Road Materials and Pavement Design

Vol. 13, No. 1, March 2012, 156–170

Crack growth behaviour in geosynthetic asphalt interlayer systems

E.K. Tschegga *, M. Jameka and R. Lugmayrb
a ViennaUniversity of Technology, Institute of Building Construction and Technology E206-4,
Karlsplatz 13, A-1040 Vienna, Austria; b TenCate Geosynthetics, Schachermayerstr. 18,
A-4021 Linz, Austria

Geosynthetic interlayers in asphalt systems have become a convenient technology for lifetime
prolongation in road construction engineering. Durability and lifetime prediction analysis of
such systems can be obtained by fatigue crack growth testing. In this paper, the fatigue crack
growth properties of three different asphalt interlayer system groups with different interlayer
functions: SAMI (=Stress Absorbing Membrane Interlayer), asphalt reinforcement and the
combination of SAMI + asphalt reinforcement are reported. For comparison non-interlayer
systems (reference) have also been tested. Instead of 3- or 4-point bending tests with beams,
wedge splitting tests with drill cores from the field are used for analyzing the fatigue crack
growth behaviour of these systems. Due to the temperature dependence of asphalt the tests have
been performed in a climate chamber at −10◦ C, 0◦ C and +10◦ C. Crack growth propagation
was determined visually. The results show that this new visual approach for evaluation of crack
growth testing together with the wedge splitting test are practicable, reproducible and allow
grading of geosynthetic asphalt interlayer systems.
Keywords: fatigue crack growth; geosynthetic interlayer; road lifetime; wedge splitting test

1. Introduction
Cracks in pavements cause massive maintenance costs for road authorities. The replacement of
asphalt layers is time consuming and may lead to problems in traffic infrastructure. Geosynthetic
interlayers are able to contribute to the strength of a pavement system. Products like geocom-
posites (SAMI + asphalt reinforcement) can lead to stress relief and reinforcement. The benefits
of geosynthetic interlayer systems are that they provide economical rehabilitation solutions and
3–4 times (Lugmayr et al 2009) delayed appearance of reflective cracking which results in long-
lasting road networks. On one hand, cracks are generated by overloads. This kind of crack can
be simulated by fracture tests (Tschegg et al., 1998). On the other hand, cracks originate, in the
long term, by fatigue and fatigue crack growth. Fatigue is one of the most frequent reasons for
cracks. Fatigue characterization of asphalt-aggregate mixtures and interfaces is an important but
difficult issue of material design in road construction. It is a major factor for accurate lifetime
prediction. Knowing the fatigue crack growth behaviour of asphalt and asphalt interfaces (in this
paper ‘asphalt interfaces’ stands for asphalt–interface–asphalt or asphalt–interface–interlayer)
could help to distinguish between long-lasting and short-lasting products. Overall, preventing
crack growth is the most efficient way in order to reduce maintenance costs. Therefore, materials
research in combination with structural design will be crucial for improving road lifetime.

*Corresponding author. Email: tschegg@ifp.tuwien.ac.at

ISSN 1468-0629 print/ISSN 2164-7402 online

© 2012 Taylor & Francis
http://dx.doi.org/10.1080/14680629.2011.644414
http://www.tandfonline.com
 Road Materials and Pavement Design 157

