Construction and Building Materials 113 (2016) 506–512

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: [Link]/locate/conbuildmat

Mechanical performance of mechanomutable asphalt binders under

cyclic creep and recovery loads
F. Moreno-Navarro a,⇑, G.R. Iglesias b, M.C. Rubio-Gámez a
a
Laboratory of Construction Engineering ([Link]), E.T.S. Caminos, Canales y Puertos, Spain
b
Physics of Interfaces and Colloidal Systems Department of Applied Physics, School of Science University of Granada, 18071 Granada, Spain

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 MAB modify their mechanical

behavior when they are activated by a
magnetic field.
 MAB behave as smart materials
offering a wide range of engineering
applications.
 MAB improve their resistant against
plastic deformations.
 An increase in the magnetic particle
content improve its mechanical
behavior.

a r t i c l e i n f o a b s t r a c t

Article history: Mechanomutable asphalt binders are able to modify their mechanical behavior when they are activated
Received 28 December 2015 by a magnetic field. Under the effect of the magnetic field, these binders can increase their stiffness and
Received in revised form 7 March 2016 perform elastically, and once the magnetic field is removed, the material recovers its original visco-elastic
Accepted 8 March 2016
properties. Because of this fact, these smart materials could offer a wide range of interesting engineering
applications. One such use could be its incorporation into structures that are required to support cyclic
loading, such as asphalt pavements, as they can be punctually activated to minimize the impact of traffic
Keywords:
loads. Based on these considerations, this paper studies the mechanical performance of mechanomutable
Creep and recovery
Asphalt
asphalt binders under cyclic loads similar to those occurring in real pavements during their service life.
Smart material For this purpose, a magneto-rheometer has been used to carry out different multiple stress creep and
Pavement recovery tests under different magnetic field strength. The results obtained show that, in spite of the fact
Mechanomutable binder that these binders do not have a significant recovery capacity, they can reduce the cumulative permanent
Magnetic particles deformations produced in them due to the effect of cyclic loads when they are under the effect of a mag-
netic field.
Ó 2016 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.
E-mail addresses: fmoreno@[Link] (F. Moreno-Navarro), igleisas@[Link] (G.R. Iglesias), mcrubio@[Link] (M.C. Rubio-Gámez).

[Link]
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
 F. Moreno-Navarro et al. / Construction and Building Materials 113 (2016) 506–512 507

