Development of Durable
Porous Asphalt, Mixes From Laboratory Experiments. Road Directorate/Danish Road
Institute, Danish Asphalt Industries, NCC Roads. Proceedings 3rd Eurasphalt & Eurobitume
Congress Vienna 2004 – Paper 090.
Nielsen, C. B.; Nielsen, E.; Raaberg, J. (2005): Durability of porous asphalt (in Danish
Holdbarhed af drænasfalt). Report 139, 2005, Danish Road Institute, Road Directorate
Nielsen, E.; Raaberg, R.; Bendtsen, H. (2006): Noise reducing SMA pavements – Mix
design for SILENCE - F2. Technical note 39, 2006, Danish Road Institute, Road Directorate
Olsson, Kenneth (2010): Interview of Mr Kenneth Olsson, Skanska AB, Farsta, Sweden, by
Mr Erik Olesen, DRI.
Perveneckas, Zigmantas; Vaitiuk, Valerija (2009): “Pilot Sections of “Hot On Hot” asphalt
Paving Technology in Lithuania”. Lithuanian Road Administration.
Resendez, Yadhira A.; Sandberg, Ulf; Rasmussen, Robert O.; Garber, Sabrina (2007):
"Characterizing the splash and spray potential of pavements". Report from the Transtec
Group, Inc., Austin, TX, USA (produced for FHWA, May 2007).
Page 72 of 124
RMV (2006): Prescription for the Measurement and Calculation13) of Traffic Noise, RMV
2006 (in Dutch). May be downloaded from www.stillerverkeer.nl. The site is an initiative by
The (Dutch) Ministry of Housing, Spatial Planning and the Environment, the (Dutch)
Ministry of Transport, Public Works and Water Management, and CROW, P.O. box 37, 6710
BA Ede, the Netherlands (www.crow.nl).
Rungruangvirojn, Pichaya; Kanitpong, Kunnawee (2009): "Measurement of Visibility
Loss Due to Splash and Spray: Comparison Between Porous Asphalt, Stone Matrix Asphalt,
and Conventional Asphalt Pavements". TRB presentation 09-1913, Transportation Research
Board, Washington, DC, USA.
Sabita Guideline (2008): “Guideline for thin layer hot mix asphalt wearing courses on
residential streets”. Manual 27, May 2008, ISBN 978-1-874968-33-7, published by Sabita,
Howard Place 7450, South Africa.
Sandberg, Ulf; Ejsmont, Jerzy A. (2002): "Tyre/Road Noise Reference Book". Informex,
SE-590 39 Kisa, Sweden (www.informex.info).
Sandberg, Ulf; Masuyama, Yukieei (2005): “Japanese machines for laying and cleaning
double-layer porous asphalt – Observations from a study tour”. Swedish National Road and
Transport Research Institute (VTI), and Chalmers University of Technology, Gothenburg,
Sweden and Laboratory of Seikitokyu Kogyo Co., Iwafune-Cho, Japan.
Sandberg, Ulf (2010): "Asphalt rubber pavements in Sweden – Noise and rolling resistance
properties". Proc. of Inter-Noise 2010, Lisbon, Portugal, 13-16 June 2010.
Sandberg, Ulf (Editor) (2011): "Rolling Resistance – Basic Information and State-of-the-Art
on Measurement methods". Deliverable # 1 in MIRIAM SP 1; see [MIRIAM, 2011].
Schulze, C.; Gharabaghy, C. (2008): “Kronausbrüche bei Offenporigen Asphalten –
Prognoseprüfung und Bewertung”, Aachener Strassenbau- und Verkehrstage, Institut für
Strassenwesen – RWTH, Aachen, Germany.
Schwanen, W.; van Leeuwen, H.M.; Peeters, A.A.A. (2007): "Acoustic optimization tool
RE3: Measurement data Kloosterzande test track". Report M+P.DWW.060.44.8. of the
subproject 1.2.4. of the Dutch “Noise Innovation Programme” Rijkswaterstaat (RWS), Dienst
Verkeer en Scheepvaart (DVS), P.O. box 5044, 2600 GA Delft, the Netherlands.
Setra (1997): “Bétons bitumineux très minces et ultra-minces”, Observatoire des techniques
de chausses n°94.
Sousa, J. B. (2006): State of the Art on Asphalt Rubber, Asphalt rubber 2006 Proceedings,
Palm Springs, California, October 25-27, 2006, ISBN 962-405-091-0.
Sousa, J. B., Way, G. B., and Carlson. D. D. (2006): Energy Consumption of Alternative
Scrap Tire Uses, Proceedings of the Asphalt Rubber 2006 Conference, Palm Springs,
California, October 25-27, 2006, ISBN 962-405-091-0.
RMV = Reken- en Meetvoorschrift = Calculation and Measurement Prescription
Page 73 of 124
SPENS (2009): "Sustainable Pavements for European New Member States – Summary
Report". Report available from FEHRL, Blvd. de la Woluwe, 42, 1200 Brussels, Belgium.
Also downloadable from: http://spens.fehrl.org
Sridhar, R.; Sureshia, S.N.; Bose, S.; Veeraragavan, A. (2005): “Study of polymer and
fibre modified open graded friction courses”, Indian Roads Congress Journal vol. 66(2), 291318.
Svedberg, Bo (2010): Call for seminar 2010-05-12 on "Framtidens materialförsörjning, en
utblick (Future supply of material – a prospective survey)". Institutionen för samhällsbyggnad, Luleå Tekniska Universitet, Luleå, Sweden.
Tækker, Christina (2009): ”Messina bridge heads into basic design stage”. COWI,
Denmark. Document downloadable from the COWI website
Thomsen, S.; Bendtsen, H.; Andersen, B. (2005): "Thin noise reducing pavements for urban
roads - Measurement report". SILVIA-DTF-DRI-009-02-WP4-20050801, SILVIA deliverable, downloadable from www.trl.co.uk/silvia
Thomsen, S.; Bendtsen, H.; Andersen, B. (2006): Noise reducing thin pavements. Urban
roads. Danish Road Directorate/Road Institute, DRI report 149, 2006 (www.roadinstitute.dk).
Thomsen, S.; Bendtsen, H.; Andersen, B. (2008): Optimized thin layers - urban roads, the
Kastrupvej experiment. Technical note 66, 2008, Danish Road Institute, Road Directorate
van Vliet, W.J. (2007): Personal communication, 2007, between Mr H. Bendtsen and Mr
van Vliet, Rijkswaterstaat, Dienst Verkeer en Scheepvaart, Postbus 5044 2600 GA Delft, the
Walsh, Ian (2009): “London asphalt specification: Guidance on the selection and
recommendations for the use of road surfacing materials and European Standards for
Asphalt”. Job No. 11412, The London Technical Advisors Group (LoTAG), Transport for
London; published by Jacobs UK Ltd. Downloadable at the website (check for latest edition):
Way, G. B.; Carlson, D. D.; Sousa, J. B.; Kaloush, K. E.; and Biligiri, K. P. (2010):
Introduction To Asphalt-Rubber Pavement Noise Reducing Characteristics, Inter Noise 2010,
Lisbon, Portugal, 13-16 June 2010.
VBW (2004): "Richtlijn dunne asfaltdeklagen". Vereniging tot Bevordering van Werken in
Asfalt, April 2004, Breukelen, the Netherlands. Downloadable from:
Wegan, Vibeke (2001): "Surfacing of concrete bridges". Report 106, (2001), Danish Road
Directorate/Danish Road Institute, Hedehusene, Denmark. Downloadable at:
Page 74 of 124
Wirtgen (2003): "Material and Process Procedure Principles for the Hot Recycling Process".
Report No. 27-30, Wirtgen GmbH, Windhagen, Germany.
VROM (2002): “Guideline Low Noise Pavements, background document for the “Regulation
for the Stimulation of the Application of Low Noise Pavements” (in Dutch), Dutch Ministry
of Housing, Spatial Planning and the Environment, 2002.
Zareh, A.; Way, G. B. (2009): Asphalt-Rubber 40 Years of Use in Arizona, Asphalt rubber
2009 Proceedings, Nanjing, China, November 2-4, 2009, ISBN: 978-988-18681-1-4.
Page 75 of 124
project OPTHINAL - Optimization of thin asphalt layers
Task 2: Inventory
Author: Luc Goubert, BRRC
1. Introduction
Thin asphalt layers have been used in several European countries and abroad for more than 15
years with mostly good results. They seem to be pavements that are cost effective, fast lying
and may have a good surface. Developments in recent years show that thin asphalt layers can
reduce noise, increase traffic safety (skid resistance and visualization during wet condition)
and be durable.
In the frame of ERANET ROAD II 14), a call was issued end 2008 for a comprehensive study
of this road surface type. The purpose of the project is the optimization of thin and very thin
asphalt surfacing in thicknesses from 15 to 30 mm.
A consortium of the Danish Road Institute, the Swedish National Road and Transport
Research Institute and the Belgian Road Research Centre has won the tender and the project
was initiated on the 1st of July 2009 15).
Gathering of detailed information about use of and experiences with thin layers in the
different European countries (and if possible also abroad) are an important part of the study.
A literature survey is carried out, but it is our experience that by far not all knowledge and
experience can be found in regular literature, but remains often hidden e.g. within
unpublished research reports of institutes, contractors etc. In order to make our study as
complete as possible, we would like to ask a few questions to a selection of distinguished
specialists in the field. The questions are listed on the subsequent pages.
We thank you in advance for your time and effort to complete them. If the answers can be
found in documents (like research reports), please feel free to simply refer to that document
More information on ERA NET ROAD can be found on the website: www.eranetroad.org
the effective start of the project has been delayed later on, due to administrative issues
Page 76 of 124
and enclose them (or tell how we can obtain it). Please note that it could also be very useful
for the project if you only answer to some of the questions listed in this document.
The completed forms should be returned to:
Luc Goubert
Research scientist
Belgian Road Research Centre
Woluwedal 42
B-1200 Brussels
E-mail: [email protected]
2. Questions
2.1. How much m² of thin layers do you have in your country/region? Which percentage
of top layers consists of thin layers? Do you have information about ratio thin
layers 16)/very thin layers 17)?
2.2. Which are the motivations to use thin layers in your region/country? If there is more
than one, please rank in order of importance. In case you don’t use thin layers, do you
have a motivation not to use them?
2.3. Which types are used?
2.3.1. thickness
2.3.2. void content
2.3.3. modifier
2.3.4. tack coat
2.3.5. type of skeleton/grading curve
2.3.6. which lab tests are performed (a priori tests)
2.3.7. which is the basic document specifications system or EU standard
2.4. Which a posteriori (in situ) performances are imposed and how and when are they
2.4.1. skidding resistance
2.4.2. noise
2.4.3. evenness
2.4.4. texture (MPD…)
2.4.5. durability
2.5. Which good or bad experiences did you have with thin layers (if any)? Are there
recurring problems?
2.6. Did you find specific solutions for problems?
2.7. Which are the remaining problems/research needs?
2.8. How do you see the role of thin asphalt layers in the future?
20 – 30 mm thick
< 20 mm thick
Page 77 of 124
List of addressees for questionnaires
The OPTHINAL questionnaire was sent to the experts listed in Table A1 together with their
affiliations and e-mail addresses.
Table A1: Receivers of the questionnaire on TAL distributed to a selection of European
pavement experts.
M. Ahadi
[email protected]
T. Ahokas
Tero Ahokas
[email protected]
J. Aksnes
Statens vegvesen Vegdirektorat
[email protected]
E. Andersen
Kolo Veidekke
[email protected]
P.J. Andersen
Vejdirektoratet (Road Directorate)
[email protected]
J. Bork
Sønderborg Kommune
[email protected]
Y. Brosseaud
[email protected]
K. Eskola
[email protected]
A. Fossbakk
[email protected]
J. Groenendijk
[email protected]
R. Hofman
[email protected]
B.-B. Jensen
NCC Danmark
[email protected]
L. Ladehoff
[email protected]
K. Laukkanen
[email protected]
G. Licitra
Arpat Toscara
[email protected]
K. Lundström
Vägverket (National Road Administration)
[email protected]
J.C. Nicholls
[email protected]
K. Olsson
[email protected]
B.W. Sluer
BAM Wegen bv
[email protected]
G. van Bochove
[email protected]
E. Van den Kerkhof
[email protected]
B. Wamsler
Vejdirektoratet (Road Directorate)
[email protected]
Page 78 of 124
Results of the interviews
A.3.1 Summary of interview with an expert from the Netherlands
Interviewers: Luc Goubert and Anneleen Bergiers
Dutch specialist: Wim van Keulen (van Keulen advies, Vlijmen)
Location and date of the interview: Vlijmen, May 18 2010
The original motivation in the Netherlands to use thin layers was adapted from France,
namely their capacity of noise reduction. Later, their low cost became the main motivation.