Several geosynthetic interlayer systems from many manufacturers worldwide are available,
ready to be applied in road constructions. Interlayer systems differ in properties as well as
in costs. A major task of this paper is to provide information on these various systems in
order to help road designing engineers to find the right product solution for their particular
purpose.
Known testing procedures to characterize fatigue crack growth behaviour of asphalt in fracture
mode I are the three-point- and four-point-bending tests. Drawbacks of the bending test on asphalt
are the big specimen size and the difficult handling because of heavy self-weight. Ramsamooj
(1980) stated that they supported the beams by an elastic solid during fatigue crack growth
measuring, to prevent them from breaking by their own weight. In addition, cutting beams directly
out of the road necessitates special equipment and much effort or specimen preparation in a
laboratory. Another disadvantage concerns the way of keeping environment properties constant
during fatigue crack growth testing. This can only be achieved by a climate chamber surrounding
the beam and the testing machine. These drawbacks make the bending test difficult and expensive
in practical use.
To study fatigue crack growth behaviour of asphalt and asphalt-interfaces, it is not enough
to know its properties at room temperature and above as in Ramsamooj (1991) between 16◦ C
and 35◦ C. At 35◦ C a huge healing effect has to be considered. Chabot et al. (2008) stated in
their work examining the pavement bound interface between cement concrete and asphalt that
the temperature is one of the most damageable factors for asphalt systems. In practice, asphalt
systems become sensitive to crack initiation and crack growth at lower temperatures – especially
in the range of −10◦ C to +10◦ C, where the properties of asphalt change dramatically from brittle
and tough (at −10◦ C) to ductile and weak (at +10◦ C). In Tschegg et al. (1995) and Tschegg
(1999) it is shown that no significant changes occur in fracture behaviour in asphalt below −10◦ C
and above +10◦ C.
A suitable method for determining fatigue life with compact specimens is desirable to overcome
the disadvantages of the bending test. For the indirect tensile test, drill cores are used as specimens,
but no reasonable fatigue crack growth values could be measured (Werkmeister & Wellner 2004).
The wedge splitting method according to Tschegg (1986) also uses compact specimens, but was
usually applied only for static fracture tests and not for fatigue crack growth tests. Tschegg et al.
(2010) was the first attempt to use the wedge splitting method successfully for fatigue crack
growth testing on asphalt and asphalt interfaces. In this work, the knowledge from Tschegg
et al. (2010) has been extended to asphalt interlayer systems with several different geosynthetic
interlayers. The specimen for the wedge splitting test are drill cores, which can either be taken
directly out of roads or produced in laboratories. The specimens are small enough to be handled
by one person only. Moreover, a climate chamber can be built in a very compact way around the
specimen.
Lifetime calculations on the basis of previously published calculation models (Francken et al.,
1996) lead to results that sometimes fit the practical observations (Lee et al., 2000) and sometimes
do not (Ramsamooj, 1999). As part of the international conference ‘Reflective Cracking of Pave-
ments in 1996’ in Maastricht (Francken et al., 1996) this situation was particularly highlighted
and mentioned as an unsolved problem in the ‘Concluding remarks’. At the 15th ASCE Engineer-
ing Mechanics Conference, Bodin et al. (2002) presented an elasticity-based nonlocal continuum
damage model for fatigue failure of asphalt concrete. They were able to predict the fatigue process
until macrocrack propagation for uniaxial conditions and found their model also in good corre-
lation with fatigue bending tests of trapezoidal cantilever samples. Chkir et al. (2009) described
an inverse analysis technique for the calibration of their fatigue model parameters by also testing
trapezoidal cantilever samples and applying successfully the improved model to asphalt samples
with different binders.
158 E.K. Tschegg et al.

1.1. Main aim of this research

This paper will show that fatigue crack growth in asphalt interlayer systems during fatigue testing
is possible by means of the wedge splitting test (Tschegg, 1986). Furthermore this paper will
prove that the wedge splitting method for fatigue crack growth is able to characterize asphalt
systems with different interlayers such as SAMI (= Stress Absorbing Membrane Interlayers),
asphalt reinforcement and a combined SAMI and asphalt reinforcement. An optical method for
analysing crack growth propagation rate was used by several authors e.g. Hartman and Gilchrist
(2004) and Tschegg et al. (2010) for determining the crack growth progress.

2. Wedge splitting method for fatigue crack growth tests

Initially developed for fracture tests (Tschegg, 1986), the wedge splitting test is now also used
in the laboratory of Material Sciences at the Vienna University of Technology for fatigue crack
growth testing of asphalt interfaces and interlayers (OeNORM B3592, 2011). The principle of
this method is shown in Figure 1(a). Besides shapes such as cylinders and beams (Tschegg, 1991)
(Figure 1(b)), rectangular solids have been chosen for this testing series. Cubes were cut out of
drilling cores, so that the interfaces between base- and cover-layers were in the centre of the
specimen. A symmetrical groove on top of the sample can either be prepared by cutting or, which
is more common, by gluing rectangular pieces of stone onto the top (see Figure 9 later). To assure
crack initiation at the interface, a so-called ‘starter notch’ has to be cut at the bottom of the groove.
From here, the crack starts to grow along the geosynthetic interlayer in a stable manner, driving
by a load.
A vertical compressive force, FM (see Figure 1(a)), generated by a push-testing machine is
transformed into a horizontal force FH by a wedge and transmitted to the groove of the specimen
by load transmission pieces. To minimize friction, roll bodies are used for this load transmission
process. Splitting of the sample is caused by the horizontal force component FH , which can be
calculated easily from the equation FH = FM /(2 tg(α/2)), where α is the angle of the wedge. If
slim wedges (α = 5–12◦ ) are used, the vertical force component FV is small in comparison with
the horizontal force FH and helps to constrain the crack to propagate within the area between
starter notch and linear support.