1. Introduction the mechanical performance of MAB under cyclic loads – which

combine stress periods and rest periods, where the magnetic field
In recent years, pavement engineering research has focused on is only activated under the presence of the load and switched off
the development of new smart materials that could render the during the rest period.
road a more efficient infrastructure [1–4], and which could This paper constitutes an evaluation of the mechanical behavior
increase its functionality [5–10]. In this respect, a new pavement of MAB under loads comparable to those that would occur during
material (mechanomutable asphalt binders, MAB) has been the real application of these materials in a road pavement. For this
recently developed [11] and it could provide an interesting tool purpose, several MAB (manufactured with different bituminous
for improving the mechanical performance of the main civil engi- matrix and particle concentrations) have been tested using a
neering infrastructures (roads, highways, airports, ports, etc.). magneto-rheometer (which can apply a magnetic field over the
These binders (MAB) are able to adapt their mechanical response sample). The main objective of this research was to assess the
to the stresses that they are required to support, and they could mechanical response of the MABs once the load and the magnetic
provide pavements that are sufficiently rigid to reduce the appear- field are removed.
ance of plastic deformations, whilst flexible enough to avoid the
formation of cracks. 2. Methodology
MAB are based on similar concepts to those used in the devel-
opment of magnetorheological (MR) fluids or elastomers [12–15]. 2.1. Materials
These binders are composed of a bituminous matrix that has been
modified with magnetic particles that are homogeneously dis- Based on the results obtained from previous research [11], two
persed over it. When these particles are activated by an external different neat binders, named B 20/30 and B 50/70, and two differ-
magnetic field, the binder can considerably increase its stiffness, ent magnetic particle concentrations with 1% w/w and 10% w/w (%
and when the magnetic field is removed, its initial rheological concentration over the weight of the binder) were selected to man-
properties are restored [11]. The results obtained in previous ufacture representative MABs. In order to compare the perfor-
research have shown that the type of bituminous matrix (harder mance of these smart binders, a high-performance commercial
or softer) used in the manufacturing of the binders has an impact binder (commonly used in the construction of long-life pavements)
on the degree to which they can modify their mechanical proper- composed of Styrene-Butadiene-Styrene polymer modified bitu-
ties. In addition, as the strength of the magnetic field applied or men (PMB) was also tested. The characteristics of the neat bitumi-
the concentration of the magnetic particles increases, the degree nous matrix used for the manufacture of the MABs, as well as the
of change produced in the properties of the binders also increases. characteristics of the high-performance polymer modified bitumen
During the service life of a real pavement, a given zone is only (PMB) used to compare the results obtained, are summarized in
under mechanical loading when the wheel of a vehicle passes over Table 1. Spherical HQ carbonyl iron powder was used as magnetic
it – for the remainder of the time this zone is free of mechanical particles to manufacture the MABs. These particles have a median
loads. These mechanical loadings L1, L2, etc., and the lag time t1, particle size of d50 = 2.3 lm (particle diameter ranges from 0.5 lm
t2, etc., are shown in Fig. 1. The load supported at a specific point to 3 lm) and an iron content of 97% (density = 7.5 g/cm3).
‘‘A” is plotted as a function of time in the lower graphs, where the The manufacture of the MABs was conducted at a constant tem-
rest periods and the mechanical loading can be seen in detail. perature of 120 °C, where the iron powder particles and the neat
Thus, for the effective application of these types of binders binders were mixed using an agitator at 200 rpm for 10 min. Based
(MAB), the magnetic field should only be applied during the instant on these circumstances, four different MABs were manufactured: B
that this zone is altered by the presence of the wheel, and then 20/30-1% (composed of a B 20/30 matrix and 1% over the weight of
switched off until a new wheel is detected. Studies carried out with the binder of iron powder particles); B 20/30-10% (composed of a B
MR fluids have demonstrated that the changes produced in their 20/30 matrix and 10% over the weight of the binder of iron powder
rheological behavior occur in a drastic, reversible, and rapid way, particles); B 50/70-1% (composed of a B 50/70 matrix and 1% over
typically with an order of magnitude of a millisecond [16–18]. the weight of the binder of iron powder particles); and B 50/70-
Nonetheless, there is still little information available regarding 10% (composed of a B 50/70 matrix and 10% over the weight of

Fig. 1. Sketch of the mechanical loads suffered by a given point in the pavement.
508 F. Moreno-Navarro et al. / Construction and Building Materials 113 (2016) 506–512

Table 1 characterization was carried out in a rheometer, provided with a

Properties of the bituminous matrix tested. magneto-rheological device, at 60 °C. A parallel plate (25 mm in
Type of bituminous matrix B 20/30 B 50/70 PMB diameter) configuration was used, with a thickness of 1 mm. Dur-
Penetration at 25 °C (mm, EN 1426 [19]) 22 65 62 ing the test, a one-second creep load was applied to the asphalt
Softening point (°C, EN 1427 [20]) 64.8 51.8 68.4 binder sample upon which the sample was allowed to recover
Fraass fragility (°C, EN 12593 [21]) 1 8 17 for 9 s. The test began with the application of a low stress
Flash point (°C, ISO 2592 [22]) >245 >230 258 (0.1 kPa, where the binder offers a linear response) for 15 creep/
recovery cycles, upon which the stress is increased (3.2 kPa, non-
the binder of iron powder particles). Fig. 2 shows scanning electron linear response) and applied for 15 further cycles.
microscopy pictures of a MAB with the iron particle homogenously In order to reproduce characteristics similar to those assumed
distributed at various scale lengths. to impact on the performance of a real pavement, during the
one-second creep load, a magnetic field of 430 mT was generated
2.2. Testing plan by a coil through which a controlled current was passed, located
under the lower plate of the rheometer [11]. After the one-
The mechanical performance under cyclic loading and variable second creep load, the magnetic field was switched off and the bin-
magnetic field was evaluated by using the Multiple Stress Creep der allowed to recover for 9 s without the presence of the field
and Recovery Test (MSCRT) [23]. This test method allowed for an (Fig. 3).
assessment of the response of asphalt binders at high tempera- The results obtained in this test are expressed as a measure of
tures, as well as the delayed elastic response [24,25]. The MSCRT how much the sample returns to its previous shape after being

Fig. 2. Scanning electronic microscopy pictures of a MAB at different scale lengths: (a) 100 lm; (b) 50 lm; (c) 20 lm; (d) 10 lm. Details of pure bituminous matrix and
magnetic particles when observed at scale lengths of 2 lm.