The main future objective is a higher durability. The functional life time is 7 years; at which
time an increase of noise by 2 dB at 50 km/h has been reported. A bitumen layer may be
applied on the surface or a fine very open asphalt concrete may be installed, which extends
life time with one to three years.
Some laboratories already have developed new thin layers with longer life time (10-15 years)
but current market conditions do not stimulate launching it on the market. Newly developed
thin layers contain a higher percentage of bitumen and fewer voids, namely 11-14 %
accessible voids, by adding a specific small sand fraction which fills up the voids.
Wim van Keulen expects the application of thin layers to remain limited even when durability
is improved because of the limited need. An important problem is the restricted time frame to
use planned budgets which stress pressure on installations even within bad weather conditions
which causes problems with durability. A possible solution is a budget which is time independent.
A.3.2 Summary of the interview with a Norwegian panel of specialists
Interviewer: Luc Goubert
Panel of Norwegian specialists: Jostein Aksnes (NPRA Road Directorate), Geir Refsdal,
Rolf Johansen and Jan Lindahl (NPRA, Eastern Region), Olle R. Larsen (Kolo
Location and date of the interview: Oslo, June 29 2010
One has to distinguish between three “types” of asphaltic layers which are thinner than 30 mm
and which are nowadays used in Norway:
o Ultra thin (Novachip etc.) (also referred to as type 1 hereafter)
o Thin surfacings 18) according to the “Norwegian Pavement Design Guidelines” (type
o “Ordinary” asphalt concrete or SMA applied as a thin layer (typically 50 – 60 kg/m²)
(type 3)
These are in principle not in the scope of the OPTHINAL project
Page 79 of 124
Reasons why it is (not) used
In Norway, about 6.2 million m² of TAL are applied, mainly on roads with AADT less than
5000. In that case, the wear of the road is not much influenced by studded tyres, but rather by
other factors like frost/heave. The use and hence negative impact of studded tyres on roads
has decreased by 25 % in Norway, due to improvement of “ordinary” winter tyres, the use of
de-icing salt and the introduction of fees for entering certain areas with studded tyres.
The reason why TAL of type 3 are quite widely used in Norway on these low volume roads
are mainly economical, rather than environment (noise) or safety driven. With TAL, simply
less material is needed per m². Maximum aggregate size is 11 or 16 mm, mainly 11 mm. In
performance related contracts 16 mm is used.
Ultra thin layers (type 1) are not so widely used, as their ranking lifetime/cost is not very
good, in none of the AADT classes. When they are applied, also rather large aggregate sizes
are used (11 or 16 mm). Remarkably, 16 mm aggregate sizes are used in a 20 mm Novachip
ultra thin surface layer.
Special attention is paid to the tack coat, especially when a thin asphalt layer (or a surface
dressing) is applied in any kind of crossing.
Thin asphalt layers are often applied on roads showing ruts. The intervention level for ruts is
in Norway 25 mm of rut depth on at least 10 % of the section. Specific problems occur when
applying a thin asphalt layer or a surface dressing:
o in ruts, the emulsion is not reaching the top of the layer (for surface dressings)
o compacting is difficult in the ruts (the use of a compacter with pneumatic tyres could
probably solve this problem)
o thin asphalt layers cool very quickly after application. The use of more than one roller
can be a solution for this problem
Solutions applied in Norway to address the specific problem of ruts are the prior milling of
the surface or the application of a levelling layer. A technique commonly used in Norway is
preheating with infrared before applying the thin asphalt layer, which allows the large
aggregate to be partially pressed into the sub layer, allowing the use of large aggregates in
relatively thin layers.
Noise becoming more and more an issue in densely populated areas, thin layers with
maximum chipping size 8 mm are more and more used (of all three above mentioned types).
To counter the increased susceptibility to wear due to the smaller chipping size, special
attention is paid to the aggregate quality and PmB is used.
Bad experiences
The main bad experience with thin layers is the relatively quick cooling when applied at lower
ambient temperatures, e.g. in autumn.
Other doubts are about the skid resistance: Norwegian experience is that in general the skid
resistance is lower the larger the maximum aggregate size is. Studded tyres tend to improve
Page 80 of 124
skid resistance, but too much use of them is bad for the durability of the road surface.
Discussions are ongoing about the optimal percentage of studded tyres in the car population.
Another observation which is made in Norway is that the ultra thin layers tend to wear out
before the threshold for rutting is reached.
TAL are not used on a number of roads because of the poor bearing capacity of the
underground. A thicker wearing course is in that case needed to provide the necessary bearing
capacity of the road.
A priori tests
A priori tests routinely carried out are:
aggregate quality
grading curve
binder content
void content (not for ultra thin type)
Laboratory tests for durability and deformation resistance are under development.
Tests in Norway have shown that adding a polymer to the binder results in 30 – 35 % less
deformation for dense asphalt at 50 °C and 10000 wheel passes. Nevertheless, polymers are
not routinely used for economic reasons on thin layers of the DAC type. Craftsmanship is
considered to be at least as important as mixture design; especially to avoid inhomogeneity,
mainly caused by segregation of aggregates and/or of binder. To stimulate craftsmanship
more and more performance based contracts are used.
The methods used are standardised in a Norwegian handbook for road building, but differ
from one region to another due to varying climatic conditions. In the north of the country preheating is systematically used. CEN standards about asphalt are strictly followed in Norway.
A posteriori tests
Skid resistance is only measured when there appears to be a problem. Initial rutting is
measured in some cases, namely on highly trafficked roads. Evenness is also measured
sometimes, but this is not a standard requirement.
The whole road network is systematically monitored once a year. Pictures are taken of every
20 m section and put into a database. This allows following the evolution of the state of the
roads and contractors get access to this database. MPD values are also systematically
measured and fed into the database, but these data are not used so far.
Noise is only measured in the frame of research projects. It is done with the NPRA CPX
trailer, which is normally operated by SINTEF.
Page 81 of 124
Remaining problems/research needs
Thanks to the systematic annual monitoring, a large volume of data is available and lots of
answers can be distilled from these figures. E.g. the average service lifetime for ultra thin
layers (type 1) as a function of the AADT:
AADT < 1500
3000 – 5000
5000 – 10000
AADT > 10000
13.5 years
11 – 13 years
8.5 – 10.5 years
6 – 7.5 years
Research should aim at extending the service life and or reducing the costs. As noise is
becoming a more and more important issue, it should also be addressed. The balance between
the wear of the stones and the mortar should be studied in order to avoid (or at least reduce)
the increase of noise by the wearing away of the mortar: the wearing away of the mortar
causes an increase of the texture. A better “balanced” wearing would maintain a smoother and
less noisy road surface. Moreover, a road surface with larger stones (> 10 mm) at the surface
tends to be slippery.
It is believed that mixtures with stronger mortars (more wear resistant) and softer binders
could be interesting. Stones could gradually be pressed into the mortar during lifetime,
leading to a permanently smooth road surface, even under the action of studded tyres.
The future
It is expected that TAL will be more and more used in the cities to abate traffic noise. The
wide use of TAL is also a matter of volumes and political choice. If the contractors would be
given a guarantee that a substantial amount of TAL will be constructed the coming years, it
would encourage them to invest in machinery. It does not pay if every now and then “a little
bit” of TAL is ordered by the road authority.
Concluding remarks
After all, the scope of application of TAL is rather narrow:
o not too high AADT (maximum 3000 – 5000)
o enough bearing capacity of the under layers
o without severe levelling problems
But a slight increase of the use of TAL is, nevertheless, expected the coming years.
It is also mentioned that in 2011 a 4 – 5 years Norwegian national research project will start
aiming to improve lifetime of road surfaces in general. It is not specially focused on TAL, but
they will be considered. Budget of the project will be between 20 and 40 million NOK.
Page 82 of 124
A.3.3 Summary of the interview with a Swiss specialist
Interviewer: Luc Goubert
Swiss specialist: Alain Jacot (Société d’Analyses & Contrôles Routiers, SACR, Zürich)
Location and date of the interview: Zürich, June 30 2010
In 2008 there was a reorganization of the management of the major road network in Switzerland. The total network of national roads, comprises a network of 2000 km of highways, was
brought under the responsibility of the Federal Authority. The transfer caused a major impetus
with respect to the abatement of road traffic noise in Switzerland. A global noise map was
made, based on point measurements and calculations (interpolations), giving rise to actions to
abate the traffic noise on black spots along the national roads.
The decision was taken to use henceforth only noise reducing wearing courses on the
highways and the standard wearing course applied is a Swiss TAL version “AC MR8” 19). PA
8 20) is no longer used due to its vulnerability to damage caused by the widely used snow
chains in winter period.
In Switzerland between 5 and 6 million m² of AC MR8 are constructed each year. Up to now
it has been applied on about 50 % of the highway network, which represents between 25 and
30 million of m².
Reasons why it is (not) used
The reason why it is used is entirely the noise reduction.
Bad experiences
In Switzerland, one sees hardly other advantages for TAL than noise reduction. One is not
sure that TAL are cheaper than conventional road surfaces, when one considers the whole life
cycle. Another disadvantage is the more difficult construction: compaction is delicate due to
the fast cooling down of the mixture once it is applied. Also problems with initial skid
resistance have been reported.
Types used
AC MR8 is laid with a layer thickness of 20 up to 30 mm and has between 5.5 and 13 % of
voids. PmB is always used. For the tack coat also modified bitumen is applied (dose 100 –
200 g/m²).
In Switzerland, one uses the available European Standards for the specifications of the
materials for bituminous mixtures, which are adapted by the National Swiss Association for
asphalt concrete “macro-rugueux” with 8 mm maximum chipping size
porous asphalt with 8 mm maximum chipping size
Page 83 of 124
Standardization (SNV). These standards are given a national annex “NA” which fixes limits.
A series of Swiss standards describes the conception, construction and requirements for the
laid asphalt courses. An overview of the relevant EN and Swiss standards can be found in the
Swiss standard SN 640 420b. The standard describing the conception, construction and
requirements for the laid AC MR8 is the SN 640 430b. Requirements for the mixture are fixed
in SN 640 431-1b-NA.
A priori tests
Contractors must have a certificate for their mixtures, which remains valid for a maximum of
five years. In these certificates, one can find:
granulometry (see example in Figure A1)
binder content
binder specifications (PEN value etc.)
voids content
results of Marshal test
results of a rutting resistance test (for mixtures intended only for heavily trafficked
Figure A1: Sieving curve limits for the Swiss gap-graded TAL AC MR8, which is adapted as
a standard wearing course on Swiss highways (Source: SN 640 431-1b-NA)
Page 84 of 124
Road authorities take samples of the mixture which are tested in certified laboratories for
compliance testing. During the construction of the road, compaction is tested. Producers of
asphalt mixtures must have their products tested by a certified laboratory and calculate the
moving average of the parameters.
A posteriori tests
Skid resistance is measured 2 – 3 months after construction of new wearing course on highways; for other roads it is not tested systematically. Longitudinal evenness is tested quite
often by means of the APL 21), but transversal evenness is only measured in case of suspected
problem. The “water value 22)” must not exceed 4 mm on highways or 6 mm on other roads.
The acoustic quality of the new wearing courses are not checked systematically either, but
every four years the federal authorities assess the quality on 7000 km of the main road
network. Parameters measured or observed are:
visual inspection
longitudinal and transversal evenness
skid resistance
acoustic quality by means of noise measurement by the CPX method (first time in 2010)
Remaining problems and research needs
As already mentioned, compacting properly is a problem with TAL and a second remaining
problem is the uncertainty of the lifetime of TAL.
The laboratory of traffic facilities in Lausanne (LAVOC) is doing research to solve these
problems. One will look for the optimal position between noise reduction and voids contents.
For the solution of the compacting problem one is working on lukewarm mixtures. The use of
these mixtures would allow a longer time for compacting the layer. The inconvenience would
be the hardening time, which is two to three days with a lukewarm mixture, which is much
longer than the two to three hours needed for a conventional mixture to harden.
Another alternative to ease the compaction of TAL which is studied is the “bituminous foam”
which is obtained by using humid aggregates for the mixture preparation or by adding
chemical products to the mixture (like wax).
Pilot research has been carried out by LAVOC; now the testing by means of test sections is
being tendered.
The future
Swiss highways will soon be covered by 100 % with TAL, but also on other roads it will be
used more and more (also the variant AC MR4 with maximum aggregate size 4 mm and a
layer thickness of 15 – 20 mm). Politically, priority is given to traffic noise abatement and one
accepts the less good durability.