2.1. Evaluation of results

In fracture mechanics, the stress-intensity-factor concept is used to predict the stress state near
the tip of a fatigue crack caused by a load. It is applicable to homogeneous elastic materials.

Figure 1. (a) Wedge-splitting method according to Tschegg (1986, 1991); (b) possible specimen
geometries.
 Road Materials and Pavement Design 159

Figure 2. Dimensions of specimen cross-section for evaluation.

For 0◦ C and below the fracture mechanics behaviour of asphalt is usually almost linear elastic.
However, the assumptions of linear elastic fracture mechanics are not appropriate for asphalt at
temperatures above 0◦ C. In order to make experimental results comparable within a temperature
range of 0◦ C to +10◦ C, it makes sense to use the K-concept and not the J-integral for +10◦ C,
even if the process zone is large in relation to the crack length and the J-integral would fit better.
The magnitude of K depends on sample geometry, the size and location of the crack, the
magnitude, and the cyclic change of loads on the material. From Murakami (1987) the calculated
geometry factor for a compact tension specimen has been adapted to the wedge splitting specimen
which has more or less the same shape (equations (1)–(3)) (Figure 2).
FH
KI = √ · fI (α) (1)
t· W
a
α= (2)
W
(2 + α) · (0, 886 + 4, 64 · α − 13, 32 · α 2 + 14, 72 · α 3 − 5, 6 · α 4 )
fI (α) = √ (3)
( 1 − α)3
where
KI = Stress intensity factor [MPa m1/2 ]
FH = Splitting load [N]
t = Specimen thickness [m]
a = Distance of load application line to crack tip [m]
W = Distance of load application line to bottom of the sample [m]
fI = Geometry factor [dimensionless]
The results were evaluated according to equations (1) to (3) in the Paris-Law regime (Figure 3).

2.2. Determination of crack length

The major technical challenge to determine the fatigue crack progress is to measure the crack
length. From literature, procedures with ink (Liang and Zhou, 1997) are known, but they are
not suitable for permanent crack growth observation. The ink-method can only be applied when
fatigue testing is stopped and will not allow the continuation of testing afterwards.
Continuous measuring of the crack length without influencing the sample during the crack
growth process should be an aim. An electrical and an optical procedure were used in Tschegg
et al. (2010).
160 E.K. Tschegg et al.

Figure 3. Paris-Law (Paris et al., 1961).

Figure 4. Camera observing sample front.

Jacobs (1995) determined the crack length during fracture and fatigue also by using the electrical
crack length measuring method. A net of thin wires was applied to the specimen, and the length
of the crack was detected immediately as soon as the wires broke. The net of wires was destroyed
during each test and could not be used for further testing. Lugmayr et al. (2009) states, that an
electrical method for measuring crack growth length was tested but did not succeed, especially at
low temperatures. At −10◦ C the crack opening at the crack tip was too small to induce fracture
of the wires. A crack could be observed visually long before the electric circuit was influenced.
The optical method for crack length determination is accomplished by evaluating photos of
the specimen front, taken by a camera positioned outside the climate chamber (Figure 4) and
capturing the sample through the window of the chamber, as in Hartmann and Gilchrist (2004).
The sample front is covered by a white, highly reflective paint, which is additionally marked
with a grid-pattern. The pattern serves as a scale indicator on the photos for later evaluation.
During crack growth, the crack tip appears black and can be observed continuously by the high
black-white contrast to the white paint on the sample front. Photos are taken every minute and
saved on a computer. Each photo is stored together with a time stamp, which allows calculation
of crack growth rate at any time (Figure 5). Due to observing only the specimen front, scattering
of data points cannot be prevented, because neither the position of the crack tip at the back nor in
the inner parts of the sample can be determined with this method. Details of an image of the crack
line of a broken sample and the course of the crack path can be found in Lugmayr et al. (2009).