Fig. 3. Sketch of the rheology test carried out to evaluate the binders with the presence of a magnetic field.
 F. Moreno-Navarro et al. / Construction and Building Materials 113 (2016) 506–512 509

Fig. 4. Sketch of the different situations that might be expected to appear during the test.

(a) (a)
1000 50 165
48 163
100 46 161

Strain B50/50+1%
44 159
Strain PMB

10 42 157
Strain

40 155
1 38 153
36 151
0.1 34 149
32 147
0.01 30 145
0 50 100 150 200 250 300 350 0 2 4 6 8 10
Time (s) Time (s)
PMB B50/70+1% Inactive B50/70+1% Active PMB B50/70+1% Active

(b) (b)
1000 45 88
44 87
100 43 86

Strain B50/70+10%
42 85
Strain PMB

10 41 84
Strain

40 83
1 39 82
38 81
0.1 37 80
36 79
0.01 35 78
0 50 100 150 200 250 300 350 0 2 4 6 8 10
Time (s) Time (s)
PMB B50/70+10% Inactive B50/70+10% Active PMB B50/70+10% Active

Fig. 5. Results of the creep and recovery tests carried out with the MABs Fig. 6. Final cycle of the creep and recovery tests carried out with the MAB’s
manufactured with a B 50/70 bituminous matrix: (a) with a magnetic particle manufactured with a B 50/70 bituminous matrix: (a) 1% w/w magnetic particles; (b)
content of 1% w/w; (b) with a magnetic particle content of 10% w/w. 10% w/w magnetic particles.

repeatedly stretched and then relaxed (recovery percent). It is con-

sidered that the higher the degree of elastic recovery under the As observed in previous studies [11], at the temperature of test-
MSCRT, the higher the fatigue resistance of the binder [24]. Fur- ing used (60 °C), the magnetic particles contained in the MABs can
thermore, a measure of the amount of residual strain left in the create an internal structure when they are activated by a magnetic
specimen after repeated creep and recovery (relative to the field (that induces a more elastic response of the MAB). Therefore,
amount of stress applied) is made (Non-Recoverable Creep Compli- during the tests developed here, two different situations appeared
ance, Jnr) in order to characterize the propensity of the asphalt bin- (Fig. 4): (a) an isotropic distribution of the magnetic particles with-
der to resist permanent deformations in the pavement [26]. out applying the magnetic field; (b) an anisotropic distribution of
510 F. Moreno-Navarro et al. / Construction and Building Materials 113 (2016) 506–512

Table 2
Average recovered strain under 3200 Pa.

Binder Current intensity Average of recovery (%) Non-recoverable creep compliance Jnr, (1/Pa)
PMB Non-active 29.51 0.0011
B 20/30 – 1.0% Non-active 6.46 0.0007
B 20/30 – 1.0% Active 6.99 0.0005
B 20/30 – 10% Non-active 4.86 0.0008
B 20/30 – 10% Active 12.97 0.0002
B 50/70 – 1.0% Non-active 0.00 0.0061
B 50/70 – 1.0% Active 0.00 0.0038
B 50/70 – 10% Non-active 0.00 0.0037
B 50/70 – 10% Active 1.35 0.0023

the magnetic particles when the magnetic field is on. One of the (a)
key aspects to be examined during this study can be derived from 100
these situations, and is concerned with what happens with this
internal structure after each load cycle (Fig. 4b): it returns to the
10
isotropic distribution (returning the binder to its initial state) or
remains in an anisotropic one (inducing a more rigid response of