Analyseur de Profil en Long
maximum depth of puddles on the road surface
Page 85 of 124
A.3.4 Summary of the interview with a panel of Austrian specialists
Interviewer: Luc Goubert
Panel of Austrian specialists: Johann Litzka (Austrian Association for Research on
Road - Rail – Transport, FSV), Jürgen Haberl (PMS-Consult, Engineering Office for
Traffic and Infrastructure), Peter Riederer (BPS, Oberösterreichische Boden- und
Baustoffprüfstelle GmbH), Michael Kostjak (Swietelsky BaugesmbH), Ronald Blab
(Vienna University of Technology)
Location and date of the interview: Vienna, July 2 2010
Figures about the area of TAL in Austria are available only for highways. The total area of
very thin layers (thinner than 20 mm) on Austrian highways is 1.1 million m² and the area of
thin layers (thickness between 20 and 30 mm) amounts to 1.45 million m². The ratio between
the two is hence 1:1.3. Between 10 and 12 % of the road surfaces on the highways consists of
No information is available for secondary roads, which are managed by the states of Austria
(Länder). Neither is information available about roads owned by the cities.
Types used
Three types of TAL which are most common in Austria:
LDDH8 (Lärmmindernde Dünnschichtdecke Heiβ, Noise reducing thin layers
“hot” 23), with maximum aggregate size 8 mm)
LSMA8 (Lärmmindernde Splitt Mastik Asphalt, Noise Reducing Stone Mastic Asphalt,
with maximum aggregate size 8 mm)
Voids content for LSMA8 is typically 10 – 11 %, which is considered to be a good
compromise between sound absorption and durability. The LSMA types have binders with
rubber or polymer modified bitumen.
Also DAC and SMA are sometimes applied in wearing courses with a thickness lower than or
equal to 30 mm: DAC 4 24) is applied in layers with thickness between 20 and 30 mm; DAC 8
and SMA 8 are applied in thicknesses between 25 and 35 mm and AC 11 and SMA 11 in
layers between 30 and 40 mm. There are also thin layer types “BBTM 25)” and AC deck D A3,
which are applied in thicknesses up to 25 mm. Maximum aggregate size is in both cases 4 or 8
indicating that the mixture is applied hot
DAC is indicated in RVS as “AC deck”
beton bitumineux très mince, very thin asphalt layer
Page 86 of 124
For the mixtures the available EN standards are used, but for the constructions a national
guideline RVS 08.16.01 26) is used.
For the roads managed by the Federal Ministry, complying with RVS 08.16.01 is mandatory;
for the state owned roads this is recommended.
Reasons why TAL are used
There are several reasons why TAL are being used in Austria:
o good resistance to rutting
o economy (materials used for the top layer are very expensive; high quality aggregates
and high quality bitumen must be used)
o especially for highways, noise reduction is very important
On highways, TAL types with 11 mm maximum chipping size are used.
The types are specified in the RVS, while the performance is specified in standards.
A priori tests
The PSV value of the aggregate for the wearing course is always checked.
A posteriori tests
Noise is sometimes assessed; namely on road surfaces which were intended for noise
reduction; not on the other types of roads. Measurements are done with an Austrian CPX-like
device (see Figure A2), which resembles strongly the ISO CPX method, but not completely.
The Austrian CPX method is described in an RVS. Skid resistance is always tested on
highways. Conformity with (longitudinal) evenness requirements is always checked by means
of the 4 m straight-edge. Transversal evenness is not checked.
There is a texture requirement; the MTD 27) value must be at least 0.4 mm. First, a visual
inspection is made and if there seems to be “suspect” parts of the road surface, MTD is
checked by means of the sand patch method.
The durability is not checked, but by means of an a priori test, the ageing of the bitumen is
Bad experiences
The problem which is most common is debonding, by failure of the tack coat. This occurs
mostly when a TAL is applied on a cement concrete under-layer and when paving is not done
correctly. Rutting is a problem when TAL is applied in a too thick layer. A general problem
with bituminous wearing courses which occurred the last years is the varying quality of the
bitumen. Due to the expensive oil, one tries to extract as much as possible fuel during the
Richtlinien und Vorschriften für den Straβenbau, Guidelines and Prescriptions for Road Construction, drafted
by the FVS, not to be confused with national standards, indicated with ÖNORM, which are made by ÖNI (Österreichischen Normungsinstitut)
MTD = Mean Texture Depth, measured by the patch method
Page 87 of 124
refining process, leading to lower quality of the remaining material. In spite of the mentioned
problems, TAL are not considered as especially problematic wearing course types in Austria.
Figure A2: The Austrian CPX-like trailer (photo by IFS Ziviltechniker Ges.m.b.H.)
Specific solutions
For a quicker compaction, more compactors are used at the same time, which also requires
specific training of the personnel. One tries to limit the distance between the asphalt plant and
the construction site and one uses covered trucks to prevent excessive cooling of the mixture
during transportation.
Enough tack coat has to be foreseen for the application of TAL on cement concrete and the
concrete surface is brushed before the TAL application.
The debonding problem is addressed in research and a wedge-splitting test has been
developed to assess this aspect in laboratory. Experiments are carried out on different textures
for the under-layer to see which one that gives an optimal bonding. An Austrian standard for
this wedge-splitting test is under preparation.
The future
The experts assume that TAL will have a very important role in Austria, both for economical
and environmental (noise reduction) reasons.
Page 88 of 124
A.3.5 Summary of the interview with an Italian specialist
Interviewer: Luc Goubert
Italian specialist: Gaetano Licitra (ARPAT, Agenzia Regionale per la Protezione
Ambientale della Toscana, and University of Pisa)
Location and date of the interview: Pisa, July 9, 2010
One has no experience at all with TAL in Tuscany and this may be generalized to the whole
of Italy.
Reasons why TAL are not used
On highways, porous asphalt is generally used for its noise-reducing and draining capacity.
One of the main inconveniences encountered in the more northern European countries for
porous asphalt is its problematic winter maintenance, but this is hardly a problem in the mild
Mediterranean climate in Italy. There is hence no incentive to explore other types of noisereducing pavements for the highways.
As to the secondary roads and city streets, there has been little interest and funding in the past
to keep these in a good shape. Multiple spot repairs of wearing courses are very common,
which is not very compatible with the use of low noise pavements.
The future
Nevertheless, things are changing, among others under the influence of the European Noise
Directive and the action plans foreseen in it. Tuscany region has adopted a project called
Leopoldo and its aim is to characterize vehicle emission to serve as input for noise mapping.
Part of the project is also the study of low noise pavements and therefore one has planned the
construction of ten test sections; six low noise pavements and four reference pavements. One
of the test sections, near the village Capolona in the province Arezzo, will have a “microdrain” wearing course, which is a porous asphalt applied as a thin layer. The thickness will be
30 mm and the length of the test section will be 200 m. Its acoustical and structural behaviour
will be monitored and in case of success, it may be used on a large scale to abate traffic noise
in the future. Acoustical monitoring will be carried out with the ARPAT CPX device, which
is a system with a vehicle-mounted microphone outside a test tyre; see Figure A3.
Page 89 of 124
Figure A3: The ARPAT CPX
A.3.6 Summary of interviews with Danish specialists
Interviewer: Erik Olesen
Danish specialists: Bjarne Bo Jensen, NCC Roads A/S, Lars Ladehoff, Colas Danmark
A/S, Uffe Mortensen, Pankas A/S, Niels Christoffersen and Uffe Mortensen, Inreco A/S
Location and date of the interviews: Telephone interviews 7 – 11 June 2010
Thin asphalt layers are widely used in Denmark. Common types are stone mastic asphalt
(SMA), open- or dense-graded asphalt concrete (AC) or UTLAC, all with 6 mm or 8 mm
nominal maximum aggregate size.
Recently, very thin asphalt layers with 4 mm maximum aggregate size have been introduced
by Colas. Inreco remixes existing pavements with new asphalt; in particular for laying SMA
6+ with 6 mm nominal maximum aggregate plus a small amount of 8 mm aggregate. Inreco
used to apply this process in Denmark and Sweden but work in Sweden has been
Amounts applied
In 2007 the total production of hot mix asphalt in Denmark was 3.3 million tonnes, 40 % of
which were wearing course asphalt [EAPA, 2007b]. The corresponding numbers for 2008
were 3.1 million tonnes and 44 %. Based on the distribution of various layer thicknesses
provided by NCC and shown in Table A1, the number of square metres laid 2007 - 2008 in
Denmark is estimated in Table A1, assuming the following typical rates:
thickness < 20 mm: 45 – 55 kg/m² (average: 50 kg/m²)
thickness 20 -30 mm: 50 – 70 kg/m² (average: 60 kg/m²).
Page 90 of 124
The total area of Danish paved roads is in the order of 400 million m2, 40 million m2 of which
constitute the motorways and other national roads.
Colas estimated its application of very thin layers (< 20 mm) to be significantly lower than
that of NCC and Pankas, see further Table A1, while Inreco did not provide an estimate at all.
The last columns of Table A1 show the estimated percentages of the national road network
presently paved with thin asphalt layers.
Typical layer thicknesses and void contents are given in Table A2.
Table A1: Estimated masses, areas and percentages of various layer thicknesses laid per year
in 2007 – 2008 on the national and municipal roads given by three contractors, and the
estimated total coverage on the national Danish road network.
Total TAL built per year
(all roads)
Total TAL area
Colas Pankas (only national
[103 t]
[106 m2]
[103 t]
[106 m2]
< 20
20 - 30
(> 30)
[%] [106 m2]
Table A2: Typical thicknesses and void contents of the common Danish types of thin asphalt
Type designation
In EN 13108
TB 6k
AB 6t
AC 6d
TB 8k
AB 6å
AC 6o
AB 8t
AC 8d
AB 8å
AC 8o
Range of Normal range of
void content
15 – 20
Void content when used as
noise-reducing pavement
No requirement
No requirement
2 -5
20 – 30
Motivation for using thin asphalt layers
The major reasons for applying thin asphalt layers in DK are – in that order – low price, high
quality, noise reduction. The lower cost is, among others, connected with the reduced need for
adjusting kerbstones.
Page 91 of 124
According to Danish specifications [AAB, 2006], which are based on EN specifications, the
following amounts of tack coat of residue binder from modified emulsion shall be applied:
UTLAC 6 (TB 6k): > 600 g/m²; UTLAC 8 (TB 8k): > 700 g/m². For all other types of thin
asphalt layers the tack coat shall be 300 – 400 g/m² of 50 % non-modified bitumen emulsion.
Colas A/S and Pankas A/S increase the amount of emulsion for open-graded wearing courses.
NCC rarely applies modifier in thin asphalt layers while Colas and Pankas to some extent
applies such modifiers.
Requirements on aggregate grading curves are defined in Danish national specifications. As
an example, Figure A4 shows the centre points of tolerance intervals specified for AC 8d, AC
8o and SMA 8, respectively.
Total passing
[% by weight]
AC 8d
AC 8o
Sieve size [mm]
Figure A4: Danish target curves for aggregate grading of AC 8d, AC 8o and SMA 8.
A priori tests
A priori tests are routinely carried out for:
Bitumen softening point
Aggregate quality
Aggregate grading curve
Stability and plastic flow (Marshall test) - not for UTLAC
Binder content
Air void content - not for UTLAC or open-graded types
Light reflection – when specified.
Page 92 of 124
When the bitumen is modified or if the asphalt is modified in-situ, the contractor must specify
the improved properties of the asphalt relative to non-modified asphalt.
A posteriori tests
Skidding resistance requirements are part of Danish national specifications for thin asphalt
layers, except for AB 6t (AC 6d). On national roads, measurements are carried out before the
road is opened for traffic to verify the requirements.
Danish national specifications for thin asphalt layers comprise requirements on evenness. On
the national road network this is verified at the completion of laying the wearing course.
Hence, for construction and acceptance of new pavements a 5 m Viagraph simulation is used
on measured surface profiles. No direct requirements are specified for surface texture.
Once a year the national road network is monitored (by the Danish Road Directorate
Profilograph and ARAN vehicle) to determine pavement condition expressed in evenness
(IRI), rut depth, and texture. Based on the texture measurements (MPD), the national road
network is screened for skidding resistance problems. If indications of lacking skidding resistance are found, then actual skidding resistance measurements are carried out using the
Danish Road Directorate's ROAR equipment. This is usually the case for 5 – 10 % of the
national road network.
If local (municipality) roads look fine, no further monitoring is done; else some measurements
may be carried out. Often roads are maintained according to performance-based contracts.
Such a contract may specify measurement every year or every third year of a) skidding
resistance, b) evenness, c) cracks and potholes, and d) rutting.
Contractors shall issue a 5 year guarantee and the durability of thin asphalt layers is 10 – 15
Danish experience with applying thin asphalt layers is generally excellent. For example,
UTLAC (TB k) has proved to be resistant to rutting and SMA types are perceived as very fine
solutions in many cases.
