Académique Documents
Professionnel Documents
Culture Documents
2
3
TITLE:
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Authors:
Felipe Halles A., (* corresponding author)
Centro de Ingeniera e Investigacin Vial (CIIV)
Pontificia Universidad Catlica de Chile
Departamento de Ingeniera y Gestin de la Construccin. Escuela de Ingeniera.
Vicua Mackenna 4860, Edificio San Agustn 3 piso, Santiago - Chile
E.mail: fhalles@ing.puc.cl
Phone: 56-2-3544245; Fax: 56-2-3544806
Guillermo Thenoux Z.
Escuela de Ingeniera
Pontificia Universidad Catlica de Chile
Departamento de Ingeniera y Gestin de la Construccin. Escuela de Ingeniera.
Vicua Mackenna 4860, Edificio San Agustn 3 piso, Santiago - Chile
E.mail: gthenoux@ing.puc.cl
Phone: 56-2-3544245; Fax: 56-2-3544806
lvaro Gonzlez V.
Facultad de Ingeniera
Universidad del Desarrollo
Avenida La Plaza 680
Email: aagonzalez@ingenieros.udd.cl
Phone: 56-2-3279756; Fax: 56-2-9253
Submission date:
Length of text:
Number of figures and tables:
TOTAL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Abstract: From the literature it is possible to find two trends regarding the stiffness evolution of
the Foamed Bitumen Stabilized/Recycled mixtures. The first trend indicates that once the
foamed bitumen mix reaches a constant value due to curing process, the stiffness decreases with
time due to load cycles. The second trend indicates that stiffness remains constant after curing
process. In this research, the stiffness evolution of foamed bitumen mixes stabilized with
different bitumen and cement contents is studied. The stiffness was measured using the Indirect
Tensile Fatigue Test (ITFT). Results indicate that once the foamed bitumen mix reaches a
constant value due to curing process, the stiffness will decrease or keeps constant depending on
the stress-level applied to the foamed bitumen layer. If the stress-level is lower than a specific
value, the stiffness of the mix will remain constant at a value very close to the initial stiffness. If
the stress-level is greater than a specific value the stiffness of the mix will decrease gradually.
Also, the reduction rate of the stiffness will be greater as the stress-level is higher. The analysis
of results from mixes with different bitumen and cement contents allow identifying the effect of
both stabilizing agents in the long-term stiffness.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
GENERAL DISCUSSION
Introduction
Stabilizing or recycling of asphalt concrete pavements using Foamed Bitumen (FB) is a widely
used rehabilitation technique worldwide. However, there is still no structural design
methodology fully validated by practitioners and researchers. One reason lies in the fact that it
has not been possible to determine the elastic modulus that FB stabilized layer develops with
time. Without a reliable modulus or stiffness value, the structural analysis of a FB pavement is
uncertain. Although there are several laboratory tests for assessing the stiffness of FB mixtures,
none of them allows identifying the observed stiffness evolution with time that has been
observed in the field.
There are two trends found in the literature regarding the in-situ stiffness evolution of the
FB mix. The first comes from studies carried out by Loizos (1) who analyzed back-calculated
stiffness from Falling Weight Deflectometer (FWD) tests from a FB recycled project built in
Greece (Figure 1). Results indicated that stiffness gradually increases from the day of
construction until it reaches a constant value after a period of approximately 12 months. A fact
that explains this behavior is the loss of moisture from the mixture that occurs during the curing
period. The loss of moisture and increase in strength or stiffness has also been observed by
Bowering (2), Jones et al. (3) and Fu et al. (4), among others.
After 12 months the backcalculated stiffness remains constant despite constant traffic of
heavy-weight vehicles (1).
32
33
34
35
36
37
38
39
40
41
42
Conversely, research studies carried out in South Africa (5) indicated that FB layers
affected to traffic load show a gradual decrease in stiffness. Figure 2 shows results of the
accelerated pavement testing performed in a recycled pavement with a FB layer using 1.8%
bitumen and 2.0% cement content. After pavement construction a 40 kN traffic load was applied
using the South African Heavy Vehicle Simulator HVS (6,7). Multidepth Deflectometer (MDD)
were installed in the pavement structure to measure deflection and to calculate the effective
stiffness of each layer.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Elastic Modulus
(MPa)
Greece
South Africa
Based on the results obtained from the pavement structural analysis, it can be seen that
the FB layer of the pavement structure in Greece was subjected to a SR of 0.20 while the FB
layer of the pavement structure in South Africa to a SR of 0.55. Both values are very different
and could explain the behavior observed using the concepts for the thickness design of perpetual
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
pavements (23). The perpetual pavements concept states that if the deformations that occur at
critical points are below a specific value (endurance limit), then the asphalt layer will not suffer
fatigue and therefore its structural lifetime could be extended to periods of even 50 years. Under
this scenario, only functional maintenance actions should be conducted for keeping the standard
expected. Results of studies carried out to validate this affirmation have suggested that the
limiting tensile strain at the bottom of the asphalt concrete layers should be no greater than 60
and that at the top of the subgrade the vertical strain should be limited to 200.