Strain
the binder). 1

0.1
3. Results and discussion

Fig. 5 summarizes the mean results obtained during the mea- 0.01
0 50 100 150 200 250 300 350
surement of the strain produced in each load cycle when the MABs
Time (s)
manufactured with a bituminous matrix B 50/70 were tested
PMB B20/30+1% B20/30+1% Acve
(magnetic particle content of 1% w/w and 10% w/w). It is clear that
at both stress levels tested (100 Pa and 3200 Pa), the activation of
(b)
the magnetic field during the presence of each stress load reduced 100
the level of strain reached by the material. This effect is in accord
with similar studies conducted in MR plastomers or elastomers
10
[15], and was more marked as the magnetic particle content
increased, as shown in Fig. 5b. In addition, there was a higher
Strain

capacity of the MABs to change their properties under these cir- 1

cumstances [11]. Nonetheless, the deformations produced in the
MABs manufactured with a B50/70 bituminous matrix were higher 0.1
than those obtained by a high-performance polymer modified bin-
der (PMB).
0.01
Fig. 6 shows the ability to recover deformations in the materials 0 50 100 150 200 250 300 350
tested, measured from the percentage of recovered strain in the Time (s)
final cycle of the test at 3200 Pa. It is clear that PMB (which incor-
PMB B20/30+10% B20/30+10% Acve
porates elastomers) was more capable of recovering part of the
strain produced by the stress load than the MABs. Thus, it can be Fig. 7. Mean results of the creep and recovery tests carried out with the MABs
said that regardless of the percentage of magnetic particles incor- manufactured with a B 20/30 bituminous matrix: (a) 1% w/w particle content; (b)
porated into the binder, B 50/70 MABs are not able to recover a sig- 10% w/w particle content.

nificant amount of the deformations when the load is removed.

Table 2 summarizes the average level of recovery strain produced appearance of permanent deformations in asphalt binders could
in the different binders tested under 3200 Pa of stress effort. Based be influenced more by the penetration grade of the bitumen than
on these results, it can be suggested that the reduction of the by the recovery ability of the modifiers used.
cumulative deformations produced in the MABs are not due to Fig. 8 shows the curves described in the final cycle of the test at
an increase in their recovery and it is due to a decrease in their 3200 Pa for the different B20/30 MABs tested. Once again it is clear
propensity to deform (Jnr values decrease). that the PMB was more able to recover the strain produced by the
Fig. 7 shows the mean results obtained during the measure- stress loads than the MABs. Nonetheless, in this case, the recovery
ment of the strain produced in each load cycle in the case of the ability of the B20/30 MABs was slightly higher than that offered by
MABs manufactured with the B20/30 bituminous matrix (with a the B 50/70 MABs (Table 2). Similarly, the propensity to produce
magnetic particle content of 1% w/w and 10% w/w). As occurred plastic deformations (Jnr values, Table 2) in B20/30 MABs was infe-
in the case of the MABs manufactured with the B50/70 matrix, rior to that offered by the PMB.
the magnetic field exerted a positive effect over the binders, and Based on these findings, it can be concluded that MABs do not
its presence reduced the strain produced in the material at any show a considerable strain recovery capacity. Nonetheless, it is
given stress level (100 Pa and 3200 Pa). In this respect, the effect observed that as the magnetic particles content increased, the
produced also increased as the amount of magnetic particles recovery ability of these binders also increased slightly when they
increased (Fig. 7b). However, in this case, due to the greater stiff- were activated due to the presence of a magnetic field (Table 2).
ness of the B20/30 bituminous matrix, the cumulative deforma- This fact could be due to the presence of residual movements
tions produced in the MABs were inferior to those that appeared appearing in the bituminous matrix due to the action of the parti-
in the reference PMB binder. Therefore, it must be noted that the cles when the magnetic field disappears. The particles could
 F. Moreno-Navarro et al. / Construction and Building Materials 113 (2016) 506–512 511

(a) cannot significantly recover the strain produced under cyclic

45 25 loads, and they mainly reduce cumulative deformations under
44 24 cyclic loading because they are more able to resist their
43 23 development.

Strain B20/30+10%
42 22 – In spite of the penetration grade of the bitumen exerting a sig-
Strain PMB

41 21 nificant influence on the amount of deformations produced in

40 20 each cycle, the magnetic particle content used in the manufac-
39 19 ture of MABs will also play an important role. As the magnetic
38 18
particle content increases, the positive effect that reduces the
37 17
appearance of plastic deformations is also increased.
36 16
– After the disappearance of the magnetic field effect, the internal
35 15
0 2 4 6 8 10 structure created by the particles included in the MABs could be
Time (s) partially decomposed (the particles could try to return to their
PMB B20/30+1% Acve initial positions), and this phenomenon could induce internal
residual movements in the binder that could lead to the recov-
(b) ery of part of the deformations produced.
45 15
44 14 Taken together, these findings suggest that mechanomutable
43 13 asphalt binders (activated by an external magnetic field only under

Strain B20/30+10%
42 12 the presence of an external load) could be regarded as an interest-
Strain PMB

41 11 ing alternative for reducing some of the main pathologies that uni-
40 10 versally appear in asphalt pavements (plastic deformations,
39 9 indentations, etc.).
38 8
37 7
36 6 References
35 5
0 2 4 6 8 10 [1] Q. Liu, E. Schlangen, A. García, M. van de Ven, Induction healing of asphalt
Time (s) mastic and porous asphalt concrete, Constr. Build. Mater. 25 (2011) 3746–
3752.
PMB B20/30+10% Acve [2] A. García, E. Schlangen, M. van de Ven, G. van Bochove, Optimization of
composition and mixing process of a self-healing porous asphalt, Constr. Build.
Fig. 8. Details of the curves obtained in the last cycle of the creep and recovery tests Mater. 30 (2012) 3746–3752.
carried out with the MAB’s manufactured with a B 50/70 bituminous matrix: (a) 1% [3] M. Carmen Rubio, Fernando Moreno, María José Martinez-Echevarria, Germán
w/w magnetic particles content; (b) 10% w/w magnetic particles content. Martinez, Jose Miguel Vázquez, Comparative analysis of emissions from the
manufacture and use of hot and half-warm mix asphalt, J. Clean. Prod. 41
(2013) 1–6.
[4] M. Pérez-Martínez, F. Moreno-Navarro, J. Martín-Marín, C. Rios-Losada, M.C.
Rubio-Gámez, Analysis of cleaner technologies based on waxes and surfactant
additives in road construction, J. Clean. Prod. 64 (2014) 374–379.
attempt to return to their initial positions, and therefore produce [5] A.K. Batra, S. Bhattacharjee, A.K. Chilvery, Energy harvesting roads via
internal movements in the bituminous matrix that induce a pro- pyroelectric effect: a possible approach, J. Energy Harvest. Storage: Mater.,
cess of recovery within the material. Devices Appl. II (2011).
[6] V. Bobes-Jesus, P. Pascual-Muñoz, D. Castro-Fresno, J. Rodriguez-Hernández,
Asphalt solar collectors: a literature review, Appl. Energy 102 (2013) 962–970.
4. Conclusions [7] A. García, M.N. Partl, How to transform an asphalt concrete pavement into a
solar turbine, Appl. Energy 119 (2014) 431–437.
[8] N. Lajnef, K. Chatti, S. Chakrabartty, M. Rhimi, P. Sarkar, Smart Pavement
This paper summarizes the results obtained during the evalua- Monitoring System FHWA-HRT-12-072, Michigan State University, 2013.
tion of MABs (Mechanomutable Asphalt Binders) under load condi- [9] E. Levenberg, Estimating vehicle speed with embedded sensors, Transp. Res. C:
Emerg. Technol. 46 (2014) 300–308.
tions similar to those that they would be expected to support
[10] C.Y. Chang, H.S. Tan, Evaluation of Magnetic Markers as a Position Reference
during their service life in a road pavement. For this purpose, var- System for Ground Vehicle Guidance and Control, California Partners for
ious MABs composed of variable bituminous matrix and magnetic Advanced Transit and Highways, UC Berkeley, 2003.
particle contents have been studied, and their performance has [11] F. Moreno-Navarro, G.R. Iglesias, M.C. Rubio-Gámez, Development of
mechanomutable asphalt binders for the construction of smart pavements,
been compared with that of a high-performance polymer modified Mater. Des. 84 (2015) 100–109.
binder. For their evaluation, multiple stress creep and recovery [12] GR Iglesias, S. Ahualli, J. Echávarri Otero, L. Fernández Ruiz-Morón, J.D.G.
cycles (which are composed of 1 s of creep and 9 s of free recovery) Durán, Theoretical and experimental evaluation of the flow behavior of a
magnetorheological damper using an extremely bimodal magnetic fluid,
have been applied on the binders, with the variable presence of a Smart Mater. Struct. 23 (2014) 11.
magnetic field when the creep stress is applied. Based on the [13] M. Zubieta, S. Eceolaza, M.J. Elejabarrieta, M.M. Bou-Ali, Magnetorheological
effects observed in the binders used during this study, the follow- fluids: characterization and modeling of magnetization, Smart Mater. Struct.
18 (9) (2009) 6.
ing conclusions can be drawn: [14] P.P. Phulé, M.P. Mihalcin, S. Genc, The role of dispersed-phase remnant
magnetization on there dispersibility of magnetorheological fluids, J. Mater.
– Mechanomutable asphalt binders can reduce their propensity to Res. 14 (7) (1999) 3037–3041.
[15] L. Chen, X.L. Gong, W.H. Li, Microestructure and viscoelastic properties of
deform under cyclic loads similar to those that appear in road anisotropic magnetorheological elastomers, Smart Mater. Struct. 16 (2007) 1–6.
pavements when they are activated by a magnetic field in the [16] D. Baranwal, T.S. Deshmukh, MR-fluid technology and its application- a
same instant that the load is applied. This effect could consider- review, Int. J. Emerg. Technol. Adv. Eng. 2 (12) (2008) 563–569.
[17] Y. Xu, X. Gong, S. Xuan, X. Li, L. Qin, W. Jiang, Creep and recovery behaviors of
ably increase the extent to which asphalt pavements can resist
magnetorheological plastomer and its magnetic-dependent properties, Soft
plastic deformations, thereby extending their service life. Matter 8 (2012) 8483–8492.
– In contrast to traditional high-performance polymer modified [18] X. Qiao, X. Lu, W. Li, J. Chen, X. Gong, T. Yang, W. Li, K. Sun, X. Chen,
binders (which can reduce the plastic deformations appearing Microstructure and magnetorheological properties of the thermoplastic
magnetorheological elastomer composites containing modified carbonyl iron
in bituminous materials, based on their ability to recover part particles and polystyrene-b-ethylene-ethylenepropylene-b-styrene matrix,
of these deformations), mechanomutable asphalt binders Smart Mater. Struct. 21 (2012) 13.
512 F. Moreno-Navarro et al. / Construction and Building Materials 113 (2016) 506–512