To obtain full advantage of a thin asphalt layer the underlying base course must be even.
The following research and development needs have been identified:
o UTLAC (TB k) types: increase their durability
o For all types:
- reduce the effects of acoustic ageing
- reduce tyre/road rolling resistance.
The future
Thin asphalt layers are already used extensively in Denmark and this is expected to continue.
Page 93 of 124
B.1.1 Ageing
It is generally accepted that TAL wearing courses are particularly exposed to ageing in road
pavements. This phenomenon can be explained by both a layer thickness below 30 mm (large
impact of the surface as compared to the total mass of the layer) and the high void content
typical of a large number of TAL (especially those characterized by a stony skeleton).
With respect to ageing, two major categories can be identified: asphalt mixes with low rather
low void content (typically below 7 %) and asphalt mixes with a rather high void content
(above 7 %). The ageing of the first family is limited to the very upper layer (5 mm) since the
exposure of the binder to oxygen in the air is restricted to the surface, even when the asphalt
mix is characterized by an open surface texture. The voids of such a mixture are not
connecting, so the bulk of the asphalt layer can be considered as closed and therefore
inaccessible to oxygen. In contrast with the first family, mixtures characterized by a rather
high void content (interconnecting pores) are exposed entirely to the action of oxygen. The
ageing is likely to proceed homogeneously throughout the whole layer. A typical example is
porous asphalt.
Generally speaking, ageing of the binder (both short and long term ageing) can be studied
either by evaluating the ageing of the binder (e.g. following the procedures as described in EN
12607 part 1 to 3, EN 15323 or EN14769) or by ageing compacted specimens (conditioning
in a heated oven). Short term ageing does occur during bitumen storage, mix production and
laying, while long term ageing occurs during the service life of a pavement.
With respect to TAL only a very limited number of studies are reported. The results of a
Polish study have been published [Judycki & Jaskula, 2003]. They stated that the ageing of
asphalt mixes (Marshall specimen) designed for thin wearing courses did not show any
significant differences between the mixes containing different bitumens (unmodified or
polymer modified). Both GAP graded asphalt concrete (GAC) and SMA indicated similar
ageing properties (by measuring resilient modulus of aged and unaged specimens).
The HAPAS certification procedure does mention in appendix A.4 a methodology for
accessing the ageing characteristics of thin surfacing systems [BBA, 2008]. The protocol
describes a method for measuring the ageing characteristics by determining fatigue properties
(using Indirect Tensile Fatigue) before and after long term oven ageing of cylindrical cores
(preferred option is to drill bore core samples from the road). It should be noted however, that
the method has yet to be proven and shown to be valid (establishing a relation between lab
simulation and ageing in the field). The method is therefore unsuitable for use in specifications. The PIB (Product Information Blad) used in the Netherlands for the certification of
DAD (“dunne asphalt deklagen” or thin surface asphalt courses) is quite similar to the
HAPAS system. It also recognizes the need for field validation of the ageing procedure and
therefore the results are considered as informative.
Page 94 of 124
B.1.2 Acoustical durability
Danish long time experiment on acoustical durability
There are not many measurement results published on the long time acoustical performance of
thin layers [Bendtsen et al, 2005]. The results of a Danish experiment have been published in
[Bendtsen et al, 2009]. The test sections on highway M10 near Solrød in Denmark were
established in 2004. Figure B.1 shows a photo from this site. The purpose was to test different
types of noise reducing thin open graded pavements on a motorway. Yearly SPB noise
measurements have now been conducted over a 5 year period; see Figure B.2 and B.3. There
are six test pavements, a dense graded pavement and five noise reducing thin layers.
Figure B.1: Thin layers have been tested since 2004 under heavy traffic load on highway
M10 at Solrød southwest of Copenhagen in Denmark.
The traffic volume at this six lane highway is around 90,000 vehicles per day. Table B.1 gives
an overview of the slope of the trend lines found at each of the six M10 pavements. For
passenger cars the DAC 11 reference pavement and the SMA 8 pavement have the lowest
noise level increase of 0.5 to 0.7 dB/year. For the more open graded pavements, the increase
varies between 0.8 and 1.3 dB/year. The increase for multi-axle trucks is generally much
lower. Figure 3.3 shows the surface of one of the TAL included in this set of trial pavements.
Page 95 of 124
DGAC11 - Multi axle
y = 0,72x + 80,82
LAmax [dB]
LAmax [dB]
DGAC11 - Passenger cars
y = 0,28x + 88,29
R = 0,76
R = 0,97
Age [years]
Age [years]
Figure B.2: Pass-by noise levels for passenger cars (left, reference speed 110 km/h) and for
multi-axle vehicles (right, reference speed 85 km/h) on the DAC 11 [Bendtsen et al, 2009].
UTLAC 8 - Multi axle
y = 1,06x + 78,78
R = 0,97
LAmax [dB]
LAmax [dB]
UTLAC8 - Passenger cars
y = 0,35x + 86,32
R = 0,58
Age [years]
Age [years]
Figure B.3: Pass-by noise levels for passenger cars (left, reference speed 110 km/h) and for
multi-axle vehicles (right, reference speed 85 km/h) on the UTLAC 8 [Bendtsen et al, 2009].
Table B.1: Average noise level increase per year for passenger cars and multi-axle vehicles
for the six test pavements on M10, using reference speed 110 km/h for passenger cars and 85
km/h for multi-axle vehicles [Bendtsen et al, 2009].
Passenger cars
Multi-axle vehicles
DAC 11
0.7 dB/year
0.3 dB/year
1.1 dB/year
0.4 dB/year
AC 8o
0.8 dB/year
0.1 dB/year
0.5 dB/year
0.2 dB/year
SMA 6+
0.9 dB/year
0.6 dB/year
SMA 8+
1.3 dB/year
0.7 dB/year
One task in the European project SILENCE was to look into the development with time of
pavement acoustic performance [Kragh, 2008]. Among the available data were French data
for several types of thin layer asphalt. Figure B.4 shows data from 17 sections of road with
very thin asphalt concrete (BBTM 6 Type 2). For each section the time history of passenger
Page 96 of 124
car pass-by noise levels are shown. All of these have been forced to origin at 0 dB and the
curves show the increase in noise level with increasing pavement age. 0 dB was defined as the
intercept of the linear regression of noise level on pavement age. The figure also shows the
regression line for all the individual data points. The spread of observations is large. The
average slope is 0.78 dB per year and the standard deviation s R of the residuals in the ydirection is 1.9 dB.
Similar figures can be found in [Kragh, 2008] for several other types of pavement. Table B.2
summarizes the results for thin layer asphalt pavement. The table states the average slope, the
standard deviation s R , the number N of road of sections and the maximum pavement age in
the database.
AFmax_90 [dB]
BBTM 6 Type 2
y = 0.78x
sR = 1.9 dB
Age [yrs]
Figure B.4: Passenger car pass-by noise level at each of 17 sections of road with BBTM 6
Type 2 in the French database; see further the text. From [Kragh, 2008].
Table B.2:
data base
Summary of ageing performance for thin layer asphalt derived from the LCPC
Pavement v [km/h] Slope sR [dB/yr] [dB]
N [‐] Max age [yrs] Very Thin Asphalt Concrete 0/6‐type 1
0.25 1.6 11 11 Very Thin Asphalt Concrete 0/6‐type 2
90 0.78 1.9 17 8 Very Thin Asphalt Concrete 10‐type 2 0.42 1.6 7 8 Ultra Thin Asphalt Concrete 0/6 0.58 0.6 2 8 Page 97 of 124
The following conclusions for highways were drawn in [Bendtsen et al, 2009]:
The noise level on asphalt pavements normally increases with time
The increase occurs continuously and before significant pavement deterioration with
ravelling and cracks etc. begins
A linear regression gives a good fit of the relation between pavement age and noise
both for passenger cars and multi-axle heavy vehicles. This was also seen in the European SILENCE study
The yearly increase in noise level is generally around twice as high for passenger cars
as for heavy vehicles.
Different parameters have been used to characterize the increase in noise levels which is often
expressed as dB per year. Two main factors affect the changes in pavement noise properties.
One relates to the physical/chemical changes in the materials caused by weather elements and
time, and the other has to do with the wear and tear caused by traffic. It can be argued
[Bendtsen et al, 2009] that the combined effects of both the physical age of a pavement and
the wear and tear from traffic determine the increase in noise levels. The age reflects an
accumulated effect of changing weather conditions like sun radiation, rain, ice, freeze/thaw,
oxidation, etc.
Norwegian long time experiment on acoustical durability
A research and development project named ‘Environmentally friendly pavements’ has been
conducted under the auspices of the Norwegian Public Roads Administration and in close
cooperation with research institutions and the road industry. The project focused on the noise
and dust properties of road surfaces. The project was started in 2004 and was completed in
2008; however, some measurements have continued through 2010.
A number of different types of thin layer pavements were tested as part of this project, all of
which were dense and two of which were proprietary. NMAS was either 6 or 8 mm.
The Editor has processed data found in [Aksnes et al, 2010] and plotted them as "noise
reductions" versus the age of the surfaces in Figure B.5. The data were collected with the
CPX method using various reference tyres specified for that method during various years
(Tyre A or Tyre SRTT). The tyres were of different ages and, although the Norwegians
attempted as well as possible to account for possible changes from year to year, there is
unavoidably an uncertainty due to the stability of the reference. The reference level is an
average level measured on various SMA11 surfaces during various years.
The Figure shows that the deterioration of noise reduction is fast. Half of the initially high
noise reduction is lost after only one year and after three years approx one dB of noise
reduction remains. Note that all these roads are exposed to wear of studded winter tyres
during the winter season. These data should be fairly representative also of Swedish and
Finnish conditions. What remains to study now is how the structural condition and rut depths
develop over time for these thin layers.
Page 98 of 124
Noise reduction vs ref level [dB(A)]
Ref level = 0 dB = Mix of SMA11 surfaces 1 year or older
SMA6 50 km/h E18
SMA6 80 km/h E18
SMA8 50 km/h E18
SMA8 80 km/h E18
SMA8 50 km/h E6
SMA8 80 km/h E6
T8s 50 km/h RV20
ViaStab8 50 km/h RV118
Year of measurement after laying
Figure B.5: Noise reduction versus age of Norwegian TAL compared to a reference level of
SMA surfaces one year or older. Data processed by the Editor from [Aksnes et al, 2010].
Acoustical durability of Dutch thin layers
Figure B.6 shows the development of noise levels of Dutch TAL over time, compared to twolayer porous asphalt and special low-noise paving blocks [Groenendijk, 2011].
Figure B.6: Noise
level changes with
age of surfaces in the
Netherlands. From
[Groenendijk, 2011].
Page 99 of 124
B.1.3 Wear by traffic
TAL characterized by a low void content usually show good to excellent resistance to wear by
traffic, as long as the bonding with the adjacent underlying layer is assured, and therefore no
delamination occurs (see B.2.5.1 for discussion on tack coats). Typical examples include
asphalt concrete or SMA.
However, TAL as designed according to EN 13108-2, or porous asphalt mixes, are very much
prone to wear by traffic due to their high void content and therefore open structure. It is
generally recognized that these types of asphalt mixes are not suitable for paving at locations
such as intersections, roundabouts, locations with turning movements (e.g. parking lots, bus
stops, etc), sharp curves or other adverse geometric sections. In the latter cases, surface
damage can occur quickly due to the high tangential forces that may occur. Such forces cause
the loss of aggregates at the surface resulting in ravelling. It should be noted however that
ravelling is a complex phenomenon and other parameters such as for example the ageing of
the binder, the stripping of the binder from the aggregate (loss of adhesion) or the exposure to
low temperatures during winter may contribute as well.
Finally, too high tangential forces may even cause the displacement of asphalt material at the
surface, especially at sections with slowing traffic.
Several test methods, some of them still under development and undergoing field validation,
reflect the importance of the impact of traffic on TAL. Illustrative in this context are:
The Cantabro test as described in EN 12697-17 used to evaluate the cohesive strength of
porous asphalt [VBW-asfalt, 2004] or Open Graded Friction Courses (OGFC) [Sridhar et
al, 2005].
The tribometer or T2R as developed at LCPC in France [Hammoun et al, 2008]
The Rotating Surface Abrasion Test or RSAT designed by Heijmans in the Netherlands
[Hartjes et al, 2008].
The Aachener Rafeling Tester (ARTe) developed by the ‘Institut fur Strassenwesen –
RWTH" in Aachen [Schulze et al, 2008]
Danish project HOLDA
Porous pavements are normally considered to have a shorter lifetime than dense pavements.