3. Materials, specimens, shape and size

In order to allow comparison of different geosynthetic interlayer systems, all other parameters had
to be kept constant. Therefore, a large-scale test field of 30 m × 3 m was prepared for specimen
creation. This test field was a driveway of a non-public gas station with heavy goods vehicle
 Road Materials and Pavement Design 161

Figure 5. Video image, taken during crack propagation (crack tip at about 2.5 cm).

traffic. The road was built and then used for 3 months. Then, drill cores were cut out of the road
for testing.

3.1. Asphalt
The base and cover-layer were made of bituminous material. The base layer was an AC22T
(B70/100) and the top layer an AC11D (B70/100). Using the polymer modified bitumen emul-
sion O65 K/PM-adhesive (OeNORM B 3503), geosynthetic interlayer systems from different
manufacturers were installed according to the manufacturers installation guidelines between base
and top layer.
Before installing the cover layer, the parameters of the asphalt surface of the base layer were
measured with a stereoscope: two photographs of the same surface area are taken at different
angles. The different light/shadow-contrasts of these photographs make it possible for computer
software to recalculate the original topography (Possler, 2009).
In the past, different milling groove distances have been used for milling asphalt roads before
installing the cover-layer. Tschegg et al. (2008) showed that the influence of the line width is
considerable and should be in the range of the aggregate diameter. This knowledge was already
included in designing the driveway for this study. The relevant parameters according to the EN
ISO 4287 standard for the milled road are listed in Table 1.

Table 1. Measured parameters (EN ISO 4287) of the milled base

layer surface before installing the cover-layer.

true area/projected area [1] 1.425

max. profile height [mm] 9.745
max. profile depth [mm] 11.69
total profile height [mm] 21.425
arithmetic average roughness Ra [mm] 0.545
maximum peak height Rp [mm] 1.8925
maximum valley depth Rv [mm] 2.175
maximum height of the profileRt [mm] 4.067
162 E.K. Tschegg et al.

Figure 6. SAMI-geosynthetic interlayer [scale in cm].

3.2. Geosynthetics interlayer systems

Each geosynthetic interlayer system belongs to one of these three major functional categories.

3.2.1. Interlayer 1: function = stress absorbing membrane interlayer (SAMI)

Geotextile, 140–150 g/m2 mechanically bonded nonwoven made from PP (Polypropylene) with
optimum bitumen retention capacity (Figure 6).

3.2.2. Interlayer 2: function = asphalt reinforcement

Geogrid, coated by PVC or latex with low bitumen content, made from glass fibres or polyester
yarns, with or without a light-weight calendared Geotextile fixed on the grid by gluing, just used
for the installation (Figure 7).

Figure 7. Asphalt reinforcement interlayer [scale in cm].

 Road Materials and Pavement Design 163

Figure 8. SAMI plus asphalt reinforcement interlayer [scale in cm].

Figure 9. Dimensions of wedge splitting specimens (typically: a = 20 mm, b = 30 mm, c = 5 mm,

HLig = 105 mm).

3.2.3. Interlayer 3: function = stress absorbing membrane interlayer plus asphalt

reinforcement
Geocomposite, combining the reinforcing properties of glass filaments with the bitumen retention
capacity of the mechanically bonded nonwoven polypropylene (Figure 8).
Besides these asphalt interlayer systems, reference specimens without any interlayer were also
prepared for comparison purposes.
The size and dimensions of the wedge splitting specimens can found in Figure 9. Cubes were
cut out of the drill cores with a diamond saw. Subsequently, a starter notch was made, using
a diamond saw. The last step of preparation was gluing two marble blocks onto the sample’s
top surface to obtain the groove for mounting the load transmission pieces (Figure 9). It is also
possible to cut the groove as shown in Figures 1(a) and 1(b).

4. Test condition and equipment

Testing was performed with a hydraulic fatigue testing machine made by Carl Schenck AG with a
load capacity of 40 kN. The sinusoidal loading frequency was 2 Hz. The maximum load amplitude
164 E.K. Tschegg et al.