Using a similar approach to FB layers, it may be assumed that if SR is lower than the
endurance limit, then the FB will maintain its stiffness.
Based on the above discussion, it is expected that after a period the FB stiffness plateaus,
which will depend of the stress/strain levels applied to the layer depending on the traffic loads
applied in the road, similarly to the behavior observed in the pavements analyzed in this article.
In the case of the pavement structure of Greece, it is probable that a SR of 20% is lower than the
SR required for generating some damage to the bonds or mastic produced by the bitumen and
cement particles.
This paper presents results of a research work carried out to represent and quantify the
stiffness evolution of FB mixes subjected to different SR levels. This information was used for
two purposes:
For establishing the effect of the different bitumen and cement content on the long term
performance of the FB mixtures, in order to determine the appropriate content of each
one.
Define the maximum stress level that must be accepted in the FB layer in order to
maintain constant its stiffness.
19 mm
6%
13%
46%
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Crushed/Fractured Particles:
Plasticity Index:
Optimum Moisture Content:
Maximum Density:
100%
Non Plastic
6.0%
2190 kg/m3
44
45
(1)
Where; P is the vertical load, t is the thickness of the specimen and d is the diameter.
1
2
(2)
3
4
5
Where; S is the elastic modulus, x the effective tensile stress and the measured tensile strain.
Where; x is the effective tensile stress and ITS is the maximum tensile stress.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
(3)
Figure 3 shows how the stiffness of the mix decreases due to the progressive increase in plastic
and elastic deformations under cyclic loading.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
FB Content (%)
1.0
2.0
3.0
2.0
2.0
2.0
ANALYSIS OF RESULTS
Effect of the Foamed Bitumen Content on the Stiffness Evolution
Figure 4, shows results of the ITFT carried out on mixes FB1C1, FB2C1 and FB3C1 in order to
evaluate the effect of the bitumen content on the stiffness evolution. Results correspond to the
average of two specimens evaluated.
Trend lines were fitted and extrapolated to each mix for each SR with the objective of
estimating the stiffness for additional load cycles. Examples of the trend lines for FB2C1are
shown in Figure 4. Table 3, shows details of trend lines of each mix as well as the stiffness
expected at100,000 and 1,000,000 load cycles. In addition, the quotient between the Initial and
Long Term Stiffness (RSIE) is presented in the same Table 3.
It must be noted that a couple of test were carried out using more than 100,000 cycles in
order verify the extrapolation made.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
MixFB2C1
(2% FB+1%Cem)
MixFB3C1
(3% FB+1%Cem)
Ratio
(%)
20
30
40
20
30
40
20
30
c
c
c
c
c
c
c
c
c
-0.013
-0.136
-0.755
-0.031
-0.140
-0.384
-0.019
-0.157
50.7
83.4
96.6
88.5
85.3
94.5
61.2
90.1
1098 (92%)
654 (55%)
131 (11%)
1310 (82%)
828 (52%)
407 (25%)
1220 (87%)
682 (49%)
1066 (89%)
478 (40%)
23 (2%)
1220 (76%)
600 (38%)
168 (11%)
1168 (83%)
475 (34%)
-0.581
96.3
228 (16%)
60 (4%)
40
Mix FB2C0
23
24
i = 100,000 (RSIE)
i = 1,000,000 (RSIE)
20
c
-0.031
70.2
849 (85%)
791 (79%)
(2% FB+0%Cem)
30
c
-0.261
95.1
341 (34%)
187 (19%)
40
Failure
Mix FB2C1
20
c
-0.031
88.5
1310 (82%)
1220 (76%)
(2% FB+1%Cem)
30
c
-0.140
85.3
828 (52%)
600 (37%)
40
c
-0.384
94.5
407 (25%)
168 (10%)
Mix FB2C2
20
c
-0.013
61.2
1538 (90%)
1493 (88%)
(2% FB+2%Cem)
30
c
-0.139
90.1
844 (50%)
613 (36%)
40
c
-0.576
96.3
324 (19%)
86 (5%)
y: Stiffness of the Indirect Tensile Fatigue/Modulus Test; x: Load cycles; c: Constant value not used in
the analysis.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
10
Based on the results shown on Figure 4 and Table 3, the following observations can be made:
In general, all mixes show similar trends in terms of stiffness evolution at different stress
ratios. According to results presented in Table 3, curves were fitted to measured values.