One of the reasons for this is the very open pavement surface structure where the binder is
exposed to oxidation, causing the binder to become harder as time goes by. When the binder
becomes harder the risk of ravelling increases. This ageing phenomenon might also apply to
TAL pavements designed with very open surface structures in order to reduce splash and
spray, as well as noise. The Danish Road Institute has together with Danish Asphalt Industries
and NCC Roads carried out a research project "HOLDA" on optimization of the life-time of
porous pavements [Nielsen et al, 2004; Nielsen et al, 2005].
The aim of this project was to design a porous pavement, which has better durability than a
reference pavement built in the Øster Søgade experiment in Copenhagen [Bendtsen & Elle-
Page 100 of 124
bjerg, 2009], evaluated only by tests of laboratory produced materials. The research was
organised in two parts; designing and testing asphalt mixes and optimizing the bituminous
mortar. An analysis of variance of the particle loss data from the Cantabro tests was
performed. It is assumed that the Cantabro particle loss represents the durability of the mix
but it should be borne in mind that the durability of the actual pavement could be different.
The results indicated, as expected, that durability is improved if the voids content in the mix is
slightly reduced. However, what is more interesting is that the highest durability was obtained
when a highly modified SBS binder developed from a soft virgin binder was used. Therefore,
this new binder was estimated to be better than the highly modified binders used earlier in
various projects. Although field tests are required for validation, it was concluded that an
estimated improvement of the durability of 1.5-2 years could be achieved [Nielsen et al, 2004;
Nielsen et al, 2005]. It is not known what possible sacrifices might be needed regarding noise
reduction. These results might also be relevant when designing porous or semi-porous TAL
pavements, as well as when designing TAL pavements for roads with heavy traffic and maybe
also for roads where studded tyres are used.
B.1.4 Typical lifetimes
As mentioned previously, TAL pavements are applied because of their functional properties
such as noise reduction and/or the beneficial economics of the system (savings by thinner
layer thickness as compared to more traditional approaches). TAL pavements, due to their
limited thickness are not supposed to contribute significantly to the dimensional stability of
any road construction.
In order to achieve the majority of its functional properties, most TAL are characterized by
rather high (mixes as defined by EN13108-2) to even very high void content (e.g. porous
asphalt). Consequently, it is generally known and accepted that the average lifetime of such
asphalt mixtures is rather limited. Typically, TAL are supposed to last for about 10 years
although functional properties may be compromised sooner (such as noise reduction).
In Denmark the lifetime of TAL is estimated to be one year shorter than the lifetime of
standard DAC 11 or SMA 11, i.e. approximately 12 years. The European average durability
of UTLAC and AC (25-30 mm) is 10 years [EAPA, 2007a].
For Swedish conditions, which logically should be fairly similar to Norwegian and Finnish
conditions, an attempt to estimate typical service life times for five major pavement types is
presented in Table B.3. It must be noted that for TAL in Sweden the experience so far covers
only a few years and not so many types of TAL. It is also difficult to define what service life
means; it depends on the type of damage, etc. Except for traffic volume (AADT), the service
life will also depend on speed, but no distinction for speed has been made in the table.
Consequently, these figures should be considered as very rough, preliminary and based on
expert judgement rather than comprehensive statistics. This will be further dealt with in the
later part of this project.
Page 101 of 124
Table B.3: Rough estimate of service life times (in years) for five major pavement types in
Sweden and as a function of traffic volume (AADT). Times (values) in red colour shall be
considered with extra caution as they are expert judgement by one of the co-authors (RK),
while the commonly used black values are expert judgements by a number of experts at the
Pavement type (English) PAC SMA AC TAL Soft asphalt Pavement AADT AADT
type (Swedish) ≤ 250 750 ABD 14 12 ABS 15 15 ABT 15 17
TSK 14
MJOG 14 16
1500 10 15 12 AADT
3000 8 13 10 AADT
5000 7 11 9 AADT AADT
AADT 9000 13000 >13000
6 4 3
9 7 5
6 4 2.5
11 8 5 2 1 3.5
B.1.5 Possibilities of restoring a deteriorated TAL
As a part of the pavement maintenance restoring of deteriorated TAL by repairing the
pavements must be executed with suitable methods. The purpose of the repairing is to extend
the lifetime of the pavement, maintain a smooth and comfortable surface and prevent water
from penetrating.
Cracks and potholes are repaired by crack sealing or crack filling and potholes by pothole
patching respectively.
Larger areas with many or bigger cracks and areas with many and bigger potholes are repaired
by full-depth patching. In this method the deteriorated TAL is removed by milling and
replaced with new asphalt concrete of the same or similar type. Often the asphalt instead of
with a paver is laid by hand in selected areas. The method is the same as for other types of
asphalt pavement with the exception of the layer thickness. If the temperature is low or in
windy weather the thin layer of asphalt may cool down and the quality of the compaction may
be insufficient.
After repair of a TAL the surface has construction joints, areas with different types of asphalt,
areas with different thickness and areas paved by hand work. Especially for TAL of low noise
types repairs may cause the noise level to increase.
Page 102 of 124
B.2.1 Other surface characteristics
B.2.1.1 Megatexture and evenness
Most thin asphalt layers are able to compensate variations in layer thickness up to ± 5 mm.
Some rather exceptional thin asphalt layers allow variations up to ± 10 mm. Therefore the
under layer should be smooth and only small unevenness may be allowed. This is especially
the case for very open asphalt layers. The under layer should be watertight. This may also be
achieved with a layer of bitumen under the thin layer. A larger transversal angle is usually
applied for very open thin asphalt layers because of the flow resistance of water. An angle of
2.5 % should be enough. Before installation, a thorough check of the under layer is
recommended and if necessary measures should be taken, such as filling up holes, improving
transversal angle, and grinding the base course to remove unevenness [VBW, 2004] .
Most thin asphalt layers show a high resistance against rutting. When the layer is not well
compacted or when it is applied too thick, it becomes more sensitive to rutting. However
because of the limited thickness of the layer also the depth of the track remains restricted.
Rutting may occur easily in under layers when the thin layer is not able to spread out the load
of the traffic. Therefore under layers should have enough resistance against deformation
[VBW, 2004].
In the Netherlands four different test sections with thin layers have been installed. Texture
measurements were performed on six positions of each test section. Texture spectra of thin
layers compared to SMA 0/8 are shown in Figure B.7. These monitored thin layers show less
megatexture than SMA. Smaller aggregates imply less megatexture. A thin layer 2/6 12 % has
more megatexture than a thin layer 2/6 8 % [Schwanen, 2007].
Figure B.7: Texture spectra of thin layers compared to SMA 28 .
Warning: Please consider the wavelength range to the right of 8 mm as most uncertain, as some problems with
the texture measurements have been detected afterwards / The Editor
Page 103 of 124
B.2.1.2 Light and glare properties
In urban areas and on major roads entering urban areas from the countryside the road
administration has installed lamp posts to improve visibility under dark conditions. Tunnels
are another area where electrical light is used – typically in the form of sodium or mercury
light bulbs. It was very early recognised that the black surface of the asphalt pavement
absorbed a lot of the light demanding increased light intensity to compensate for this. This
would result in increased energy consumption. In Denmark, for instance, in order to reduce
consumption of electricity the road administrations asked for the addition of a certain
percentage of artificial light coloured aggregate. The artificial aggregate was heat treated flint
which became shiny white during the transition of the siliceous material in the process. In the
1960's and 1970's it became normal to demand that 20 - 30 % of the aggregate above 2 mm
size consisted of this white coloured aggregate. The change of the light reflective properties is
now obtained by the addition of white or light coloured aggregates to the asphalt mix during
production. The exact content of white aggregates depends on the light properties of the
ordinary aggregates in the asphalt mix. Typically the content is 20 – 30 % of white aggregates
in a size of the largest fraction used. That means if the asphalt material is DAC 8, then the size
of the white aggregates is 5/8 mm.
Due to the energy crises in 1972 and 1979 the original artificial aggregate under the trade
name of Synopal became uneconomical to produce and it was tried to make a similar product
at lower temperature but some of the properties were inferior to the original product. Test
equipment capable of measuring the light reflection of a moist, cut surface was developed in
order to avoid the increasing percentage of this more inferior artificial light coloured
aggregate and perhaps open the use of light grey natural aggregate. During this development it
was found that a general rule of thumb is: the lighter the colour the poorer the adhesion
properties are because the adhesion between white aggregates and bitumen is not at the same
level as for ordinary high quality aggregates and bitumen.
The introduction of PAMVLE – the light reflection measurement method – improved the
durability of the surface layer for the intended use on roads with lamp posts, because different
blends of natural aggregate of grey and light grey colour could be introduced and the
objective test method demonstrated that the desired light reflection was achieved.
The experience gained showed that only the fraction above 2 mm counted. There was
virtually no effect by introducing light coloured sand or finer materials. In order to get the
larger aggregate particles more exposed to the light open texture mixes were preferred.
Energy savings for lamp posts can be achieved for urban areas, tunnels and major roads if
light coloured road surfaces are used. From a psychological point of view many drivers
complain about pitch dark road surfaces when driving at night on roads with no lamp posts.
The conclusion of this is that open textured TAL with light coloured aggregates are the
optimum choice in these cases. The open texture maximizes the area that reflects the light
(either from the lamp posts or from the vehicle itself) and the rough texture minimizes the
glare from the vehicles you meet.
From a durability point of view you introduce only the light coloured aggregate in the thin
surface layer and not in a more vital thickness of the pavement structure if the aggregate
Page 104 of 124
reveals less than optimum adhesion properties. With a content of white aggregates up to 30 %
in combination with the addition of a high quality anti stripping agent the lifetime of asphalt
with light reflection properties is at the same level as for ordinary asphalt pavements.
For TAL the experience with light reflectance and lifetime is the same. But light properties
are only incorporated in TAL when the authorities require selected specifications to the light
reflectance on sections with and without street lightning because on voluntary base the asphalt
contractors will like to avoid the additional risk to the durability of the pavement.
B.2.1.3 Drainage, splash and spray
Generation of splash and spray and how it relates to road characteristics
As stated in the review of TAL history, the reduction of splash and spray from traffic
travelling at high speeds on wet roads has been used as an argument for using TAL. As will
be discussed in the following, it is arguable whether TAL are better than conventional asphalt
pavements in this respect.
In rather general terms, the road characteristics which influence splash and spray emission are
the following:
Amount of water standing on the surface: The deeper water, the more must be moved away
from the tyre/road contact patch. A portion of the water will stick to the tyre rubber and in its
tread pattern and be pulled up and then by centrifugal forces be thrown out from the tyre and
caught in the wind wake around and after the vehicle. This causes what we call spray: fine
droplets easily caught by the wind. But there is a limit as to how much this can be; the
remaining water will be pressed away from the contact patch and form what we call splash: a
plume of water ejected out towards the side of the vehicle.
Crossfall of the surface: It is common that roads are designed with crossfalls (transversal
grade) of about 2 %, but the higher it is the faster the water runs off to the roadside and the
lower the water depth.
Depth of ruts: The amount of water may be increased substantially in ruts.
Macrotexture: Low macrotexture means that all water on the road surface must be displaced
by the tyre; higher macrotexture means that some water may remain within the macrotexture.
Porosity (voids) of the pavement: If the pavement has high voids content, with connected
voids, a certain volume of water may be accumulated in the voids, until the voids are totally
filled. The porosity also means that some of the water may run-off to the roadside through the
pavement layer. During rainfall, it may take quite a while until the surface will start looking
wet, and a pavement with high voids content may never appear to create a "water mirror" on
the surface. It should however, be noted that there is some spray also from a highly porous
surface looking dry in wet weather, as the air pressed away at the leading edge of the
tyre/road contact patch will press up some water to the side, and under-pressure at the trailing
edge will pull up water from the pores. This water will create spray; albeit not at all as intense
as on a dense surface.
Page 105 of 124
When rainfall starts, it will take some time until the road surface will look wet. Some water is
absorbed by the surface micropores, some will be turned into water vapour due to the asphalt
often being quite a lot warmer than the air. If the pavement has a significant voids content it
will take some time until the voids are filled with water, in the best case an infinite time. This
time may be counted in several minutes or in hours. During this time a porous or semi-porous
surface may look essentially dry; often for such a long time that the rainfall may finish before
the pavement is saturated with water. After rainfall has finished the water in the voids will
remain there for some time, which may be counted in hours or even in days; cooling the
surface while water vapours. This is actually used in hot climates to reduce the so-called heat
island effect.