[19] EN 1426, Bitumen and bituminous binders, in: Determination of Needle [23] AASTHO TP 70, Standard Method of Test for Multiple Stress Creep Recovery
Penetration, AENOR, Asociación Española de Normalización y Certificación, (MSCR) Test of Asphalt Binder Using a Dynamic Shear Rheometer (DSR),
Madrid, 2007. American Association of State and Highway Transportation Officials, 2009.
[20] EN 1426, Bitumen and bituminous binders, in: Determination of the Softening [24] J. D’Angelo, R. Dongre, Practical use of the MSCR test: characterization of SBS
Point – Ring and Ball Method, AENOR, Asociación Española de Normalización y dispersion and other additives in PMA binders, Transp. Res. Rec. 2126 (2009)
Certificación, Madrid, 2007. 73–82.
[21] EN 12593, Bitumen and bituminous binders, in: Determination of the Fraass [25] A. Gogalipur, Modification of Multiple Stress Creep and Recovery Test
Breaking Point, AENOR, Asociación Española de Normalización y Certificación, Procedure and Usage in Specification (Master thesis), University of
Madrid, 2007. Wisconsin, Madison, 2011.
[22] EN ISO 2592, Determination of flash and fire points, in: Cleveland Open Cup [26] A. Shenoy, Refinement of the Superpave specification parameter for
Method, AENOR, Asociación Española de Normalización y Certificación, performance grading of asphalt, J. Transp. Eng. 127 (2001) 357–362.
Madrid, 2002.