has changed during each single test depending on the crack growth velocity. Usually, one test was
finished within 24 to 48 hours (depending on sample temperature), when the crack had reached
the bottom of the sample. The load amplitude sometimes had to be adapted during testing, either
to accelerate or to reduce crack growth velocity, starting at about 70% of the fracture strength of
the samples. The absolute maximum splitting-force-amplitude did not exceed 3000 N (at −10◦ C).
The frequency of 2 Hz gave the material enough time for relaxation and was chosen due to
pre-tests, which showed no changes in results at lower frequencies.
All tests were performed in a climate chamber attached to the testing machine with a temperature
control accuracy of ±0.5◦ C. The testing temperatures of the specimen were −10◦ C, 0◦ C and
+10◦ C. Prior to testing, the samples had been stored in a refrigerator for more than two days.
Three or sometimes two identical specimens of each asphalt interlayer system were tested at each
testing temperature in order to provide enough data for evaluation.
A black/white video camera from Mintron Enterprise Co., Ltd. and a PC TV card from Pin-
nacle Systems was used for photo recording. Inductive transducers made by Hottinger Baldwin
Messtechnik GmbH were mounted on the assembly to measure the crack mouth opening displace-
ment. All signals from the numerous temperature sensors, and force transducers were amplified
and read by a data acquisition card from National Instruments and the data acquisition soft-
ware DasyLab, which recorded all data and controlled the pre-storage chamber and the main
climate chamber.
For more information on the climate chamber, see Lugmayr et al. (2009), Tschegg et al. (2010).

5. Results and discussion

The results show that the wedge splitting test together with visual crack length determination
is a convenient method to characterize the fatigue crack growth behaviour of different asphalt-
geosynthetic interlayer systems at different temperatures of the specimen.

5.1. Fatigue crack growth behaviour of asphalt interlayer systems

For the same stress intensity, the crack growth rates for asphalt interfaces and asphalt-interlayer
systems at +10◦ C are higher than at 0◦ C and much higher than at −10◦ C. This fact could be
shown by the resulting Paris-law-diagrams in Figures 10 to 13. In all cases, sample temperatures

Figure 10. Fatigue crack growth of asphalt interfaces without interlayer.

 Road Materials and Pavement Design 165

Figure 11. Fatigue crack growth of asphalt interlayer systems with ‘SAMI’ function.

Figure 12. Fatigue crack growth of asphalt interlayer systems with ‘asphalt reinforcement’ function.

of +10◦ C lead to faster crack growth (higher da/dN ) at fixed stress intensity factors K. Also
at −10◦ C, a higher K is necessary to observe the same crack growth rate as at 0◦ C.
Whereas Figure 10 shows the fatigue crack growth behaviour of asphalt interfaces without
interlayer, Figure 11 shows asphalt interlayer systems with SAMI function. Obviously, the slopes
of the linear regression lines of fatigue crack growth curves at +10◦ C, 0◦ C and −10◦ C have
changed because of the interlayer type. This means that the crack growth velocity is retarded
by the use of this kind of interlayer compared with non-interlayer systems. The asphalt system
behaves in a more ‘ductile’ way with the SAMI interlayer.
At the end of fatigue crack growth testing (close to rupture), i.e. when it comes to fast crack
growth, the scattering of visual crack length detection is high. Therefore, some data points repre-
senting this fast crack growth in Figure 11 were not integrated into the linear regression. Likewise,
the crack growth velocity is very low (sometimes in a range of mm/day) and the error of the gen-
erated data points is, thus, also high at the beginning of fatigue crack growth testing, when the
process zone starts to grow (e.g. Figures 10 and 11, data point below da/dN = 1E–7). Both
phenomena can be found in almost all fatigue crack growth testing datasets.
166 E.K. Tschegg et al.

Figure 13. Fatigue crack growth of asphalt interlayer systems with ‘SAMI+asphalt reinforcement’
function.

Figure 12 shows fatigue crack growth results of interlayer-systems with an asphalt reinforce-
ment function.
Here, the relative position of the two linear regression lines (+10◦ C and 0◦ C) is completely
different from non-interlayer systems (Figure 10) and the SAMI interlayer systems (Figure 11).
The 0◦ C crack growth velocity is much higher for this interlayer type than for the others, which
means poor fatigue crack growth behaviour at 0◦ C. At −10◦ C, no stable crack growth could
be obtained. Asphalt reinforced interlayer systems at −10◦ C are too brittle and can thus not be
measured by the wedge splitting test. Even if the crack was initiated after 24 h of fatigue testing,
crack growth was unstable, and the sample broke within a few seconds.
Figure 13 shows the fatigue crack growth behaviour of the third asphalt interlayer system –
‘SAMI+asphalt reinforcement’. The results are close to the SAMI-only interlayer systems in
Figure 11, but the slopes of the linear regression lines at 0◦ C and −10◦ C are smaller than in
Figure 11, which indicates a slightly slower crack growth than for SAMI-only interlayer systems.
The regression lines of Figures 10 to 13 are plotted in Figure 14. In this diagram, the differ-
ent systems can be compared directly in view of fatigue crack growth behaviour: The asphalt
reinforcement interlayer performs worse than the SAMI and even worse than the non-interlayer
system at −10◦ , 0◦ and +10◦ C temperatures. At +10◦ C, the non-interlayer system has a better
crack growth behaviour than the SAMI+asphalt reinforcement interlayer. Best performer (for long
road lifetime) is the SAMI-interlayer system at +10◦ C; at 0◦ C and −10◦ C, the SAMI+asphalt
reinforced interlayers are the best when it comes to higher loads (= higher stress intensity factor
K). For lower K-values, the SAMI interlayer is better than all others at all temperatures.
Beyond these results, a lot of questions still remain open, for instance the reason for the scattering
of data points on the non-linear crack growth curve (Lugmayr et al., 2009). Thus, further research
will be necessary.

5.2. Bridging effect

The bridging effect originates from the topography of the fracture surface. If the surface is smooth,
no energy consuming interaction between the already broken parts will take place. In asphalt
interfaces, the aggregate from the top layer is pressed into the base layer by a road roller. As a
 Road Materials and Pavement Design 167

Figure 14. Comparison of fatigue crack growth properties of different asphalt interlayer systems at +10◦ C,
0◦ C, and −10◦ C.

Figure 15. (a) Bitumen and interlayer bridging in SAMI-interlayer systems; (b) bitumen bridging only in
asphalt reinforced interlayer systems.

result, the top and the base layer are interlocked up to a certain extent. In this case, crack growth
requires additional energy for either disconnecting the two layers or breaking the aggregate and
pulling the bitumen filaments out. This process occurs almost completely behind the micro crack-
or process-zone (Figure 15). SAMI-interlayers show two kinds of bridging – bitumen bridging
and fibre bridging, which are both caused by the soft and fleecy structure of the interlayers
(Figure 15(a)). Interlayers for asphalt reinforcement do not show fibre bridging; only bitumen
bridging occurs (Figure 15(b)). Especially at low temperatures, the fracture energy is influenced
by fibre bridging and not by the bitumen, because the bitumen becomes brittle but the fibres still
transmit forces. At +10◦ C, the bitumen is ductile and contributes to bridging.
168 E.K. Tschegg et al.

5.3. Healing of asphalt layers

Another (less explored) effect, which is important for fatigue crack growth analysis, is crack
healing. Cracks in asphalt, which appear in the winter/spring season, may heal during the summer
period. Healing is an inverse effect to crack growth. Although asphalt has a lower strength in
summer, and thus crack initiation is more frequent, crack healing is present in summer. In the
winter/spring season, the strength of asphalt is higher but crack healing is less frequent.
Using the healing performance on reflective cracks in asphalt layers (BTI16 B100 according
RVS 8S.05.14), the mechanical fracture behaviour (notch tensile strength and specific fracture
energy) was determined in Tschegg et al. (2000). In this study, the healing process was simulated
by performing one complete cycle, including loading at 0◦ C, subsequent healing for 4 hours at
35◦ C, and then a cooling phase back to 0◦ C.
The notch tensile strength (analogue to adhesive bond strength) did not change after several
healing actions. However, the specific fracture energy (crack resistance) was reduced after each
healing action (Tschegg et al., 2000). With regard to fatigue crack propagation, analysing and
evaluating the healing effect is essential.
Until now, only a few experimental (Wool, 1995; Little et al., 1997, 1999; Tschegg et al.,
2001, 2011), numerical (Tschegg et al., 2000) and theoretical (Wool, 1995) studies in the field
of healing are available. No numerical theory yet bears this effect in mind, which can bring back
up to three times more crack growth resistance than without temperature healing (Tschegg et al.,
2000), depending on the type of asphalt and the healing temperature. In future, at least at +10◦ C
and above, this effect must be taken into account for lifetime calculation.

6. Conclusion
Up to now, beams and bending tests have been used to obtain the fatigue crack growth behaviour
of asphalt-systems. In this paper, the wedge splitting method was applied for determining the
fatigue crack growth properties of asphalt reinforced geosynthetic interlayer systems at −10◦ C,
0◦ C, and +10◦ C. Crack growth was determined by a newly developed visual evaluation method.
This approach of fatigue crack growth test evaluation is sensitive enough to observe the different
features of crack growth behaviour of three different asphalt geosynthetic interlayer systems.
It could be observed that SAMI and SAMI+asphalt reinforcement asphalt interlayer systems
outperform asphalt reinforcement systems and non interlayer systems by far, especially at low
temperatures and high loads, when not only bitumen bridging but both bitumen bridging and
fibre bridging occurs. The latter increases the crack growth energy extensively and leads to crack
growth delay. Beyond all restrictions of evaluating the crack progress visually, the wedge splitting
method managed to give feasible results for fatigue crack growth analysis at different temperatures.
This testing method can provide important data for lifetime predictions of roads in an easy way
in the future.

Acknowledgement
We would like to thank Professor Stefanie Stanzl-Tschegg for discussions and reviewing of this study, and
the Oesterreichische Forschungsförderungsgesellschaft FFG (Austrian Research Promotion Agency) for
financial support.

References
Bodin, D., Pijaudier-Cabot, G., De La Roche, C., & Piau, J.-M. (2002). Continuum damage approach to
asphalt concrete fatigue modeling. Journal of Engineering Mechanics (ASCE), 130, 700–708.
 Road Materials and Pavement Design 169

Chabot, A., Balay, J.M., Pouteau, B., & De Larrard, F. (2008). FABAC accelerated loading test of bond
between cement overlay and asphalt layers. Sixth RILEM CP Conference, Chicago. In Taylor & Francis
Group Proceedings: 13–23.
Chkir, R., Bodin, D., Pijaudier-Cabot, G., Gauthier, G., & Gallet, T. (2009). An inverse analysis approach
to determine fatigue performance of bituminous mixes. Mechanics of Time Dependent Materials, 13,
357–373.
EN ISO 4287, standard. (1997). Geometrical Product Specifications (GPS) – Surface texture: Profile method –
Terms, definitions and surface texture parameters.
Francken, L., Beuving, E., & Molemaar, A.A.A. (1996). Reflecting cracking in pavements – design and
performance of overlay systems. Proceedings of the Third RILEM Conference, 2–4 October 1996,
E&FN SPON, Maastricht, Netherlands.
Hartman, A.M., & Gilchrist, M.D. (2004). Evaluating four-point bend fatigue of asphalt mix using image
analysis. Journal of Materials in Civil Engineering, ASCE, Jan/Feb, 60–68.
Jacobs, M.M.J. (1995). Crack growth in asphaltic mixes. Dissertation, Technical University Delft.
Lee, H-J., Daniel, J.S., & Kim, Y.R. (2000). Continuum damage mechanics-based fatigue model of asphalt
concrete. Journal of Materials in Civil Engineering, 12, 105–112.
Liang, R.Y., & Zhou, J. (1997). Prediction of fatigue life of asphalt concrete beams. International Journal
of Fatigue, 19, 117–124.
Little, D.N., Lytton, R.L., & Williams, D. (1997). Propagation and healing of micro cracks in asphalt concrete
and their contributions to fatigue. In A.M. Usmani (Ed.), Asphalt Science Technology (pp. 149–195).
New York: Marcel Dekker Inc.
Little, D.N., Lytton, R.L., Williams, D., & Kim, Y.R. (1999). An analysis of the mechanism of micro damage
healing based on the application of micromechanics, first principles of fracture and healing. Journal of
Association of Asphalt Paving Technologists, 68, 501–542
Lugmayr, R., Jamek, M., & Tschegg, E.K. (2009). Mechanism of fatigue crack propagation and fracture
behavior in bituminous roads. In A. Loizos, M.N. Partl, T. Scarpas & I.L. Al Qadi (Eds) Advanced
Testing and Characterization of Bituminous Materials (pp. 807–816). London: Taylor & Francis.
Murakami, Y. (Ed.). (1987). Stress Intensity Factors Handbook, Vol. 1. Oxford: Pergamon Press. ISBN-13:
978-0080348094.
OeNORM B 3592 (Austrian Standard). 2011. Determination of notch tensile strength and specific frac-
ture energy of construction materials, their compounds and composites – wedge splitting method,
(in German). Bestimmung der Kerbzugfestigkeit und der spezifischen Bruchenergie von Baustoffen,
Baustoffverbindungen und Verbundwerkstoffen – Keilspaltmethode. Austrian Standards Institute.
OeNORM B 3503 (Austrian Standard). 1992. Bitumen emulsion for road constructions, (in German).
Bitumenemulsionen für den Straßenbau - Anforderungen - Bitumen- und Polymerbitumenemulsionen
für Haftbrücken. Austrian Standards Institute.
Paris, P.C., Gomez, M.P., & Anderson, W.E. (1961). A rational analytic theory of fatigue. The Trend in
Engineering, 13, 9–14.
Possler, H. (2009). Surface roughness of construction materials. Masters thesis, in German, Vienna
University of Technology.
RVS 8S.05.14. (1986). Bituminous layers Österreichische Forschungsgesellschaft Straße –
Schiene – Verkehr, Bituminöse Tragschichten im Heißmischverfahren. Vienna: FVS. (In German.)
Ramsamooj, D.V. (1980). Fatigue cracking of asphalt pavements. Transportation Research Record, 756,
43–48.
Ramsamooj, D.V. (1991). Prediction of fatigue life of asphalt concrete beams from fracture tests. ASTM,
Journal of Testing and Evaluation, 19(3), 231–239.
Ramsamooj, D.V. (1999). Prediction of fatigue performance of asphalt concrete mixes. ASTM, Journal of
Testing and Evaluation, 27(5), 343–348.
Tschegg, E.K. (1986). Test method for the determination of fracture mechanics properties. Patent
Specification No. A-233/86 390 328, Austrian Patent Office. Prüfeinrichtung zur Ermittlung von
bruchmechanischen Kennwerten sowie hiefür geeignete Prüfkörper, Österreichisches Patent AT-390328
(in German).
Tschegg E.K. (1991). New equipment for fracture tests on concrete. Materials Testing, 33, 338–342.
Tschegg, E.K. (1999). Concept and parameters to characterize the fracture behaviour of bituminous materials
and layer bonds. EUROBITUME Workshop 99, Proceedings of Performance Related Properties for
Bituminous Binders, Kirchberg Luxembourg, Paper No.009, pp 1–5.
170 E.K. Tschegg et al.