Splash and spray-related features of thin asphalt layers
It should be clear that all the influencing features mentioned above may be the same on TAL
pavements as on conventional and thicker AC or SMA. The features may also vary, both for
worse and for better, in relation to the thicker asphalt pavements. It is, therefore, not justified
to say that TAL in general are better or worse in terms of splash and spray. However, there
are some issues worth discussing:
By definition TAL are generally thinner than "conventional" SMA or AC. Assume that we
have a TAL which is porous and an AC which is porous, and consider the following cases (it
is assumed in all cases that the voids are connected):
Porous AC, 50 mm thick, 20 % air voids: This will store up to (effectively) an average
10 mm of rainfall (if water runoff to the side is neglected), as 50 mm with 20 % voids
is an equal volume of air as the volume of 10 mm of standing water (50x0.20)
Thin asphalt layer, 25 mm thick, 20 % of voids: This will store only half the water
volume of the previous case; i.e. 5 mm of standing water or rainfall (again, if water
runoff to the side is neglected)
Thin asphalt layer, 20 mm thick, 15 % of voids: This will store only 3 mm of standing
water or rainfall (water runoff to the side is neglected).
However, even in the latter case, to take care of 3 mm of rainfall before giving a wet surface
is a feature which will be appreciated by travellers in many cases. Depending on whether a
semi-porous TAL is compared to a dense or a porous AC, it will come out better or worse.
Nevertheless, one can say that the potential water storage property is better for a porous AC
than a porous TAL having been exposed to similar dirt volumes causing clogging of the
pores, just because of the different thickness.
In opposition to the advantage of porous AC over TAL as discussed above one can argue that
a semi-porous 25 mm TAL may be designed to provide almost the same durability as a 30-40
mm thick SMA, at least not being more expensive. In such a case it would be a competitive
alternative which would give the extra advantage of some water storage capacity, which will
reduce splash and spray.
Page 106 of 124
Apart from porosity and thickness, also NMAS (nominal maximum aggregate size) is in
general different for TAL than for conventional SMA or AC. NMAS is usually but not
consistently lower for TAL. But it is impossible to say that this makes any difference in splash
and spray generation.
Studies of splash and spray properties of TAL and other pavements
The authors first note that no measurement results of splash and spray on thin asphalt layers
have been found. Nevertheless, a few studies related to porous pavements versus dense
pavements shall be mentioned, since they provide some support for the general discussion
above about the effect of pavement porosity on splash and spray.
In 2005, DRI did a literature search on thin asphalt layers [Bendtsen et al, 2005]. With regard
to splash and spray DRI found, besides some documents discussing or presenting measurement methods, only two documents which provided some data or qualified observations
regarding comparison of pavements.
In the first study, test sections with porous pavements and SMA had been constructed on an
interstate highway in Indiana, USA and a measurement program covering noise, texture etc.
had been carried out [McDaniel & Thornton, 2005]. No quantified measurements of splash
and spray were conducted, but visual observations were made and it was concluded that the
splash and spray during one rainstorm event was considerably less on the porous pavement
than on the SMA pavement. Sight conditions for the driver were significantly improved, when
passing or being passed by trucks which typically produce large amounts of splash and spray.
A typical view comparing a porous and a dense asphalt is seen in Figure B.8. This picture is
not from any of the studies referenced here, but the view is rather typical.
Figure B.8: Typical view of splash and spray and pavement wetness on a porous compared to
a dense asphalt pavement. Picture used by permission from Mr Keizo Kamiya, Pavement
Division, Road Research Department, Central Expressway Research Institute, Tokyo, Japan.
Page 107 of 124
The second study presented a summary of results from measurements on splash and spray
carried out (starting in 1984) on test sections on A38 at Burton-on-Trent in the UK [Nicholls
& Daines, 1992]. Measurements were performed on 33 test sections with porous and nonporous pavements. The results showed that the splash and spray was reduced by 95 % on a
porous pavement compared to dense asphalt.
The variables which had been investigated in the study were the speed, rainfall, texture depth
and hydraulic conductivity (water outflow). The results showed that the splash and spray on
porous asphalt appeared to be very low over a wide range of hydraulic conductivity values;
even if the hydraulic conductivity was close to zero the splash and spray would still be only
half that on an equivalent HRA (hot rolled asphalt) surfacing.
A model was set up by [Nicholls & Daines, 1992] which can predict the splash and spray
generated on porous and hot rolled asphalt depending on the hydraulic conductivity, the
texture of the surfacing, the speed of the vehicles, the rainfall and the total recent rainfall. An
interesting observation mentioned was that splash and spray is carried over from non-porous
surfaces to porous surfaces by distances up to 100 m.
A model was also suggested for characterizing pavement splash and spray properties based on
an extensive recent literature study in which VTI took part [Resendez et al, 2007]. The
purpose was to propose how the splash and spray-reducing potential of various pavements
could be quantified in PMS or similar. The follow-up project, intending to develop this model
was given to another consortium; a work which is still ongoing.
Finally, a PhD study on splash and spray properties of various pavements is near finalization
in Bangkok 29). A first report of this work, indicating substantial advantages of the porous
asphalt, has been presented at a TRB meeting [Rungruangvirojn & Kanitpong, 2009].
B.2.2 Emission of inhalable particulate matter into the air
The effects of inhalable wear particle production (PM 10 ) from thin surface layers have not
been studied specifically. Instead, the main focus in pavement wear particle research has been
the influence of different stone materials and the influence of maximum stone size (NMAS).
Since thin surface layers normally have small NMAS, the influence of this parameter is likely
to be the most relevant. Still, little has been published internationally regarding this
relationship. Research at VTI, see Figure B.9, showed that SMA pavement with quartzite
(NMAS = 8, 11 and 16 mm) generates more PM 10 at lower NMAS at all three speeds tested
using studded tyres [Gustafsson et al, 2009]. The test with two mylonite SMA pavements
differed mainly at 70 km/h, also showing a higher PM 10 concentration at the lower NMAS
(8 mm). The results are in accordance with the hypothesis that stone properties that affect the
total pavement wear also affect the production of the inhalable fraction. From both laboratory
and field studies it is well known that lower NMAS results in increased pavement wear, see
e.g. [Jacobson & Wågberg, 2004].
Ulf Sandberg was a member of the Examination Committee for this PhD work
Page 108 of 124
PM10 (mg/m3)
PM10 (mg/m3)
Quartzite 1 SMA16
Quartzite 1 SMA11
Quartzite 1 SMA8
Mylonite SMA11
Mylonite SMA8
Road simulator speed (km/h)
Road simulator speed (km/h)
Figure B.9: Mean PM 10 concentration during stable concentrations at 30, 50 and 70 km/h for
three quartzite pavements (left) and two mylonite pavements (right) with different maximum
stone sizes. From [Gustafsson et al, 2009].
Regarding stone material choice, this is an important factor for wear particle production.
Especially obvious when using studded tyres, but notable also when using non-studded winter
tyres or summer tyres. In [Gustafsson et al, 2009] it was shown that two SMA pavements,
both with NMAS = 11 mm, differed in PM 10 production as a result of two different quartzite
materials. In the SPENS project [SPENS, 2009] it was shown that a limestone pavement’s
high calcium content contributed markedly to the elements of the coarser fraction of PM 10
while a diorite pavement gave a different composition with more influence of silica; see
Figure B.10.
Ongoing research at the Swedish National Road and Transport Research Institute (VTI) will
report more detailed results concerning influence of NMAS and stone material during 2010.
B.2.3 Structural strength and bearing capacity
In the design of flexible pavements the structural strength of the material layers is critical for
the bearing capacity.
UTLAC does not contribute to the bearing capacity. That means that the existing pavement
shall have a satisfactory evenness and bearing capacity.
For TAL between 20 and 30 mm the E-modulus is approximately 3 000 MPa using bitumen
40/60 or modified bitumen.
This means that before laying TAL - including UTLAC - it is decisive that all bearing
capacity is available in all elements of the existing pavement. Any existing structural damages
must be repaired before paving. The repair method is patching for levelling adjustment.
Frequently the damages are on a level where patching is insufficient. Then paving with TAL
is combined with milling operations in the same thickness as the new pavement.
Page 109 of 124
Diorite/summer tyre
Diorite/winter tyre
Limestone/winter tyre
Limestone/summer tyre
Figure B.10: Elemental analysis of different particle size fractions from diorite and limestone
pavements worn with non-studded winter tyres and summer tyres [SPENS, 2009].
B.2.4 Use of natural resources
Under normal circumstances the aggregate for the surface layer needs to be of better quality
than the aggregate of the bituminous binder and base course, due to the requirement as the
interface to traffic. The necessary bearing capacity for the road can be achieved by using nonpremium quality aggregates and perhaps even by upgrading inferior local aggregates.
Premium quality aggregate is seldom abundant and must be hauled in over longer distances.
By the use of TAL the necessary amount of premium aggregate will be lower and lead to
savings in natural resources and energy consumption. This will lead to an overall saving in
natural resources of aggregates or at least a slower rate of exploitation.
In some countries, where U.K. is one, it is very difficult to get access to high-quality aggregates, so ways to reduce the need for this are of particular importance. Here, it is indeed
important to reduce the thickness of the wearing course which requires such high-quality
aggregate. However, in the U.K. where hot rolled asphalt (HRA) has been very popular for
decades, HRA surfacings require even less of high-quality aggregate than a TAL, as this is
rolled into the mastic. This has caused the road administrations to partly return to the
traditional policy of using HRA at the expense of TAL [Lee, 2010].
See further Section B.3.4. The possibilities of recycling are discussed in Section B.3.3.
Page 110 of 124
Construction-related aspects
Paving equipment
With the exception of one type of asphalt surface layer TAL can be paved and compacted by
standard equipment. But that does not mean that it is unproblematic to construct a TAL of
good quality. The special case concerning UTLAC will be discussed separately.
As TAL has minimal thickness there is virtually no possibility for the material during normal
paving to compensate for insufficient transversal and longitudinal profiles as well as local pot
holes. All such deficits need to be solved by spot repair or even placing a levelling course.
Due to the small NMAS it will increase the risk of having permanent deformation (rutting) in
the surface layers if the layer thickness of the TAL locally goes above 4 times NMAS.
As explained in Annex B.2.3 it is important that all or nearly all necessary bearing capacity is
available in the underlying structure. For this reason the levelling course used to adjust the
necessary profile for the surface layer is usually performed as a strengthening layer.
Paving such thin layers of hot mix asphalt as defined by the term TAL requires full attention
to a lot of aspects. The heat content of the material itself is rather limited, when levelled off
by the screed of the paver the material quickly decreases its temperature due to heat
conduction to the underlying layer. After the screed the heat transfer to the air is very
sensitive to wind speed, so a windy day in September can have much quicker cooling effect
on the pavement than a day in November with no wind at all. The necessary roller pattern to
accomplish the desired degree of compaction has typically to be performed within 3 to 5
minutes. Warm Mix Asphalt is a concept where different technology or additives contribute to
lower both the mixing and the compaction temperature. This technique can be advantageous
to prolong the time window available for compaction. Whether or not this will prove to be
correct in the long term perspective is too early to say but the potential is there.
A recent development in paving equipment will be beneficial to preserve the heat content of
the surface layer for a longer than normal period. It is a possibility to use a double–layer
asphalt/twin paver which simultaneously paves both a strengthening/levelling course and a
thin surface layer. As the layers are placed “hot-on-hot” there will be no need for tack coating
in between. Due to the heat content of the lower layer the surface layer can be paved
extremely thin. The technology from Dynapac is called Compactasphalt® and can be seen in
Figure B.11. See also [Sandberg & Masuyama, 2005] and [Perveneckas & Vaitiuk, 2009].
A special paver is also needed for the paving operation for the material UTLAC. One of the
most important features of UTLAC is that the open graded hot mix material is paved onto a
surface which is sprayed with a polymer modified bitumen emulsion in approximately the
double amount of residual binder for normal tack coat. When the hot mix is applied the
emulsion has not broken yet and the remaining water boils up and facilitates a very intimate
contact between the substrate, the polymer modified binder from the emulsion and the new
surface layer. In order to apply this polymer modified emulsion under the paver it must have
spray bars close to the screws that distribute the hot material before the screed. Figure B.12
shows a schematic top view of the specially equipped paver for applying UTLAC. In this case
a paver from Vögele with the five spray bars (in red) for the polymer modified bitumen
emulsion and heating system for the emulsion tank is shown.
Page 111 of 124
Figure B.11: Compactasphalt® from Dynapac (www.compactasphalt.com)
Figure B.12: UTLAC paver from Vögele (www.voegele.info) (schematic top view; the five
spray bars for polymer modified bitumen emulsion are indicated in red)
When the concept of placing hot mix asphalt in a non-broken emulsion was introduced the
technique was tried out worldwide. Numerous reports exist especially from the period 19891991 highlighting initial trials probably due to the fact that professionals were sceptical
concerning the new concept. As the technique at that time was newly developed several
problems were indicated, bleeding from the spray nozzles, spraying operation continued even
after the paver had stopped, etc. These old reports (among others from several states in the
USA) are not quoted here. They have little relevance because pavers for UTLAC since then
have been improved a lot.