The curves are described by a power relationship with a,b regression coefficients. Table 3
shows that b coefficients are very similar. For example, when a SR of 30% was
applied, b coefficient for mixes FB1C1, FB2C1 and FB3C1 were -0.136, -0.140 and
-0.157 respectively. When a SR of 20% and 40% were applied, a larger variability was
observed, but b values were within the same range.
The stiffness of the FB mixes plateaus after a certain number of load cycles when the
stress ratio is in the order of 20%. For example, Mix FB2C1 showed a relative constant
stiffness of 1200 1300 MPa for the long term, which represents almost 80% of the
initial Stiffness (1600 MPa). In the case of SR equal to 30%, the mix showed a good
evolution of stiffness, with a constant decreasing line with a low slope. In this case the
stiffness of the mix was 600 MPa after 1 million load cycles, which represents the 37.5%
of the initial stiffness (1600 MPa). This fact showed that after all the load cycles were
applied, the mix still is able to keep cohesion due the effect of the stabilizing agents.
Similar results are observed in mixes FB1C1 and FB3C1. In the case of a SR equal to
40%, mixes showed a clear reduction in stiffness with load cycles. All of them showed
almost null stiffness after 1 million cycle loads.
When mixes were subjected to a stress ratio of 50%, specimens collapsed in a relative
short period indicating that stresses and strains applied are much larger than the cohesion
provided by the FB and cement to the mix.
Although trends of the stiffness for each mix were very similar, results show that there is
an optimum bitumen content that maximizes the stiffness. In this case, the better results
in terms of stiffness evolution were observed in Mix FB2C1 with 2% FB and 1% cement.
Table 4 shows the slope of the trend line for each mix when SR is 50%.It is interesting to
note that when mixes were subjected to a SR of 50%, the stiffness rate of change for each
mix were very similar, which means that the FB content did not significantly affect the
behavior of the mix under these stress conditions.
Mix FB1C1
Mix FB2C1
Mix FB3C1
Mix FB2C0
Mix FB2C1
Mix FB2C2
50
50
50
40(*)
50
50
Slope (m)
(y = m*x + b)
-0.1116
-0.1207
-0.1189
-0.0643
-0.1207
-0.2183
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
11
It can be observed that there are significant differences between mixes without cement
(Mix FB2C0) and with cement (Mixes FB2C1 and FB2C2) for the same bitumen content.
Mixes with cement (FB2C1 and FB2C2) have larger stiffness compared to Mix FB2C0
(with FB only). While stiffness of mixes FB2C1 and FB2C2 are 1310 MPa and 1538
MPa respectively for 100,000 cycles, stiffness for mix FB2C0is 849 MPa. These results
indicate that a pavement structure with FB2C1 and FB2C2 layers will have a better
structural capacity than using mix FB2C0.
Additionally, Mixes FB2C1 and FB2C2 were able to withstand more cycles than mix
FB2C0 for the same test loading sequence. While Mix FB2C0 collapsed for a SR equal to
40%, mixes FB2C1 and FB2C2 collapsed for a SR equal to 50%.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
12
The stiffness of the FB mixes plateaus after a certain number of load cycles when the
stress ratio is in the order of 20%, providing guarantee that the mix will keep its cohesion
in the long-term.
For the case of a SR equal to 30%, mixes with cement (FB2C1 and FB2C2) showed an
acceptable stiffness evolution. It is interesting to note that although mix FB2C2 initially
has a greater stiffness than mix FB2C1, when extrapolated to 100,000 cycles, both
stiffnesses tend to be the same. Stiffness of Mix FB2C1 is equal to 828 MPa for 100,000
extrapolated cycles while stiffness of mix FB2C2 for the same conditions is 844 MPa.
For the case of a SR equal to 40%, mix FB2C1 showed a better stiffness evolution than
mix FB2C2. In this case the stiffness rate of change for mix FB2C2 was significantly
higher than mix FB2C1, which gives lower stiffnesses in the long term. When comparing
trend lines fitted to the data, it can be seen that values of the variable b are -0.384 for
mix FB2C1 and -0.576 for mix FB2C2. If the fitted equations are used to extrapolate the
stiffness, at 1 million load cycles the stiffness of mix FB2C1 is 168 MPa while stiffness
of mix FB2C2 is 86 MPa. In contrast, mix FB2C0 (only with FB) collapsed for a SR
equal to 40%.
For the case of a SR equal to 50% in mixes FB2C1 and FB2C2 as well as SR equal to
40% in mix FB2C0, trend lines were fitted to the data (see Table 4). The slopes for each
equation (m) indicated that as the cement content increases the rate of change of the
stiffness also increases. This means that when increasing the cement content, mixes with
the worst behavior are obtained for a SR of 50%.
Overall, the use of 2.0% cement in a FB layer will be beneficial only if the FB layer is
loaded to a SR equal or lower than 30%. If the FB layer is loaded to a SR equal or higher
than 40%, it is recommended only to use 1.0% cement together with FB.
CONCLUSIONS
An analysis of the stiffness degradation of foamed bitumen mixtures was done using the indirect
tensile fatigue test (ITFT) aiming to identify the evolution of the elastic modulus or stiffness of
the mixture in long term. The experimental study was designed to obtain the maximum stress
level that the FB layer is capable of withstanding without significantly reducing its stiffness as
well as to study the effect of the bitumen and cement contents in the long term stiffness.
Based on the results presented the main conclusions may be summarized as follow:
The stiffness of FB mixes will evolve according the stress-level applied to the FB layer.
If the stress level is lower than a specific value the stiffness of the mix will remain
constant at a value very close to the initial stiffness. If the stress level is greater than a
specific value the stiffness of the mix will gradually decrease.
If the Stress-Ratio of the mix is lower than 20%, then the mix will have a stiffness within
a range of 75% - 90% of the Initial Stiffness after one million load cycles. The value will
keep relatively constant during the life of the pavement.
If the Stress-Ratio of the FB mix layer is between 20% and 40%, the stiffness will
gradually decrease. The higher the Stress-Ratio, the higher will be the reduction rate of
stiffness.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
13
If the Stress-Ratio is around 50% then the mix will collapse and the cohesion provided
by stabilizing agents will reduce to zero in a relative short period. In that case the
stiffness will be equivalent to the elastic/resilient modulus of the reclaimed material
without stabilizing agents.
Analysis of the bitumen content effect showed very little influence on the stiffness
evolution, but it was possible to find an optimum content that maximizes the stiffness; in
this case, Mix FB2C1 followed by mixes FB3C1 and FB1C1. Conversely, the effect of
cement content was significant on the stiffness evolution as well as in the absolute value
of stiffness. In addition, it is possible to state that cement must always be incorporated to
FB mixes for guaranteeing minimum short and long term stiffness.
Results showed that the use of 2% of cement in the FB layer will be beneficial only if the
FB layer is loaded to a SR equal or lower than 30%. If the FB layer is loaded to a SR
equal or higher than 40%, it is recommended to add only 1% of cement and FB.
The data and analysis provided in this research work can be used to estimate an effective elastic
modulus (EMM) of the FB mix in the long term based on the stress-state expected at the FB
layer. The EMM may be defined as a percentage of the initial stiffness and a shift factor must be
developed to adapt the elastic modulus provided by the ITFT to the elastic modulus that better
represents the mechanical properties of the FB mix in the field.
It must be noted that the laboratory work presented in this paper was carried out using
only one aggregate source, one bitumen source, a single active filler and a single temperature.
All these factors contribute significantly to the FB mix performance and therefore results
obtained are limited. Conclusions must be validated using a larger experimental design.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
Loizos A. In-situ characterization of foamed bitumen treated layer mixes for heavy-duty
pavements. International Journal of Pavement Engineering, Vol. 8, N 2, June 2007, 123135.
Bowering R.H. Foamed bitumen: Production and application of mixtures, evaluation and
performance of pavements. In Proceedings of Association of Asphalt Paving
Technologists, Vol 45: 45373, 1976.
Jones D., Fu P., Harvey J.T., and Halles F. Full-Depth Reclamation with Foamed
Asphalt: Final Report. Davis and Berkeley, CA. University of California Pavement
Research Center. (Research Report UCPRC-RR-2008-07), 2008.
Fu P., Jones D., Harvey J.T. and Halles F. Investigation of the Curing Mechanism of
Foamed Asphalt Mixes Based on Micromechanics Principles. Journal of Materials in
Civil Engineering, Vol. 22, pp.29-38, 2010.
Long F. M. The development of structural design models for foamed bitumen treated
pavement layers. Contract Report CR-2001/76, Gauteng CSIR, South Africa, 2001.
Verhaeghe B, Sadzik E. and Visser A. Three Decades of Development and Achievements:
The Heavy Vehicle Simulator in Accelerated Pavement Testing. 10th International
Conference on Asphalt Pavements (ICAP), 2006.
Du Plessis L., Coetzee NF and Burmas N. Heavy vehicle simulator in accelerated
pavement testing: a historical overview and new developments. 3rd International
Conference on Accelerated Pavement Testing (APT), Madrid, Spain, October 2008.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
14
Asphalt Academy Transportek. South African Interim Technical Guidelines. Design and
Use of Foamed Bitumen Treated Materials - TG2. CSIR Transportek, South Africa, 2002.
Sunarjono Sri. Tensile Strength and Stiffness Modulus of Foamed Asphalt Applied to a
Grading Representative of Indonesian Road Recycled Pavement Materials.
DinamikaTeknikSipil, Vol.7, pp 1 10, January 2007.
Thenoux G., Gonzalez A. and Jamet A. Aspectos Constructivos del Primer Proyecto de
Reciclado en Fro In-Situ con Asfalto Espumado en Chile. Revista Ingeniera en
Construccin Vol. 18, 2003, pp. 148-156, Pontificia Universidad Catlica de Chile.
Twagira M.E., Jenkins K.J. and Ebels L.J. Characterization of Fatigue Performance of
Selected Cold Bituminous Mixes. Proceedings of the 10th International Conference on
Asphalt Pavements, Quebec, Canada, 2006.
Jenkins K.J.; Long F.M. and Ebels L.J. Foamed Bitumen Mixes = Shear Performance?.
International Journal of Pavement Engineering, Volume 8, Issue 2, June 2007, pages 85
98.
Long F. M. and Theyse H.L. Mechanistic-Empirical Structural Design Models for
Foamed and Emulsified Bitumen Treated Materials. Conference on Asphalt Pavements
for Southern Africa CAPSA, 2004. South Africa.
Nataatmadja A. Some Characteristics of Foamed Bitumen Mixes. In Transportation
Research Record. Journal of the Transportation Research Board, N 1767, TRB, National
Research Council, Washington D.C., 2001, pp. 120-125.
Ramanujam J.M. and Jones J.D. Characterization of Foamed-Bitumen Stabilization.
International Journal of Pavement Engineering, Vol. 8, Issue 2, June 2007, pp. 111-122.
Leek C. Insitu Foamed Bitumen Stabilisation The City of Canning Experience.
Proceedings from 20th Australian Road Research Board Conference, Melbourne, 2001.
Saleh M. Effect of Rheology on the Bitumen Foamability and Mechanical Properties of
Foam Bitumen Stabilised Mixes. International Journal of Pavement Engineering, Volume
8, Issue 2, June 2007, pages 99 110.
Halles F. and Thenoux F. Degree of influence of active fillers on the properties of
recycled mixes with foamed asphalt. In Transportation Research Record 2095: 127-135,
2009.
Lancaster J., McArthur L. and Warwick R. Vicroads Experience with Foamed Bitumen
Stabilisation. Proceedings 17th Australian Road Research Board Conference, Volume 16,
Part 3, pp. 193-211, 1994.
Hodgkinson A. and Visser, A.T. The role of fillers and cementitious binders when
recycling with Foamed Bitumen or Bitumen Emulsion. Proceedings of the 8th Conference
on Asphalt Pavements for Southern Africa. September, 2004.
Jenkins, K.J. Mix design considerations for cold and half-cold bituminous mixes with
emphasis on foamed bitumen. Ph.D. Thesis. University of Stellenbosch, South Africa,
2000.
Gonzalez, A, Cubrinovski, M, Pidwerbesky, BD, and Alabaster, D 2009, Full scale
experiment on foam bitumen pavements in CAPTIF accelerated testing facility,
Transportation Research Record 2094, Transportation Research Board, Washington,
D.C., 21-29.
Committee on General Issues in Asphalt Technology. Perpetual Bituminous Pavements.
Circular Number 503. A2D05 Committee, Transportation Research Board, 2001.