Tschegg, E.K., Kroyer, G., Tan, D.-M., Litzka, J., & Stanzl-Tschegg, S. (1995). Characterizing of fracture
behavior of asphalt. Road Research No. 440, Federal Ministry of Economy, Vienna, Austria, Charak-
terisierung der Brucheigenschaften von Asphalt. Österreich. Bundesministerium für Wirtschaftliche
Angelegenheiten, Bundesstraßenverwaltung, Wien, Straßenforschung 440 (in German).
Tschegg, E.K., Ehart, R.J.A., & Ingruber, M.M. (1998). Fracture behavior of geosynthetic interlayers in road
pavements. ASCE Journal of Transportation Engineering, 124, 457–464.
Tschegg, E.K., Catharin, A., & Stanzl-Tschegg, S.E. (2000). Healing behaviour of reflective cracks.
Proceedings of the Fourth International RILEM Conference on Reflective Cracks in Pavements, 26–30
March, Ottawa, Canada., pp. 456–464.
Tschegg, E.K., Jamek, M., & Stanzl-Tschegg, S. (2001). Fracture and healing behaviour of asphalt
and asphaltinterfaces. Road Research No. 516, Federal Ministry for Transport, Innovation and
Technology, Bruch- und Selbstheilungsverhalten von Asphalt und Asphaltverbunden. Ausmaß und
Nutzen. Austria, Straßenforschung 516, Bundesministerium für Verkehr, Innovation und Technologie,
Bundesstraßenverwaltung, Wien, Österreich (in German).
Tschegg, E.K., Jamek, M., & Buchta, M. (2008). Mechanical and fracture-mechanical evaluation of interfaces
between milled and new asphalt layers. Road Research, 575; Straßenforschung, 575 (in German).
Tschegg, E.K., Jamek, M., & Lugmayr, R. (2011) Fatigue crack growth in asphalt and asphalt compounds
(by means of the wedge splitting method). Engineering Fracture Mechanics, 78, 1044–1054.
Werkmeister, M., & Wellner, F. (2004). Die Untersuchung der Ermüdungsverhalten von Asphaltgemischen
mit dem dynamischen Spaltzugversuch im Rahmen einer analytischen Bemessung, Bitumen, 3, 66
(in German).
Wool, R.P. (1995). Polymer Interfaces, Structure and Strength. Munich, Vienna, New York: Hanser Gardner
Publications. ISBN-13: 978-1569901335.