Laying time and traffic closure
Based on the Austrian experiences reported by Litzka the finisher should, if possible, not be
idling at any time during application [Litzka et al., 1994]. The working speed of the finisher
should be coordinated with the deliveries of the hot mix. Any stopping and restarting the
paver causes problems with “bumps” and smoothness. Therefore, logistics in general are very
Page 112 of 124
important and critical. The temperature of the material should lie between 150 and 170 °C
during application and between 120 and 150 °C during compaction. As thin layers cool down
very rapidly, the open structure does help in this respect, efficient compaction is possible only
for a few minutes (normally a minimum temperature of about 90 °C is required). Usually, it is
not possible to achieve compaction rates of more than 96 %, but the essential goal is to ensure
that the rollers press the layer onto the base in time to secure both good bonding and a good
internal structure within the layer. In this context, "Sabita’s guideline" indicates a minimum
compaction time window of 10 minutes necessary for practical reasons [Sabita guideline,
The Sabita’s guideline also illustrates quite well both the impact of the weather conditions on
the critical cooling rate of TAL (in this case a 25 mm layer). Attention is paid both to the air
and base course temperature as well as the wind velocity. The following example illustrates
quite well the impact of these parameters:
A 25 mm material paved in weather conditions of 13 °C air and 18 °C base temperature,
and a wind speed of 20 km/h, has a compaction window of only 7 minutes
Whereas at 30 °C air and 45 °C base temperature, and no wind, the compaction window is
14 minutes.
The Austrian experiences are backed up by similar findings reported in the USA dealing with
UTBWC applications [Hanson, 2001].
It is generally recognized that due to the high cooling rate, manual compaction does not result
in a satisfactory quality or performance of a TAL and should therefore be avoided whenever
possible. It is also advised to restrict the paving season to a period of April – October,
according to the Dutch guideline. In northern Europe and in the Alps this season must of
course be shorter.
Due to the high cooling rate, some TAL offer the opportunity to open the road for traffic
fairly quickly (after 30 – 90 minutes) following the paving operations. However, the newly
laid thin layer shall cool down to a temperature of 35 °C, or lower, before opening to traffic.
Nevertheless, in order to avoid any risk of compromising the long term durability of TAL, it
is not advised to consider such a practice as standard operations.
In Japan, recently, new research has been performed on ultra-thin asphalt layers in an attempt
to improve the ease of construction and traffic closure time for constructions in medium-tolow temperature range [Hatakeyama et al, 2010]. A special kind of lubricating oil, which
reacts chemically by water addition, was used which allows a good workability even at lowmedium temperatures (Figure B.13).
By spraying water from the roller at the compaction a chemical reaction is obtained between
oil and reaction agent that causes an early strengthening by increasing the binder viscosity.
Different amounts of lubricating oil, reaction agent and water were studied. Based on this
study specimens were made with the following conditions: 30 % special lubricating oil, 75 %
reaction agent, 4 % water, 60 °C at compaction and 7 curing days before tests. After all
promising laboratory results, this pavement with a thickness of 10 mm was installed with
success on a test section of 3 m width and 15 m length.
Page 113 of 124
Figure B.13: Ideal transition of binder viscosity according to [Hatakeyama et al, 2010].
These are the main findings of the research:
Good anti-skidding properties is provided by using the lubricating oil which implies
that rolling of the mix is possible around 60 °C
Viscosity of the binder is increased when spraying water at compaction, which makes
it possible to open traffic already after approx. 30 min. of curing time
Other properties and weather proofing are similar to that of dense-graded asphalt
Maintenance and rehabilitation
The purpose of the maintenance of an asphalt pavement is by periodic work – under normal
traffic and weather conditions – to keep the pavement as close as possible to its original
condition and to improve or extend the functional life or condition by the right repair at the
right time.
Conventionally the maintenance consists of the following three categories [Johnson, 2000]:
Preventive maintenance is the routine work to prevent deterioration of the TAL pavement.
The point is to improve or expand the functional life of the pavement by surface treatments
and operations to retard progressive failures and reduce the need for routine maintenance and
service activities. The methods for TAL are crack sealing, small pothole repair and fog
Corrective maintenance is performed after a deficiency occurs in the pavement. That means
the pavement needs repair for loss of fraction, moderate rutting, extensive cracking, potholes,
bleeding and ravelling. While preventive maintenance is performed when the pavement is still
in a good condition the corrective maintenance is performed when the pavement needs repair
and is more costly. The methods for TAL are crack filling, pothole repair, patching, full-deep
patching overlays, and milling in combination with overlays.
Emergency maintenance consists of the activities performed in an emergency situation such
as dangerous potholes on rods with high level of traffic. On TAL the repair methods are pot
holes repair, patching, full-deep patching overlays, and milling in combination with overlays.
Page 114 of 124
B.3.1 Sustainable construction
Thin asphalt layers can contribute to a sustainable road infrastructure. If a base course having
sufficient bearing capacity is established, a step is taken in the direction of what might be
denoted a “perpetual pavement”. TAL can be applied as an easy-to-replace wearing course
(the pavement "skin") on top of this perpetual pavement to service a variety of needs, such as
skid resistance, rolling resistance (see 7.4) and noise. Sustainability is obtained because this
perpetual pavement is easily maintained by renewing the TAL when requirements on one or
more of surface properties are no longer fulfilled.
B.3.2 Criteria for determining end-of-life
The condition of the asphalt pavement is determined according to ASTM D 5340, “Standard
Test Method for Airport Pavement Condition Index Surveys”. Even the standard is for
airports it is very useful for determining the condition of road pavements.
The test method covers the determination of the pavement condition through visual surveys of
the pavements using the Pavement Condition Index (PCI) method of quantifying pavement
The pavement of the road is divided into sections. The type and severity of the pavement
distress is assessed by visual inspection. Types of distress are alligator cracking, bleeding,
block cracking, polish aggregates, ravelling, rutting, etc. The total numbers of types of
distress are 16. The severity of distress is low, medium or high in a standardize way for each
of the types of distress.
The distress data are used to calculate the PCI which ranges from 0 to 100 with 0 being the
worst possible condition and 100 being the best possible condition.
If PCI < 25 the rating is that the condition of the pavement is very poor. If the PCI < 10 the
condition of the pavement is failed which means end-of-life of pavement.
The Pavement Condition Index is useful for all type of pavements including TAL.
The end-of-life of a pavement will not be reached as long as correct surface maintenance is
carried out [EAPA, 2007a].
B.3.3 Recycling properties
The possibility of recycling asphalt materials has been one of the large advantages of bituminous materials as opposed to the competition from the other road construction material Portland cement concrete.
With a very few exceptions old asphalt pavement can be recycled to 100 % and the capital
investment in the materials can be utilized – depending on age and constituents – also up to
Page 115 of 124
nearly 100 %. This aspect has of course a great impact of saving natural resources both on the
aggregate and the binder side.
The few exceptions that exist are linked to the use of materials in asphalt pavement in old
days before recycling became an issue. The situation today among the asphalt producers is
recognising the fact that recycling ability gives the industry a competitive edge towards the
cement concrete that must not be lost. This means that when new additives are offered to the
asphalt producers one of the first questions asked is: ”Will your product interfere with our
future ability to recycling the asphalt material in which it is used? If so forget it.” It might not
be possible to re-use the effect of the additive the second time around, but the minimum
requirement is then that it stays inert and shows no negative effect on the resulting product.
There are three groups of exceptions that to some extent either endanger or prohibit recycling
of which the first is the most predominant. The three exceptions are:
Coal tar and pitch based on coal tar (PAC/PAH). These products were earlier used as
binders or especially addition to bitumen due to their very good adhesion properties.
They impose a risk of carcinogenic nature for the workers if they are exposed to either
fumes or particulate matter. In some countries (like Denmark) the occurrence of tar and
pitch is so seldom that it is allowed to use small amounts of RA containing tar in hot mix
asphalt if Health-Safety and Environment (HSE) precautions (protective means) are taken
for the workers. In other countries (like Germany) recycling of these materials in hot mix
is not allowed but it is possible to recycle the material in cold products (like bitumen
emulsion based products). When tar compounds are present in asphalt pavement (like in
old base layers) they normally will not show any environmental hazard as their
hydrophobic nature will prevent leakage from the bituminous materials.
Asbestos fibres (either present as fibres or natural occurring in the aggregate used). Again
it is the carcinogenic risk that is the reason for the reservations. The origin of the fibres
could be from experimental use as stabilising fibres (the rumour has it that it had been
used in some of the very first SMA:s before cellulose fibres were introduced). Heat
resistant coating of the brakes in old cars may also contribute as another source of
asbestos fibres found in RA. The last source is natural occurrence in aggregates used for
asphalt materials.
Bituminous materials that for some reason more or less unknown have shown an
accelerated hardening profile to an extent that the asphalt producer will not dare to re-use
the material for new pavement. This can be associated to the presence of hardening
promoters like certain heavy metal atoms either naturally being present in the crude oil
(like Vanadium) or originating from catalysts used in the refinery for special processes.
Apart from these exceptions there is no obstacle in general in recycling asphalt materials into
new products.
But with respect to the present project of TAL it is important to highlight that there are two
sides to the coin:
Can TAL be recycled into new pavement materials?
Is it possible to use reclaimed asphalt in TAL?
Page 116 of 124
The first question is definitely answered by YES, but it is becoming virtually impossible to
recycle RA from TAL into a new TAL which answers the second question by NO. The reason
for these statements is a combination of a lot of technical, practical, economical reasons.
Under the seventh framework programme of the European Commission a four year project
called Re-Road is presently running (2009-2012) which will look into aspects of recycling
asphalt pavement and optimising the best use of RA and identifying the obstacles that
prevents higher percentages of RA in surface layers. Information is gathered from the asphalt
producers to illustrate the situation and TAL is an important segment of the road market
today. The analysis of the responses has been done yet, but some important points can be
extracted already:
The present thickness of many TAL is very small (perhaps 20 – 25 mm) compared to
the operational regime of the big milling machines, such as the machine shown in
Figure B.14. It is of course possible to mill off such small layer thicknesses. But often
no contamination from the aggregate from the layer beneath must occur in the milledoff material because their properties are unacceptable in a new surface layer. This
makes the milling process for selected materials almost impossible as the reason for
milling might be a worn and cracked surface layer to be replaced.
Figure B.14: Big milling machine in operation at Eaton Place, London, UK, removing the thin
surfacing, with a risk for the material to be contaminated by aggregate from the layer beneath.
Page 117 of 124
TAL has – as earlier mentioned – an NMAS of maximum 10 – 11 mm and even more
predominantly either 6 or 8 mm as nominal maximum aggregate size. This means that
the reclaimed material in order to be reused must be separated in a manner which will
crush the bituminous mortar between the aggregate and not the aggregate. Later this
material must be screened/sieved into smaller fractions like 0/6 mm in order to be
applicable for recycling into new TAL. Huge problems are associated with handling
0/6 mm fractions of RA because of caking due to the mortar rich material. This will
very easily result in lumps of RA in a new TAL which is totally unacceptable from a
quality point of view. For this reason he asphalt producers handling RA normally
produce a 0/16 mm fraction (and more seldom a 0/11 mm fraction) which makes it
impossible to reuse for TAL.
The particle size distribution of a 0/6 mm fraction of RA will also contain a relative
high portion of fines (filler and particle up to perhaps 1 mm) that it is extremely
difficult to utilize in the mix design of new TAL.
Perhaps the most important problem prohibiting the use of RA in TAL today is the
extremely high functionality built in the present range of TAL today. In order to
produce a noise-reducing open texture of an UTLAC, even special small size fractions
of several virgin aggregates need to be available within very tight limits of quality
The Re-Road will – when it is finished – hopefully reveal the best route to optimum utilisation of reclaimed asphalt but if a high level of functionality is required of the TAL the present
conclusion is that it has to be produced from virgin aggregate resources and that the recycling
of old TAL have to be part of either thicker surface layers or part of bituminous base and
binder courses and indirectly in that manner save natural resources of virgin materials.
B.3.4 Energy consumption and emissions during transportation
Research projects on future material supply have identified a problem with effects of transportation of materials on energy consumption and the environment. Consequently, there is
concern over the increasing transportation work to serve civil engineering construction
projects [Svedberg, 2010]. More and more material needs to be transported over longer and
longer distances and road construction is part of this problem. A recent European project has
studied this problem but does not especially address TAL [ECRPD, 2010].
Increased use of TAL may affect this problem in two opposite ways:
Less material needed in TAL will require less transportation
Higher-quality aggregate might need to be transported from quarries or other supplies
further away from the construction project
The problem applies to both construction and maintenance operations. The two features of
TAL listed above are evidently in conflict. It is difficult to say which one that will dominate;
future LCA will need to look closer at this.
Page 118 of 124
B.3.5 Potential effects of climate change
To reduce the impact of road transportation on climate change, essentially the CO 2 emissions,
all possibilities to reduce rolling resistance will have to be considered since it directly affects
energy consumption and thus CO 2 emissions. It is obvious that this calls for tyres with low
rolling resistance. However, it should be equally obvious that rolling resistance properties of
road surfaces should be considered [MIRIAM, 2011], [Cooee, 2011].
Present state-of-the-art regarding rolling resistance related to road surfaces shows that the
macrotexture of the surface, as represented by the Mean Profile Depth (MPD), is the most
influential parameter [Sandberg, 2011]. How does this relate to TAL as compared to other
common wearing courses?
The macrotexture of TAL may vary largely due to type of TAL, wear and NMAS. In general
TAL use smaller NMAS, typically in the range of 4 to 8 mm, and in general the MPD is lower
the lower the NMAS is. Figure B.6 is an illustration of this. This speaks in favour of TAL.
However, one of the concepts used when designing special TAL (such as proprietary ones) is
to create a relatively high level of macrotexture while still using a small NMAS, mainly by
using a gap-graded mix to obtain an open or even porous texture. Figure 7 in [Bendtsen &
Raaberg, 2005] about French TAL usage is an extreme example of this, where UTLAC has
mean texture depths of around 1.5 mm as compared to 0.5 mm for conventional dense
In conclusion one may say that TAL of type SMA6 and SMA8, or similar, will be desirable to
use in the future since they will reduce rolling resistance. However, proprietary ones
attempting to get the maximum possible macrotexture will not be favoured, unless they
replace surfacings with rough textures.
Page 119 of 124
CE-marked products according to the EN 13108 series
At present, it is compulsory to have TAL CE-marked if they are marketed as complying with
one of the product standards of the EN 13108 series (parts 1 to 7 as listed in Section 6.5.1). In
the near future, UTLAC applications will be covered by EN 13108-9 (the drafting of prEN
13108-9 is ongoing when this is written). It should be stressed that the latter standard only
specifies the asphalt mixes and not the final application on the road, as already discussed at
large in Section 3.2.2.
In contrast with the EN 13108 series, the ETAG 16 guideline (being drafted at present)
dealing with ultra thin layers will (in the future) take into account the entire process and
therefore includes the paving operations and the final application. This is due to the fact that
the paving operations for ultra thin layers are more critical in comparison with thicker asphalt
layers (>20 mm) in order to guarantee a good quality and durability of the application. Therefore, in the future, products complying with this guideline could be CE-marked.
Proprietary products
The CE mark on road construction products in accordance with EEC Directive M/124 has for
the product standards in the EN 13108-x series become mandatory from the start of 2010. The
impact on the market has still to be seen in daily practise of purchasing and contracting
asphalt materials. The approach may also be governed to a large extent for the coming years
by the attitude among the customers.
If the customers are public road administrations (either state, regional or community level) it
will be a question about their interpretation and willingness to abide to the public procurement
directive which direct to purchasers to (or limit them to) CE marked product – if available.
One of the loop holes concerning public procurement directive is: who is purchasing the
asphalt material (the loose mix on the lorry) that is CE-marked? The public road administration makes a contract for an asphalt pavement and that is Works (not covered by CE
marking or Eurocodes) and not a purchase according to a CEN product standard.
This has a big impact on proprietary products depending on the national attitude in the market
you operate. If the road administrations are strictly limited to purchase CE-marked products
(if available) you have as a company three options:
1. You can try to design and describe your proprietary product in one of the CE-marked
asphalt families in the EN 13108-x series. (A lot of proprietary products exist in the
material family for UTLAC which now is described in a draft product standard prEN
13108-9. This will eventually become a harmonized European product standard which
results in CE-marked products.
2. The European Organisation for technical Approval (EOTA) has for many years worked on
designing a guideline for UTLAC product which would enable them to be CE-marked
Page 120 of 124
through this route. The asphalt contractor can in this manner obtain a European Technical
Approval (ETA) which fulfils the essential requirements in the road construction
3. The third option is for the company to market their product and highlight that it covers a
technical functionality that it is not covered by the CE marking under either CEN or
EOTA. This would allow the public road administrations to purchase the product even if it
was not CE-marked, because the offered functionality was not covered. The success of
this approach would be very dependent of the national attitude/interpretation with regard
to public procurement directives.
In the first two options a third party inspection is involved in the initial type testing (ITT) and
the survey of the factory production control (FPC). In the last option you as a customer must
assure the quality by different approaches. It could be prequalification schemes which among
other look into ISO 9001 quality assurance schemes or similar at the asphalt contractor or
through making reference to appropriate national guideline for quality control or similar
accepted control schemes with regard to the specific job.
Noise-related classification
HAPAS in the United Kingdom
The following applies to the UK Highways Agency (HA) which is responsible for 4 % of the
UK road network (highways and trunk roads) carrying 30-40 % of the road traffic. For
information on “non-trunk roads” one would have to contact the County Surveyor Society or
many (i.e. thousands of) local road administrations.
For a contractor to build pavement on the national highway network he needs a noise label
certificate from the Highway Agency Product Approval Scheme (HAPAS).
The UK HA defines a noise reducing surfacing as one with a Road Surface Influence RSI ≤ 2.5 dB. RSI is defined in the HAPAS guidelines [BBA, 2008], see also Table C.1 and
Eq.(C.1) – Eq.(C.2). Noise testing must be made at two road sections with the same pavement
type. The noise level used to determine the RSI is a combined SPB noise level from light,
dual-axle heavy and multi-axle heavy vehicles. Such surfacing is denoted “Thin surface
course system (for highways)”, and it can be any surfacing as long as RSI ≤ -2.5 dB. The
producer must certify his product has an RSI ≤ -2.5 dB. There are at present 37 surface
products with a HAPAS thin surface certificate [HAPAS, 2011]; many of which also have
chosen the option to include a certification of noise. A HAPAS certificate generally has a 5
years lifetime.
The reference for comparison in the UK is a “new” (i.e. at least 12 months old) hot rolled
asphalt (HRA) with 20 mm nominal aggregate size and with an aimed mean texture depth
MTD = 1.5 mm (sand patch). The reference values were established in the 1970-1980s based
on average pass-by measurement results at many sites. Compared to the dense asphalt concrete reference pavements used in other European countries, the British reference is a rather
noisy reference pavement.
Page 121 of 124
The Road Surface Influence (RSI) for high (H) and medium (M) speed roads are:
RSI H = 10 log10 (7.8  10
L veh,L
RSI M = 10 log10 ( 11.8  10
+ 0.578  10
L veh,L
L veh,H1
+ 0.629  10
+ 10
L veh,H1
L veh,H2
Eq. (C.1)
) - 95.9
+ 0.157  10
L veh,H 2
) - 92.3
Eq. (C.2)
Table C.1: Reference speeds and reference noise levels for various categories of vehicles and
roads [Abbott, 2008].
Speed [km/h] / L veh [dB]
High speed
Medium speed
L veh,light
110 / 84.7 dB
80 / 81.1 dB
L veh,dual-axle
(L veh,H1 )
90 / 86.6 dB
70 / 83.8 dB
L veh,multi-axle
(L veh,H2 )
90 / 89.1 dB
70 / 86.6 dB
In principle, the full HAPAS certificate cannot be issued until it has been proved that 24
months after construction, the surface in fact still retains MTD ≥ 1.2 mm. In practice, the
noise level is certified shortly after 12 months.
The vehicle noise level L veh , used as a reference is 1 – 2 dB higher than the reference noise
level used e.g. in Denmark for classifying road surface noise reduction.
Concerning lifetime average noise performance, calls for tenders are based on the expectation
that HAPAS procedures will assure an average noise reduction as given by RSI (measured at
the at least 12 months old surface) multiplied by 0.7 [Highways Agency, 2006], limited to a
maximum of 3.5 dB. With the UK prediction method [CRTN, 1988] one can use this
correction in noise computations.
C road in the Netherlands
The reference pavement in the Dutch system for measuring and computing road traffic noise
levels [RMV, 2006] is dense asphalt concrete, most probably DAC 16, but the aggregate size
and the pavement age are not given explicitly in [RMV, 2006]. According to [van Vliet, 2007]
the reference at high speed roads is DAC 16 and at low speed roads the reference is a mix of
DAC 16 and DAC 11. The reference values at 7.5 m distance from the vehicle centre line, at a
height of 5 m above the road surface are given in Table C.2.
The road surface correction C road is the increase in noise emission as compared with that on
the reference surface. One may express this increase either in terms of an overall A-weighted
noise level or in terms of a correction for each octave-band with centre frequencies from
63 Hz to 8 kHz. An octave-band is a range of frequencies in which the highest frequency is
twice the lowest frequency.
Page 122 of 124
Table C.2: Reference values at 7.5 m distance at a height of 5 m [CROW, 2004].
Vehicle Category
Reference speed
L AFmax
Earlier, the road surface correction was included in the publication describing the prediction
method, but nowadays the Dutch organization CROW on its website publishes a list of correction factors and reports documenting the measurements behind them. Besides corrections for
twelve generic surfacings, the table contains corrections for 35 – 40 proprietary products 30).
For each of these products a test report can be downloaded from [CROW, 2010].
SRS system in Denmark
The Danish SRS system is based on CPX measurements according to [ISO/CD 11819-2]. It
was established in 2006 as a 1st generation system [Kragh, 2007b]. The principle is for the
contractor to build a test section with new pavement, measure CPX noise levels at a reference
speed of 50 km/h or 80 km/h, compare with defined reference noise levels, and declare a
pavement noise class; see Table C.3. Noise levels L A and L D are measured using the reference
tyres A and D defined in [ISO/CD 11819-2] with tyre A in the right side and tyre D in the left
side of the trailer. Based on these noise levels the index CPX DK is calculated as
CPX DK  0.85  ( LA 1)  0.15  LD  K
Eq. (C.3)
K is a trailer correction constant derived from a field calibration. In 2009 two trailers
participated and they were issued K = 0.0 and 0.1 dB, respectively.
Table C.3: First generation Danish SRS noise classes.
Noise class →
Class limit
97 – 99
95 – 97
< 95
89 – 91
87 – 89
< 87
In April 2011 it was 37
Page 123 of 124
Dutch COP testing system
For the Dutch main road network there is no system for testing COP of delivered surfacings.
But for the local road network, a special detailed procedure was in existence during the years
2001-2004. The testing was mainly based on CPX measurement. The idea was to measure
CPX along the delivered roadwork. From this, the average SPB level is estimated, based on a
CPX-SPB relation measured on the site or a relation known beforehand. Finally, the SPB
level is compared with the COP requirement.
A special law, the “Regulation for the Stimulation of the use of Low Noise Pavements”, was a
temporary initiative of the Dutch Ministry of Housing, Spatial Planning and Environment
(VROM). The technical background of the law was given in the background document
[VROM, 2002]. The local authority could tender and select a contractor according to its
normal procedures [VROM, 2002] but recommended that the contract should put the
responsibility for complying with the acoustic requirement on the contractor. The reason for
the strict COP procedures was that local road administrations could get a refund from the
Ministry for applying noise reducing surfacing and that the Ministry wanted to be certain that
its money was well spent.
CEN Noise Classification System
Civil servants of the European Commission are working on the development of European
standards on noise classification but it is not clear what work already has been done.
Therefore CEN/TC 227/WG 1 has created a Task Group « Noise Classification » which is
formed with members of CEN/TC 227/WG 1, WG 2, WG 3 and WG 5. The first meeting took
place 15 April 2010 in Berlin.
The task of the Task Group is described as follows: “to determine the needs and possibilities/practicalities for drafting European standard(s) for a system for the classification of noise
characteristics for surface courses and to determine the needs and possibilities/practicalities of
conformity assessment.”
A questionnaire will be sent around to CEN/TC 227 members to explore the needs and ideas
regarding a Noise Classification System. The outcome will be communicated in a report to all
CEN/TC 227 members. Based upon this report CEN/TC 227 will decide on the needs and
future approach of this Task Group.
It is not yet decided exactly what the work of the Task Group will comprise. Possible steps
1. Characterization: type approval testing using defined measuring methods and criteria
2. Conformity of production: validation of new surfaces using defined measuring
methods and criteria
3. Monitoring: validation of surfaces taking into account the time aspect using defined
methods and criteria
4. Classification
The questionnaire should make it more clear how far the Task Group should go. Probably the
work will comprise at least points 1 and 2.
Page 124 of 124
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF